SSSR 2010 Conference Abstracts
Click on the name of the first author and see the abstract, click on Top to go back.
Asfaha - Cain - Yin - Adelman - Adlof - Ahn - Al Otaiba - Allan - Allen-Tamai - Aram - Aravena - Ariasi - Aro - Azimi - Baayen - Baker - Balass - Bar-Kochva - Barnes - Barth - Barzillai - Bekebrede - Bell - Berthiaume - Betjemann - Biancarosa - Bianco - Binder - Bitan - Blomert - Bodrova - Boets - Bolger - Bowles - Brandeis - Braze - Brem - Bridges - Brodeur - Buch-Iversen - Bus - Byrne - Cain - Calderón - Capotosto - Caravolas - Cardoso-Martins - Carlisle - Carroll - Casalis - Castles - Chang - Cheah - Chen - Chen - Chen - Chow - Christopher - Chuang - Chung - Coddington - Colombo - Commissaire - Cone - Connor - Conrad - Cornelissen - Coyne - Coyne - Cromley - Cronin - Crosby - Cunningham - da Mota - Daigle - Darrow - Davidse - Davidson - Davies - Davis - de Jong - de Lissa - de Souza Batista Kida - Deacon - Deacon - Defior - Denton - Devlin - Dewitz - Dion - Doabler - Drouin - Duncan - Elbro - Elliott - Elwer - Escribano - Evans - Farnia - Farrington - Fella - Fielding-Barnsley - Folsom - Folsom - Fonseca - Foorman - Fragman - Fricke - Friedman - Friend - Gagl - Geier - Gellert - Gendt - Georgiou - Gerde - Geudens - Gill - Gonzalez - Gooch - Goodrich - Goodwin - Gottardo - Gottwald - Goudey - Grifenhagen - Guerrero - Guo - Gutierrez-Palma - Günther - Haenen - Haentjens - Hagtvet - Haikio - Hampson-Jones - Harrison - Hart - Hartmann - Hauptli - Hautala - Havelka - Hawelka - Hebert - Heggie - Heikkilä - Herppich - Hiebert - Hofer - Hogan - Holliman - Holmes - Hoonhorst - Hornickel - Hsuan - Hume - Hurry - Huss - Hutzler - Ioannou - Irannejad - Ise - Janiot - Japel - Jenner - Jiménez-Fernández - Johnston - Jones - Kahn-Horwitz - Katz - Katzir - Kaya - Keenan - Kegel - Kemp - Kendeou - Kershaw - Kessler - Kim - Kim - Kirby - Kjeldsen - Klingner - Ko - Kohnen - Korat - Korinth - Koyama - Kuhn - Kuperman - Kuppen - Kurtz - Kyle - Lam - Landerl - Lau - Lauterbach - Lawrence - Lee - Lemons - Leppänen - Lerner - Lervag - Lesaux - Lesaux - Leseman - Leseman - Leung - Levin - Levorato - Liao - Lin - Lin - Lin - Lipsky - Liu - Liversedge - Logan - Logan - Lonigan - Lonigan - Loveall - Lowe - Luk - Luo - Manolitsis - Marinus - Martinussen - Mason - Masuda - Maurer - McArthur - McClung - McDowell - McGeown - McKague - Meiri - Melby-Lervag - Middleton - Middleweek - Miller - Mol - Monaghan - Morris - Morrison - Mundy - Murray - Müller-Myhsok - Naess - Nash - Nation - Neugebauer - Nicholson - Noordenbos - Oakhill - Olofsson - Olson - Pacton - Padakannaya - Pan - Papadopoulos - Parrila - Pasisz - Pasquarella - Pauly - Pennala - Pentimonti - Petscher - Phillips - Phillips - Piasta - Pike - Pimperton - Pineau - Poelmans - Pollard-Durodola - Pollo - Poole - Poulsen - Powell - Primor - Protopapas - Puranik - Quemart - Radach - Ramirez - Rau - Rauch - Ravid - Rayner - Rickard Liow - Ricketts - Rivera - Roach - Roberts - Roberts - Robins - Roenberg - Roskos - Russak - Sabatini - Saiegh-Haddad - Salmi - Santoro - Saunders - Savage - Schatschneider - Schiff - Schulte-Körne - Schwartz - Segal-Seiden - Segers - Seidenberg - Seidlova Malkova - Serniclaes - Shankweiler - Siegel - Silva - Silva-Maceda - Sims - Sivapurapu-Pisupati - Skiba - Skibbe - Smeets - Smith - Snellings - Snow - Soden - Solity - Solomon - Spanoudis - Sparks - Sperring - Stadie - Stainthorp - Steacy - Steenbeek-Planting - Steinbrink - Stewart - Stothard - Stringer - Taibah - Taylor - Teale - Tijms - Tirovolas - Tong - Torppa - Trapman - Trotter - Tunmer - Tzeng - Ugen - Uhry - Uno - Vaessen - Vaid - van Bergen - van Daal - Van den Broeck - van den Broek - Van Dyke - van Kruistum - Vandermosten - Vaughn - Veispak - Verhoeven - Vinco - Vousden - Wade-Woolley - Wagner - Wagner - Wang - Wang - Waters - Weekes - Weekes - White - Williams - Williams - Wilson - Wilson - Wimmer - Winskel - Wise - Wolff - Wolter - Wong - Wong - Wood - Wotschack - Wydell - Xiuli Tong - Xuereb - Yang - Yang - Yin - Yulia - Zeguers - Zesiger - Zevin - Zhang - Zhang - Zhao - Zhou - Ziegler - Zigmond
Yonas Mesfun Asfaha (Department of Language and Culture studies Tilburg University, the Netherlands)Jeanne Kurvers - Spelling processes in Ge'ez and Latin scripts
Purpose This is a follow up to a reading acquisition study in syllable based Ge'ez and alphabetic Latin that aimed to investigate the impact of availability and granularity or grain size on early reading development. Method The original study sampled around 400 grade 1 children from four different linguistic groups that learned to read and write in the alphasyllabic Ge'ez (Tigrinya and Tigre languages) and alphabetic Latin (Kunama and Saho) scripts. Word reading, spelling and syllable symbol or letter knowledge results were used to compare the grain size in script and teaching. In the current study, word spellings from a sample of 197 grade 1 students learning to read and write in alphasyllabic Ge'ez (Tigrinya language) and alphabetic-syllabic Latin (Saho) are examined to see how children represent CVC syllable end consonants. Outcomes of comparisons of good and bad spellers within and across languages will be related to their phonological awareness. Results: Earlier results showed the syllable based reading and spelling in Ge'ez script progressed faster than in Latin script and that alphabetic-syllabic Latin teaching resulted in higher reading and spelling outcomes than pure alphabetic teaching. We expect analyses of word spellings in the current study to point to a rather poor acquisition of correct syllable end consonant spelling when compared to CV-syllable signs. Conclusion As long as the language structure allows, a syllable based script and a syllable based teaching of alphabetic reading and writing show advantages in initial reading and writing development.
Kate Cain (Lancaster University, UK) - Discussion
Li Yin (Tsinghua University)Rebecca Treiman - Young Chinese children differentiate between drawing and writing
Purpose: Children under the age of 3 ½ or so are commonly thought to produce the same types of scribbles for writing and drawing. We tested this idea by examining whether Chinese children aged 2 through 6 make distinctions between drawing and writing in their productions and whether these distinctions make sense to adults. Method: 109 native Chinese children aged 2 through 6 in Beijing, China, participated in the study. They were asked to write and draw four targets. In Experiment 1, 16 native Chinese college students were shown, individually, several writings and drawings from a child and asked to determine whether a new product from the same child was writing or drawing. In Experiment 2, another 16 college students were shown two sets of productions from each child and asked to assign the productions to the categories of writing and drawing. Results: Results show that Chinese children as young as 2 produce distinctions between writing and drawing in ways that adults can appreciate. The adult judges performed significantly above the level expected by chance in distinguishing the written and drawn productions of children from all age groups in both experiments. Importantly, adults performed above the level of chance even with the productions of children aged 2 to 2 ½ and even with the productions of children who could not read any characters. Conclusions: Distinguishing writing from drawing in production, a first step in literacy learning, appears to take place earlier than previously believed.
James S. Adelman (University of Warwick); Maura G. Sabatos-DeVito; Suzanne J. Marquis - Warwick Oral Reading Differences Project
Suzanne Adlof (University of South Carolina); Charles Perfetti - Word-to-text integration of spoken sentences by adults with low reading comprehension skill: an event related potentials study
Purpose: Individual differences in reading comprehension have been linked to differential ERP responses during online text integration (Yang, Perfetti, & Schmalhofer, 2005; 2007). Specifically, when skilled comprehenders encounter words that repeat or paraphrase information from previous contexts, they show a reduced N400 (relative to the same word encountered in a neutral baseline context). Less-skilled comprehenders show this reduction for repeated words but not paraphrases, suggesting a specific difficulty integrating words with prior semantic context. Whereas past studies have not controlled for word reading ability, the current study extends these observations to listening comprehension, while matching the comprehension groups on word reading. This extension allows for an examination of whether the previously observed differences between skilled and less-skilled comprehenders are specific to reading situations or associated with oral language skills. Method: Participants include two groups of adults matched for word reading skill. The 'Low-comprehension' group displays good word reading (TOWRE) but poor comprehension (Nelson Denny & Gates MacGinitie). The 'Control' group displays good skills in both. Stimuli include 120 2-sentence passages, similar to those used by Yang et al. (2005, 2007), but recorded for audio presentation. The second sentence of each passage begins with a target word that relates to information in the first sentence as an: 1) Explicit repetition, 2) Paraphrase, 3) Inference, or 4) Baseline. Results & Implications: A finding that Low-comprehension and Control groups differ in ERP responses would provide evidence differences in contextual integration are not specific to reading, but rather associated with oral language processing.
Hei Won Heidi Ahn (University of Oxford); Kate Nation; Elizabeth Wonnacott - Against morpho-orthographic decomposition in early visual word recognition: Evidence from masked priming
Purpose: This study used masked priming to investigate morphological decomposition in visual word recognition in English-speaking children and adults. The current dominant view of visual word recognition holds that a morpho-orthographic decomposition process operates early on in recognition, whereby all words giving the appearance of morphological complexity are initially segmented into their plausible morphemic constituents, without regard to their actual morphological status. This view stems from masked priming studies which suggest that genuinely morphologically complex words, such as 'builder', and those displaying pseudomorphological complexity, such as 'corner', are decomposed into {build} and {-er}, and {corn} and {-er}, respectively. Little research to date has looked at morphological decomposition in developing readers, with none investigating this in English-speaking children. The first experiment sought to investigate morphological decomposition in English-speaking children using masked priming. The second experiment aimed to extend the findings from the first experiment to English-speaking adults. Method: In Experiment 1, forty-five English-speaking Year four and five children were tested on a masked primed visual lexical decision task. Experiment 2 tested twelve English-speaking adults on a modified version of the task. Results: In both experiments, whereas genuinely morphologically complex words like 'builder' primed their base forms ('build'), pseudomorphologically-structured words like 'corner' did not prime their plausible base forms ('corn'). This suggests that pseudomorphologically-structured prime words were not decomposed. Conclusions: These results implicate an early decomposition process that does not operate purely on the basis of orthography. We argue in favour of a semantically-informed account of decomposition in visual word recognition.
Stephanie Al Otaiba (Florida State University/FCRR);Carol M. Connor; Jessica Folsom; Luana Greulich - The role of pre-kindergarten home literacy and kindergarten classroom reading instructional environments on student reading outcomes in kindergarten and first grade
Purpose: This study examined the effects of home literacy and kindergarten reading instructional environments on student reading outcomes at the end of kindergarten and first grade. Methods: This cluster-randomized control field trial included 14 schools matched on free and reduced lunch, school reading scores, and then assigned to treatment or to a wait-list control condition. 23 treatment teachers were trained to use Individualized Student Instruction-K, were provided bi-weekly research support and monthly ongoing professional development. The remaining 21 teachers provided instruction as usual. Parents completed a survey in fall about their home literacy practices. Students' alphabetics and reading was tested in fall, winter, and spring (n = 556 students) of kindergarten and reading was again tested at the end of first grade (after attrition, n = 362). Given the nested nature of our data, to address our second question regarding the student outcomes, we conducted a hierarchical multivariate linear model using the HLM version 6.02 (Raudenbush, Bryk, Cheong, & Congdon, 2004). Where appropriate, we applied a Benjamini-Hochberg (BH) correction for multiple comparisons at the p < .05 level (Benjamini & Hochberg; 1995). Classroom instruction was observed, videotaped, and coded for reading instruction and individualization, as well as for fidelity of treatment. Results: Students in treatment classrooms had significantly stronger reading scores than controls (ES = .34) at the end of kindergarten. Generally, students whose parents reported stronger home literacy environments tended to have stronger outcomes than those with weaker home literacy environments. First grade analyses are currently underway. Based on classroom observation, treatment teachers provided significantly more individualized instruction than did control. Conclusions: Implications for child characteristic by instruction interactions, including the influence of reported home literacy and classroom learning environments, will be discussed.
Nicholas Allan (); Christopher J. Lonigan - Unique associations of internalizing, externalizing, and social behaviors on emergent literacy in preschool children
Purpose: Accumulating evidence indicates that many behavioral constructs are associated with the development of emergent literacy skills. Whereas researchers have explored the contributions of externalizing, internalizing, and social behaviors independent of other behaviors in preschool children, little has been done to determine the unique relations between these behaviors and the emergent literacy skills of preschoolers. The purpose of this study was to explore the joint and unique relations between social competence, internalizing, and externalizing behaviors with emergent literacy. Method: Preschool children (N = 405; M age = 4.89 years, SD = 0.51; 41% girls) completed emergent literacy (oral language, phonological awareness, print knowledge) measures in the spring of their preschool year. At that time, teachers rated the children's behaviors using the Social Competence and Behavior Evaluation-Short Form. Results: Results from correlations and regression analyses revealed that the three behavior domains were all significantly correlated with the three domains of emergent literacy; however, when examined simultaneously, only social competence and internalizing behaviors were uniquely associated with emergent literacy skills. Conclusions: This study extends previous research in preschool children that had found evidence for relations among externalizing and social behaviors and emergent literacy. Whereas externalizing behavior had previously been found to be associated with emergent literacy skills, it appears that only social and internalizing behaviors are uniquely associated with emergent literacy skills. Longitudinal studies of the relation between these behaviors and emergent literacy will help determine whether internalizing and social behaviors are associated with emergent literacy growth as well.
Mitsue Allen-Tamai (Aoyama Gakuin University) - What affects the development of word knowledge among young Japanese EFL learners?
Vocabulary building has been a keen interest among L1 and L2 teachers of young learners. There are many layers of knowledge involved in knowing a word. The presenter is interested in examining what kind of word knowledge young Japanese EFL learners have and how they develop it. She has hypothesized that there is a significant relationship between phonological awareness, alphabetical knowledge, and word recognition and that Japanese children get benefits from trainings in phonological awareness and letter recognition to acquire their vocabulary. One of the aims of this study is to examine the English word knowledge of young Japanese EFL learners and its relationship to phonological awareness and alphabetical knowledge. The other aim is to examine the effects of a training in phonological awareness and letter recognition on word recognition. Japanese children of ages ten to thirteen (N = 428) participated in the study and their phonological awareness and alphabetical knowledge were measured. The participants were also administered three different tests to measure their 1) phonological knowledge, 2) orthographic knowledge, and 3) conceptual knowledge of words. Structural equation modeling was used to analyze the influence of the two variables on word recognition. The model suggests that Japanese young learners of English develop their phonological awareness and alphabetical knowledge and the both influences their word recognition. The effects of one-year trainings were also found and the participants developed their word knowledge by taking trainings on phonological awareness and letter recognition.
Dorit Aram (Tel Aviv University, Israel);Margalit Ziv, AlQasemi teachers' College and Tel Aviv University, Israel; Yaara Fine, Oranim Academic College, Israel - Promoting Parental Reference to Mental States via Storybook Reading Interactions with Preschooler from Low Socio-economic Status Background
Purpose: This study examined the effectiveness of an intervention, aimed to enrich parents' storybook reading surrounding internal mental states. Such knowledge has been associated with problems in social interaction, school acceptance, and literacy. Method: Parents in the intervention (n = 30) received six books and were instructed to refer to central mental states, different points of view and the child's life while reading them to their children. Comparison group parents (n = 28) received the same books without instructions. The participants were from low SES families. The intervention was based on a structured model that included four repeated readings of one book/week. Parents were asked to (1) make sure that their child understood the story; (2) discuss central mental states in the story; (3) connect the story to the child's life, and (4) the child was asked to tell the story. Home visits and video-tapes of reading were conducted for both groups. Results: Compared to the control, parents and children in the intervention group expressed a richer mental state discourse, in which the child was more active; parent-child discourse was longer and included more references to vocabulary, to mental states, to the child's life, to the causes and outcomes of mental states, and to the central mental theme of the story. Conclusions: Conversation about the causes and consequences of mental states may be important and interventions based on parental mediation of storybooks can contribute to children's mental understanding, and potentially literacy development.
Sebastián Aravena (University of Amsterdam); Jurgen Tijms - Training reading fluency in children with dyslexia: explicit, rule-based learning versus implicit, associative learning
Purpose: In our study, we transposed the discussion on explicit versus implicit learning in skill acquisition to the area of training reading fluency. As knowledge of phonetic-graphemic relations and the ability of using them instrumentally in reading are considered the core aspects of developing reading fluency, these two elements were the focus of our training study. Method: An artificial orthography was used to train both regular and irregular phonetic-graphemic relations in a computerized intervention. Children with dyslexia and non-dyslexic controls were provided with one of three experimental training conditions: a) an explicit, algorithmic rule-based training, b) an implicit, associative training using repetitive exposure, c) a combination of a and b. Both their knowledge of the phoneme-grapheme relations and their reading fluency in this artificial orthography were tested during and after training. Results: The results show that a) all training conditions are effective in teaching phoneme-grapheme knowledge and tend to be durable, b) there is transfer of phoneme-grapheme knowledge on reading fluency of the artificial orthography, c) children with dyslexia are less susceptible to acquire reading fluency skills than non-dyslexics, d) children develop better reading fluency skills with the explicit and the combined training than with the implicit, associative only training condition, and f) explicit training seems to nourish the subsequent effectiveness of implicit, associative training. Conclusions: Both implicit-associative and explicit-rule based training methods seem to bear potential in reading fluency intervention in children with dyslexia. The combination of explicit training followed by more implicit training seems to be most promising.
Nicola Ariasi (University of Padova, Italy);Lucia Mason - Text structure and learning from expository text: Evidence from eye-tracking
Purpose This study aimed to investigate the degree to which different learning outcomes produced after reading a refutation and a non-refutation text related to different cognitive processes during reading. Method 59 college students participated, who read a refutation or a non-refutation text about tides. Both texts comprised the same number of sentences, words, and characters. Eye-movement indices were: first-pass, look-back, and look-from fixation times. A pre- and post-test was used to assess learning. Results Findings from a 2(refutation/non-refutation text) x 2(pre/posttest) mixed-factor design (reading time and working memory as covariates) revealed that readers had higher learning outcomes after reading the refutation text, F(1,54)=5.697, p=.02, n2p=.10. A subsequent ANCOVA revealed that refutation text helped readers co-activate the correct and incorrect ideas (as indicated by the look from measure), F(1,54)=4.449, p=.04, n2p=.08. Finally, a hierarchical regression analysis indicated that processes as indicated by eye-movements accounted for a significant 44% of the variance in learning outcomes, F(9,18)=3.327, p=.01. The longer the time spent looking from the scientific conceptions to the text refutation parts, the higher the learning score, β=.372, t(18)=3.889, p=.001. Conclusions The study provides evidence that readers allocate attention differentially according to the text structure. Findings support the hypothesis that the refutation text facilitates the simultaneous activation of misconceptions and correct conceptions (conflict detection), which is particularly relevant to knowledge restructuring. The crucial link between online cognitive processes and offline outcomes of learning from text has been documented.
Mikko Aro (University of Jyväskylä and Niilo Mäki Institute); Jarkko Hautala; Sini Huemer - Assessing improved reading of syllables among Finnish poor readers
Purpose The effects of repeated reading of syllables were evaluated in an experimental training study with 28 Finnish-speaking poor readers. The questions addressed were whether an emphasis on either speed or accuracy during training has an effect upon training outcome, and whether the possible training effect is specific at a level of syllable. It was also of interest whether the effects of syllable training can be observed in eye movement patterns during reading. Method Two groups of children (speed/accuracy) participated individually in 10 training sessions, with altogether 50 repetitions. The outcome measures were reading times of syllable and pseudoword lists, consisting either of trained syllables, neighbour syllables (shared bigraphs with practice syllables) or control syllables (no shared bigraphs). There were three assessments. Time between 1st and 2nd assessments served as control period. The training took place between 2nd and 3rd assessments. A subsample of 18 children participated also in an eye movement study. Eye movements were recorded prior and post training when reading the same tasks that were used in assessments. Results There were no differences between the two groups in any outcome measure. Increase in reading speed was observed in reading of practiced syllables, and pseudowords containing practiced syllables. The training reduced the number of fixations in items with trained syllables. Conclusions The results imply that repeated reading of syllables might offer a way to train reading speed in Finnish. The effects of training seem to be specific to the trained materials at the level of syllable.
Mahshid Azimi (OISE/University of Toronto)Esther Geva; Fataneh Farnia; Alexandra Gottardo - Early home language and school language predictors of grade 4 reading comprehension
Purpose: We examined the association between kindergarten performance on linguistic and cognitive measures assessed in English language learners' (ELL) home language (L1) and in the societal language (English) and grade 4 English reading comprehension. Method: The sample included 112 ELLs from three typologically different language backgrounds: Chinese (45%), Spanish (41%), and Portuguese (14%). We assessed ELLs' cognitive, language and various literacy skills from kindergarten to grade 4, using multiple measures. Several language and literacy skills were also assessed in the L1 (e.g., vocabulary, word reading, phonological awareness). Results: Regression analyses suggest that phonological awareness, rapid naming, and word reading measured early either in the L1 or in English, and letter sound knowledge accounted for unique and significant variance in reading comprehension four years later. Conclusions: Implications for a comprehensive L2 reading theory, the nature of "transfer", and implications for assessment will be addressed.
Harald Baayen (University of Alberta) - The directed compound graph of English: An exploration of lexical connectivity and its processing consequences
Doris Luft Baker (Center on Teaching and Learning); Keith Smolkowski; Scott K. Baker; Sylvia Linan-Thompson; Alejandra Mielke - Reading intervention with Spanish-speaking students: Maximizing instructional effectiveness in Spanish and English in first grade
Purpose This study examines the impact of systematic and explicit teaching routines in Spanish on the reading performance of first grade Spanish-speaking English learners (ELs) in the United States. We will present overall reading achievement outcomes in Spanish and English and discuss the relationship between implementation fidelity and outcomes. Method Design: Randomized control trial with assignment at the school level Participants: 38 schools in Oregon and Texas Materials: Eighteen lesson cards designed to develop teaching routines for phonemic awareness, phonics, fluency, vocabulary, and comprehension activities. Measures: BVAT; Aprenda and SAT-10: Word Reading, Sentence Reading, and Reading Comprehension subtests; IDEL Fluidez en las Palabras sin Sentido and Fluidez en la Lectura Oral; DIBELS Nonsense Word Fluency and Oral Reading Fluency; classroom observation measures of implementation fidelity. Data Analysis: Mixed-model analysis of covariance (ANCOVA) and random coefficient analysis (RCA) were used to measure intervention impact. Results Initial findings indicate that the intervention had a significant effect on oral reading fluency and pseudo word reading in Spanish and English, favoring the treatment group. Preliminary analysis indicates that reading outcomes were higher in classrooms where treatment implementation was stronger. Conclusions: Providing ELs with systematic and explicit instruction in first grade leads to better outcomes on some measures of reading, but not others. In particular, the intervention had an impact on word reading and fluency outcomes in Spanish and English, but not on comprehension. Results will be discussed in the context of different types of reading approaches to teach reading to ELs in the United States.
Michal Balass (Hobart and William Smith Colleges); Laura Halderman; Erik Benau; Charles A. Perfetti - Uh-Oh! I made a mistake: ERP indicators of meaning processing during word reading
Purpose: We investigated error detection by measuring individuals' error-related negativity (ERN) during meaning judgments. The ERN is elicited immediately after an error response, reflecting a conflict when multiple responses compete. We tested the hypothesis that the ERN would indicate conflict in responses during meaning processing. We predicted an ERN effect for errors made to simple semantic category judgments that may measure implicit knowledge about word meanings. Method: Adult readers completed a time-pressured semantic categorization task designed to elicit errors. Foils were semantically related to the category, e.g., given the category COUNTRY, failing to respond to "yes" to India and responding "yes" to Africa would be errors, and the potential locus of ERN effects. Results: The ERPs showed greater negativity at approximately 30 to 80ms post-response for incorrect responses than for correct responses. The magnitude of this error response was modulated by readers' discrimination sensitivity between category members vs. non-members. Readers who showed higher sensitivity showed larger ERN effects and increased positivity to correct responses. Conclusions: The results show that errors in a semantic categorization task elicit an ERN response. However, the difference between high performing and low performing participants was due primarily to a greater positivity for correct responses, rather than greater negativity to errors. Thus, the data suggest that readers who are more accurate have an enhanced neural response when they are correct in semantic judgments. Thus, knowing when you are correct may help you know when you are wrong.
Irit Bar-Kochva (University of Haifa, The Edmond J. Safra Brain Research Center for the Study of Learning Disabilities); Zvia Breznitz - Predicting reading skills in the two forms of Hebrew orthography at the formal transitional stage between the 3rd and the 4th grades.
Purpose: Hebrew readers are skilled in reading two forms of script. Reading of the shallow pointed script is initially acquired, and during the 3rd and the 4th grades children are expected to proceed to reading in the deep unpointed script. The purpose of this study was to examine which skills underlie reading the pointed and the unpointed Hebrew script during this transitional stage. Predictions derived from the "Orthographic depth hypothesis" (Katz & Frost, 1992) were examined using a within-subject-and-language study design. Method: 87 children participated in a longitudinal study from the 3rd to the 4th grade. Each year parallel versions of pointed and unpointed reading tests were administered, as well as tests addressing linguistic and nonlinguistic skills. Results: Reading proficiency in both forms of script was very similar as early as the 3rd grade. Regression analyses suggest that phonological awareness, morphological awareness and RAN were stronger predictors of accuracy in oral reading of pointed words and texts than of unpointed words and texts. Similar results were obtained in the case of silent reading comprehension. However, these differences were mostly small, especially in the 4th grade. Furthermore, RAN, vocabulary and visual speed of processing explained a similar amount of variance in pointed and unpointed reading fluency in both grades. Conclusions: The results may suggest that at this transitional stage both forms of scripts are processed similarly. It is possible, however, that as the exposure to pointed texts decreases with age, reading routines of the two forms of script become more distinguishable.
Marcia Barnes (University of Texas-Houston Health Sciences);Kimberly Raghubar - Children and adolescents from the third to tenth grades see things through the protagonist's eyes during reading
Purpose. To investigate the construction of spatial situation models during reading in children and adolescents. Method. Eighty-five participants in grades 3 to 10 studied a layout of a market with objects outside the shops (e.g., hydrant and flowers outside meat shop). Then they read a story about a protagonist moving through the market. Reading was periodically interrupted by the presentation of two words; participants decided if these objects were from the same or different shops. Objects were from the location where the protagonist ends up (Goal); the unmentioned location the protagonist passed through to get from one shop to another (Path); the location from which the protagonist started (Source); or another location not mentioned in the text (Other). Results. Participants in grades 9-10 learned the layout of the market more quickly than those in grades 3-8, but there were no grade-related differences in later retention. Grades were similar in accuracy and more accurate on items from the Path locations than the Source and Other locations. Response time findings supported those for accuracy except that participants in grades 9-10 were faster than all other grades. Conclusions. Greater accuracy and speed in judging objects in the unmentioned pathway signifies that readers built a mental model of the story based on the viewpoint of the protagonist. The findings are similar to those reported for fluent adult readers and suggest age-invariance in making implicit spatial inferences during reading. Findings are discussed in relation to models of text processing and theories of grounded cognition.
Amy Barth (University of Missouri-Columbia); Marcia Barnes; Karla Stuebing; Carolyn Denton; Jack Fletcher - Language Profiles of Treatment Nonresponders, Treatment Responders, and Typically Developing Students
Background. Students with accurate decoding in the presence of poor reading comprehension are impaired on a wide range of language tasks (Cain & Oakhill, 2006). Purpose. The purpose of this study is to compare the language profiles of second grade poor comprehenders who failed to respond to a multi-component reading intervention, students who responded to intervention, and typically developing readers. Method. The sample comprises 32 poor comprehenders who failed to respond to intervention (TXNR), 88 students who responded to intervention (TXR), and 52 typically developing readers (TYP). All students performed above 93 on the WJ-III Basic Skills composite. Poor comprehenders obtained standard scores of less than or equal to 93 on the WJ-III Passage Comprehension. The following skills were assessed: decoding, passage comprehension, phonological awareness, listening comprehension, syntax, vocabulary, verbal working memory, and bridging inference making. Results. A mixed model ANOVA was conducted to obtain profiles for students' language abilities. Using Wilk's criterion, a significant shape effect was observed F(14, 294) = 2.91, p < 0.0004, eta square= 0.21. Follow-up pairwise contrasts with group status as the between subjects factor indicated that TXNR performed significantly lower on phonological awareness, syntax, listening comprehension, vocabulary, and verbal memory than TXR and TYP and also lower than TYP on bridging inferences. TXR performed lower than TYP on phonological awareness, syntax, listening comprehension, and vocabulary. Conclusions. These findings suggest that poor comprehenders, who fail to respond to intervention, continue to present deficiencies on a wide range of language tasks.
Mirit Barzillai ()Maryanne Wolf; Sasha Yampolsky; Robbin Morris; Maureen Lovett - The role of rich, flexible, semantic knowledge in reading intervention
Purpose: The role of semantic knowledge within the reading process has become of increased importance in intervention research for children with reading disabilities. The current study evaluated the effects of a multi-componential intervention (RAVE-O) that emphasized the formation of high quality semantic representations and their connections to other reading subskills (i.e., phonology, orthography,morphology, syntax)in a group of struggling second and third grade readers. The investigation focused on whether such an intervention would foster increases in word reading and fluent comprehension as well as gains in semantic knowledge when compared to phonological-decoding interventions without semantic emphases. Method: 279 second and third grade struggling readers were randomly assigned to interventions, each of which was taught by trained teachers for 70 hours. A battery of standardized and experimental language and reading measures was administered to the children pre, mid (35 hours) and post intervention, as well as at one year follow-up. Results: Hierarchical linear modeling analyses revealed significant gains for the RAVE-O group on both standardized (Word-R) and experimental measures of polysemous word knowledge and semantic flexibility immediately after the intervention, as well as one-year post intervention. These gains in semantic knowledge were in addition to significant increases in fluent comprehension and word reading. Conclusion: The present results attest to the importance of incorporating an emphasis on rich semantic knowledge and its connections to the other reading subskills within reading interventions. Implications for multidimensional intervention approaches and semantic emphases within interventions for reading fluency will be discussed.
Judith Bekebrede (Center for Educational Training, Assessment, and Research VU University); Aryan van der Leij; Truus Schijf; David Share - Dyslexia across three decades
Purpose This study investigates whether dyslexia is comparable across ages. Does the manifestation of dyslexia and underlying subskills vary when dyslexic students are compared to age matched controls in primary and secondary school and adulthood? Method In the present study carefully selected dyslexics and controls are compared at four different ages: 8/9, 12/13, 15/16 and 37/38. At every age, 35-50 dyslexics are compared to controls (matched on verbal and non-verbal intelligence) on equivalent tasks tapping (pseudo)word reading fluency, rapid naming, phonological awareness, (speed of) orthographic processing. Results The results indicate that the dyslexics differ from the controls on all measurements regarding reading (related) skills at all ages. There appears to be a larger difference in (pseudo)word reading fluency and phonological awareness at the age of 15/16 and 37/38 than at age 8/9 and 12/13, whereas the differences in rapid naming and orthographic processing comparisons are similar. In addition, within tasks and across ages, word reading fluency and orthographic processing of both groups are better with age, but pseudoword fluency of dyslexics at higher ages is comparable. Conclusions The results indicate that, in comparison to age controls, the impact of the phonological core deficit is larger at higher ages. In contrast, orthographic processing seems to be less affected when they are older.
Nanci Bell (Lindamood-Bell Learning Processes) - Intervention that integrates imagery and language improves adolescents' reading skills.
Purpose: The question addressed was: Can theoretically-based reading instruction that integrates mental representations and language successfully increase word and context level reading skills in adolescent poor readers. Method: A random sample of 10th graders who scored below proficiency on state standards testing were assigned to either experimental or control conditions. Based on diagnostic testing, the experimental group received focused, small group instruction during one semester (mean=118.7 hours) and the control group attended the regular 10th grade English course. Analysis was restricted to those who also scored below average on an individually administered broad reading test (27 experimental, 28 control). Results: Although the two groups did not differ at baseline, the experimental group had significantly greater changes on measures of phonemic and orthographic awareness, decoding, fluency, and spelling but not on listening or oral reading comprehension. Where groups differed, effect sizes were moderate to strong. Groups did not differ on end-of-year standards testing; however the experimental group had significantly higher GPAs. Conclusions: High school students who receive a semester of instruction focused on specific weaknesses can make substantial gains in phonological processing and word level skills in a short time, but greater intensity and duration of instruction may be needed to significantly affect context level skills. That state standards test scores have proved intractable in this and other studies, which in this case runs counter to evidence of improved grades, may be a reflection of achievement test construction.
Rachel Berthiaume (Département de didactique)Daniel Daigle; Elisabeth Demont - Morphological processing and learning to read: a comparison between deaf and dyslexic children
Deaf children and dyslexic children encounter great difficulty in learning to read (Ecalle & Magnan, 2002; Gallaudet Research Institute, 2004). If phonological processing is mainly pointed out as the source of their reading deficit (Demont, 2003; Musselman, 2000), there are no apparent reasons to believe that these children would have difficulties in processing the morphological structure of words. Indeed, morphological units are visually accessible units, as opposed to units of sound. The objectives of this study are to investigate and compare deaf students' and dyslexic students' knowledge of morphology in written French. Deaf subjects and dyslexic subjects (n=32) aged 9 to 12 were matched to subjects of the same chronological age (CA) and subjects of the same reading age (RA). We used a lexical probability task (LPT) where subjects had to determine which of two pseudowords (e.g. prépareur or papiereur) resembled the most to a real word in written French and a segmentation task (ST) where subjects had to segment words into morphemes. For all groups, results are above the threshold of chance for the LPT only. Results from ANOVAs indicate that all groups had better scores on the LPT than on the ST, that deaf subjects and CL have comparable results, but perform more poorly than CA, and that dyslexic subjects performed more poorly than CL and CA. These results constitute the basis for further research on the role of morphological structure in deaf and in dyslexic children's reading development and may lead to more adapted literacy teaching practices.
Rebecca Betjemann (Regis University); Sally J. Wadsworth; Richard K. Olson; Erik G. Willcutt - A longitudinal investigation of GPA, ACT, and SAT scores as Outcome Measures related to reading ability.
This study investigated the relation between reading-related skills at two time points and outcome measures that are important for college admission. The sample included 94 participants from a larger twin study, with a mean age at initial testing of 13.4 years, and a mean age at follow-up of 19 years. Correlations were computed between three outcome measures from follow-up (Current GPA, ACT scores, and SAT scores) and five reading-related measures at both time points (Word Reading, Spelling, Vocabulary, Processing Speed, and Reading Comprehension). Results revealed low to moderate correlations between GPA and the reading measures at follow-up (r = .21 - .41). Whereas ACT Scores were moderately and significantly correlated with all follow-up measures (r = .38 - .64), SAT scores had significant moderate correlations with Word Reading and Spelling (r = .52 and .61), but not Vocabulary or Reading Comprehension (r = .25 and .28). Reading and speed measures were highly stable over the 5.6-year interval between initial and follow-up assessments (e.g., r = .56 - .77) and correlations between the reading-related measures at initial assessment and the three outcome measures were surprisingly similar to those between follow-up reading-related measures and outcome, ranging from .24 - .55, and all were significant. Results suggest that these key scores used for college admission may be differentially related to reading-related skills, but these relations are quite stable, even with reading measures from 6 years previous. Thus, early reading and reading-related skills may predict performance on measures that are important for college admission.
Gina Biancarosa (University of Oregon); Anthony S. Bryk - Summer growth in reading from kindergarten through second grade
Purpose: This study explored individual- and school-level contributions to children's longitudinal change in reading ability during the two summers occurring between kindergarten and second grade. Method: Participants included all children enrolled in kindergarten through second grade (n=8,520) attending 17 schools across the Eastern U.S. over four years. Children were assessed each fall and spring with grade-appropriate subtests of the Dynamic Indicators of Basic Early Literacy Skills and each spring with the Terra Nova test of reading. A single scale measure of reading ability was derived from the resulting scores using Rasch analysis allowing reading to be modeled as a continuous outcome across all grades. Longitudinal cross-classified modeling provided the baseline model for analyses. Because the data come from a larger study of the effects of an American school-wide literacy reform model, analyses controlled for intervention effects and also for the value-added by the specific teachers and schools children experienced. Results: Findings revealed that on average students lost ground in reading ability during the kindergarten-to-first-grade summer but grew during the first-to-second-grade summer. Variance in these effects was significant only at the school level (not at the student or classroom levels) and was greater for the first-to-second-grade summer than at any other period. Subsequent analyses explored school-level socio-economic status (SES) and proportion of minority students as predictors of change in reading ability during the summers. Conclusion: Findings suggest that much of the progress made in closing reading achievement gaps in lower SES schools during the academic year is lost during the summer.
Maryse Bianco (Université P. Mendès France)Lima,L., Bressoux, P., Pellenq, C. - Early training at code and comprehension skills and reading achievement at first grade
Purpose: The relationship between oral language development, early training and word identification as well as reading comprehension at first grade was examined in a sample of 687 French children in a 3-year longitudinal study. Method: Children's phonological awareness and oral language comprehension development were assessed at 4 years of age, at the onset of the study. Children were then entered in two experimental groups. The first one was trained at phonological awareness and the second one received a comprehension skills training. The experimental groups were further divided in 2 subgroups. The first one was engaged in a two semesters training and the second was trained during one semester only. Word identification and reading comprehension were assessed nine month after the training period ended, as children were at first grade. Results: Hierarchical linear models showed that both phonological awareness and oral comprehension at 4 years of age explained a unique part of variance for the two outcome measures, while revealing a symmetrical influence: phonological skills explained more variance for alphabetic reading skills and oral comprehension explained more variance for reading comprehension. The results also showed that the training programs had positive medium term effects when they lasted two semesters. In this case, the phonological awareness training had a positive effect on alphabetic scores and the comprehension training program had a positive effect on reading comprehension. Conclusion: These results provide some insight for early oral instruction and are discussed in the light of the theoretical debate relating to the linguistic predictors of literacy acquisition.
Katherine Binder (Mount Holyoke College)Brooke Magnus - Children's use of morphologically complex words in written and oral story-telling
Purpose: The purpose of this study was to examine the developmental trajectory of children's acquisition of morphological awareness, and their use of morphologically complex words in written and oral storytelling. Method: We assessed 70 children (grades 3rd - 5th) on measures of morphological awareness, spelling, phonological abilities, and vocabulary knowledge. Additionally, participants composed oral and written narratives based on picture prompts. For the written stories, children were also provided with a revision period during which time they could correct their stories. The stories were analyzed for the use of morphologically complex words. Results: Results revealed that increases in grade level were associated with increases in the use of morphologically complex words, and participants used more morphologically complex words in their oral stories than in their written stories. In addition, participants produced more inflected words compared to derived words. While we hypothesized that the gap between inflected and derived words would decrease with increases in grade level, the gap actually increased such that older students were using more inflected words than the younger students, but there had been little growth in the use of derived words. When we examined the data produced during the correction period, we found that the number of corrections that involved morphologically complex words was correlated with both a participants' morphological awareness score and with the total number of morphologically complex words produced in the written narrative. Conclusions: For the age groups we tested, children's knowledge of inflected morphemes is stronger than their knowledge of derived morphemes.
Tali Bitan (Dept. of Communication Disorders, University of Haifa)Adi Morag; Einav Yehezkel; Tami Katzir - Does the mode of phonological representation make a difference? The effects of vowel letters vs. diacritic marks in reading Hebrew words at different developmental stages
Cross linguistic studies, that go beyond the English language, have shown that readers of more consistent orthographies, rely on smaller grain size units in the conversion of orthography to phonology, even in early stages of reading acquisition (Ziegler & Goswami, 2005). Hebrew provides an opportunity to study these effects within language. Our goal was to examine effects of orthographic transparency on oral word reading in various stages of development. Children in second and fifth grades and adults participated in an oral naming task, in which word length, the presence of a vowel letter, the presence of diacritic marks and frequency, were manipulated. The effects of these parameters on accuracy and RT were tested, as well as their correlations with standardized measures, and a systematic error analysis. Our results show greater facilitating effects of diacritic marks in 2nd grade, and greater facilitating effects of vowel letters in 5th graders and adults. We also find that all children read short words faster than long words, but the opposite is true for adults reading words with no vowel letters. Altogether these results reveal a developmental transformation from relying on sublexical decoding in children to relying on larger whole word units in adult Hebrew readers. Moreover, children initially rely on full phonological information provided by diacritic points, and gradually develop their ability to extract phonological information from partial and more ambiguous cues provided by vowel letters. Finally, while vowel letters provide phonological information for children, adults may also benefit from their orthographic contribution.
Leo Blomert (Dept. Cognitive Neuroscience, Faculty Psychology & Neuroscience Maastricht University, the Netherlands); Nienke van Atteveldt; Vera Blau; Dries Froyen - The multisensory brain basis of normal and abnormal reading
Purpose: A prerequisite for learning to read is the establishment of a link between speech and script; in alphabetic languages an association between letters and speech sounds. Two widely accepted assumptions about the necessary prerequisites/causes of normal and abnormal reading acquisition will be questioned. First, it is generally assumed that the acquisition of grapheme-phoneme correspondences is a minor event in comparison to the development of phonological awareness and takes only a few months to learn in relative transparent alphabetic languages. Second, the dominant theory of dyslexia assumes that phonological problems cause reading problems. Methods: The results of several ERP and fMRI studies will be integrated. To study the degree of automation of letter-speech sound association we conducted ERP (MMN) experiments with beginner (7-8 yrs), advanced normal (11 yrs) readers and dyslexic children (11 yr) and adults (22 yr). To study the locus and nature of letter-speech sound association in dyslexia we conducted fMRI experiments with normal and dyslexic children (8-9 yr) and adult readers (22 yr). Results: The electrophysiological results show that the automation of letter-speech sound processing takes many years to develop, because it is not just the learning of an association, but a neurobiological integration process. The neuro-imaging results show that the proximate cause of reading problems in dyslexia resides in inadequate letter-speech sound processing in dyslexia. Conclusion: Reading acquisition is essentially multisensory in nature and the failure thereof consequently can be found in inadequate multisensory integration.
Elena Bodrova (Mid-continent Research for Education and Learning; Denver, Colorado, USA)Deborah J. Leong - Self-Regulation and the link to literacy: A pilot study of the Tools of the Mind program
Purpose: The presentation will describe a pilot study of a program designed to impact self-regulation and the possible effects this might have on literacy development. Tools of the Mind is a comprehensive early childhood program that applies Vygotskian and post-Vygotskian ideas to teaching preschool- and kindergarten-aged children. One of the defining characteristics of this program is a system of self-regulation scaffolding (Bodrova & Leong, 2001, 2007) that includes focal self-regulation activities as well as activities where self-regulation is embedded into the teaching of content skills, such as literacy and mathematics (for example, Buddy Reading where children use external mediators to help them take turns). Methods: The presenters will share the results from a small pilot study of 37 preschool children in Head Start classrooms located in the Pacific Northwest Children's level of self-regulation were measured mid- and end-year using the Head Toes Knees Shoulders task (Ponitz et al., 2008), and the literacy level of each child was assessed by analyzing the child's Play Plan, collected in the same week of the self-regulation measure. Play Planning occurs every day, has minimum teacher support, and produces a tangible product to be coded. Results: Children with higher levels of self-regulation were more advanced in their writing-demonstrating an understanding of concept of word, representing a message that expressed the intent of what they would do, phonemic awareness, and the use of sound-to-symbol correspondence to represent sounds. Conclusion: There is some preliminary evidence that there may be a link between literacy knowledge and self-regulation.
Bart Boets (Active Member, Regular); Maaike Vandermosten; Hanne Poelmans; Heleen Luts; Jan Wouters; Pol Ghesquière - Longitudinal relations between auditory processing, speech perception, phonology and literacy development in children at risk for dyslexia
Purpose: The auditory temporal processing theory postulates that dyslexia results from a deficit in the processing of auditory temporal stimuli. This basic deficit is hypothesized to hamper accurate speech perception, hence disrupting subsequent development of adequate phonological representations and literacy skills. The present longitudinal study aimed to verify the presence of these postulated deficits and causal relations. Methods: 31 children with a family history of dyslexia and 31 matched low-risk controls were followed up from preschool to third grade. Low-level auditory processing, speech-in-noise perception, categorical perception, phonological ability and reading and spelling were assessed at various time moments. Results: A preliminary analysis where children were categorized in groups based on family-risk status and first-grade literacy achievement, indicated that those children showing both the family-risk and the literacy-impairment presented preschool deficits in FM-detection, speech-in-noise perception, phonological awareness and rapid automatic naming. Investigation of the relations indicated that auditory processing was related to speech perception, which itself was related to phonological awareness and first-grade literacy development. In the current presentation, we will further elaborate these data by analyzing groups based on formal third-grade dyslexia diagnoses, and by modeling longitudinal relations using causal path analysis. Conclusions: This longitudinal study indicates that the auditory deficit generally precedes and predicts the literacy problem.
Donald Bolger (University of Maryland)Say Young Kim - The effect of sub-syllabic units in the orthographic processing of Korean
Purpose: The Korean orthography is uniquely designed to combine consonants and vowels into a square-shaped cluster. In addition, it was well-known that a syllable structure in Korean is a left-branching (i.e., body-coda) model which contrasts to English (i.e., right-branching). The present study addresses whether or not Korean speakers favor of body-coda structure over onset-rime structure within a monosyllabic word can be confirmed by a rapid orthographic processing. Method: We examined the role of sub-syllabic units in the orthographic processing of Korean during a masked prime lexical decision task with one of three different SOAs (19ms, 30ms, and 50ms). Native Korean participants were given CiVCc structured monosyllabic words as targets preceded by one of six different types of primes: Onset consonant (Ci), body (CiV), rime (VCc), non-match (CcVCi), identical (CiVCc), and control (CVC). Results: The main effects of SOA and prime type were significant. The priming effect size increased as SOA increased. Identical primes induced larger facilitation effect on the targets than other prime types. Although the priming effect of both sub-syllabic units of interest, body (CiV) and rime (VCc), were significant, the magnitude of priming effects for the rime were larger than those of the body across all SOAs. Lastly, non-match primes induced an inhibition effect at the longest SOA (50ms). Conclusions: The current data suggest that the role of sub-syllabic units in Korean orthography may be different from that predicted by the left-branching model of syllable structure under masked priming circumstances.
Ryan Bowles (Michigan State University)Kevin Grimm; Lori Skibbe; Fred Morrison - Statistical methods for the study of schooling effects
Purpose: We present and compare three statistical methods for examining the effects of schooling on literacy outcomes. Growth in literacy results not only from instruction within a school setting, but also from non-schooling effects such as maturation and learning within the home, and methods for assessing schooling effects must differentiate between these two sources of literacy skill growth. Method: We evaluate the statistical and theoretical properties of three methods: (1) The school cutoff design treats the age cutoffs between grades as a form of random assignment, yielding a natural experimental design; (2) Regression discontinuity analysis includes age as a covariate to adjust for non-schooling effects within a standard regression analysis with schooling as the predictor variable; (3) Nonlinear growth curve analysis uses a smooth nonlinear growth curve model to account for non-schooling effects, allowing schooling to be considered as a time-varying covariate. To illustrate the methods, we employ data from 383 children measured up to 10 times from preschool through second grade on literacy measures assessing decoding, phonological awareness, and vocabulary. Results: The school cutoff design provides the strongest causal conclusions but has strict data requirements. Regression discontinuity analysis requires the least complex research design, but offers relatively weak theoretical conclusions. Nonlinear growth curve analysis offers the greatest flexibility and strong theoretical conclusions, but requires longitudinal data and complex statistics. Conclusions: Although examining schooling effects is challenging, effective methods are available. The methods are appropriate for different research designs, and differ in complexity, flexibility, data requirements, and strength of conclusions.
Daniel Brandeis (Dept. Child and Adolescent Psychiatry, University of Zürich, Switserland); Sylvia Brem; Martin Kronbichler; Matthias Schurz; Vera Blau; Joel Reithler; Sanne van der Mark; E. Schulz; K. Bucher; E. Martin; Rainer Goebel; Leo Blomert; Heinz Wimmer; Urs Maurer - Multicenter imaging of visual print processing deficits in dyslexia
Purpose: The visual word form system (VWFS) in the inferior occipito-temporal cortex of the left hemisphere often shows activation during print processing and reduced activation in dyslexia. Contrasting print to control stimuli such as symbol strings with matched tasks has revealed a posterior-to-anterior gradient of print specialization, but also resulted in less robust reading and dyslexia effects in the small samples used, which may be due to severity differences of dyslexia. Here we aim to investigate these effects in a larger, multicenter sample of children that also allow for selecting extreme groups. Method: Across the 3 centers, 154 children (7.8-12.2 years) viewed words and symbol strings and performed matched tasks while functional MR images were recorded. After excluding subjects with excessive motion and poor task accuracy, data from 48 dyslexic (<10th percentile) and 59 control children (≥20th percentile) were analyzed (SPM5) using literature-based ROI and voxel-based whole brain group and regression analyses. Group comparisons were also performed with extreme groups (dyslexia: <5th percentile; controls: ≥50th percentile). Results: Reduced word-specific activation in inferior occipito-temporal regions including the VWFS was found in the regression analyses and extreme group comparisons, but proved less robust without strict group definition. Robust compensatory effects in frontal regions in dyslexic children did not depend on extreme group selection. Conclusions: Deficits of visual print processing in inferior occipito-temporal regions increase with severity of dyslexia, and may not be prominent in milder forms of dyslexia.
David Braze (Haskins Laboratories);David Braze; James Magnuson; Einar Mencl; Kenneth Pugh; Whitney Tabor; Julie Van Dyke; Donald Shankweiler - RAN Predicts Reading Fluency in Oral but not Silent Reading
Purpose: This study contributes to understanding the specific nature of the association between Rapid Automatized Naming (RAN) and reading fluency. Method: We recruited young adults (N=155), oversampling poor readers to encompass a wide range of reading skills (cf. Braze et al., 2007). We focused on reading fluency outcomes at the passage- (GORT4), sentence- (WJIII, reading-fluency-subtest) and word- (TOWRE) levels. These provide indices of both oral (GORT4, TOWRE) and silent (WJIII-RF) reading fluency. Primary predictors of interest were RAN (CTOPP: letters, digits & colors), articulation rate (Ackerman & Dykman, 1993) and visual factors (visually-guided-saccade vs. anti-saccade task, Muñoz & Everling, 2004; visuo-spatial-memory, Corkin, 1974). Important covariates include: age, decoding skill, verbal-memory, print-experience, IQ. Results: In two models targeting oral reading fluency, RAN accounted for significant variance even controlling for articulation rate and visual factors, and other covariates (above); not so in the model targeting silent reading fluency. Visual factors predicted unique variance only in the model targeting silent reading fluency. Articulation rate accounted for unique variance only in the model targeting oral word reading fluency. Of other covariates, only print experience captures unique variance in fluency (all three fluency measures). An additional model targeting RAN shows that only articulation rate predicts unique variance after accounting for Age and visual factors. Conclusions: Articulatory and visual factors dissociate in their relationship to fluency. Articulatory factors dominate in oral reading; visual factors come to the fore only in silent reading. Print experience is an important precursor of both oral and silent reading fluency.
Silvia Brem (Department of Child and Adolescent Psychiatry, University of Zurich);Silvia Brem; Ulla Richardson; Silvia Bach; Christoph Hofstetter; Ernst Martin; Daniel Brandeis - Development of audiovisual integration of print and speech in normal and poor readers
Purpose The integration of multisensory information from speech and print is highly automatized in good readers but deficient in poor readers (Blau et al. 2009). Recent studies consistently point to the crucial involvement of the superior temporal sulcus (STS) in integration (van Atteveldt et al. 2004) and conceptual matching (Hocking and Price 2008) of incoming audiovisual information. In this study we aimed to track the development of audiovisual integration processes from kindergarten to adulthood in normal and poor readers. Methods Non-reading kindergartners (n=16, 5.7-7.1y), normal (n=31, 7.8-9.2y) and poor reading (n=15, 7.8-8.9y) 2nd graders as well as normal reading adults (n=15, 21.9-34.5y) were examined with functional magnetic resonance imaging (fMRI: 3T, standard ER-fMRI processing with SPM5). Incongruence effects determined by the statistical difference between incongruous and congruous audiovisual conditions are reported for each age group and for the differences between groups. Results While kindergarten children did not show an incongruence effect, normal reading 2nd graders and adults exhibited more pronounced activation in inferior frontal and superior temporal brain regions to incongruous than congruous items. The inferior frontal activation of the 2nd graders was furthermore correlated with reading performance. Conclusions In summary our results show that incongruous presentation of speech and print evokes a conceptual mismatch effect in frontal and temporal brain areas as soon as children's basic reading skills allow the matching of written and spoken information. The modulation of the matching effect by children's reading skills points to deficient integration of audiovisual information in poor readers.
Mindy Bridges (University of Kansas) ;Hugh Catts; Diane Nielsen - The use of a dynamic screening of phonological awareness to predict reading risk for kindergarten students
Purpose: This study was carried out to investigate the usefulness and validity of a dynamic screening of phonological awareness (DSPA) for the early identification of reading disabilities in kindergarten students. Method: The DSPA and the DIBELS Initial Sound Fluency (ISF) subtest were administered to a large sample of students at the beginning of kindergarten (N=161). Approximately 70% of the sample had scored in the at-risk range in initial school-based screening. Reading outcomes were measured at the end of kindergarten by administering several subtests from the Woodcock Reading Mastery Tests-Revised. Results: The distributional characteristics of the DSPA were compared to those of the ISF. Results showed that the ISF had a floor effect in the distribution of scores like that found in other studies (Catts et al., 2009). The DSPA, on the other hand, did not have a floor effect in its distribution. The predictive validity of the DSPA was also compared to that of the ISF. Results from both hierarchical and logistical regression analyses indicated that the DSPA added significantly to the prediction of reading outcomes over and above the ISF. Conclusion: These results provided preliminary support for the usefulness of the DSPA for the early identification of reading disabilities. Additional data will be presented from this sample, using outcome measures obtained at the end of first grade (to be collected in March/April 2010).
Monique Brodeur (Université du Québec à Montréal)Eric Dion; Marie-Ève Campeau; Anne Barette - Incidental vocabulary learning in a kindergarten phonological awareness and alphabetical program
Purpose: There remain controversies regarding the content and nature of the kindergarten curriculum. It has notably been argued that focusing instruction on the alphabetic code (AC) and phonological awareness (PA) is detrimental to language development. However, AC and PA are often taught by presenting words and teachers typically discuss the meaning of these words. This raises the possibility that AC and PA activities might actually foster language development, a possibility that was explored in our study. Method: Kindergarten classrooms (N = 19) were randomly assigned either to a control (regular activities) or an intervention condition. In this last condition, teachers implemented with adequate fidelity La forêt de l'alphabet, a French adaptation of Optimize Intervention Program (Simmons & Kame'enui, 2002). Student's (N = 256) knowledge of a random sample of words presented during AC and PA was individually assessed at pre- and post-test. Analyses were conducted separately for students considered not at risk, at low-risk, and at high-risk of reading problems. Results: Low- and high-risk students in the intervention condition made more gains on the vocabulary measure than their control counterparts, with small to moderate effect sizes (.24 - .46). Gains did not differed by conditions for not at risk students. Conclusions: AC and PA activities do not appear to slow down verbal development, on the contrary. Also, our results suggest that systematic AC, PA and vocabulary instruction could be profitably combined.
Ida Buch-Iversen (); Carsten Elbro - Effects of inference training on 6th graders' reading comprehension
Purpose: The study investigated possible causal links between inference making and reading comprehension. The main questions were whether a brief course in gap filling inferences would have robust, positive effects on inference making during reading and even on a standard measure of reading comprehension. Method: A quasi-experimental training study was conducted among 236 Norwegian 6th graders. Sixteen classes were randomly assigned to either an experimental or a control condition. Students in the experimental condition received inference training in eight 30 minute sessions provided by their ordinary teachers. The training focused on gap filling inferences and used graphic organisers to highlight the students' contribution to comprehension. The control group received ordinary teaching in language and literature during the same time. Results: Mixed between-within subject ANOVA's showed large interaction effects indicating that the inference training had a considerable impact on students' inference making skills and reading comprehension. The effects were robust across variation in gender, vocabulary, decoding, non verbal IQ, and completion of the training. Positive, delayed effects were found five weeks after completion of the training, indicating that the positive effects were maintained. No effects were found on students' maths abilities. Conclusions: The results suggest that inference making can be trained with specific effects on inference making during reading and with a sizeable impact on reading comprehension. The explicit focus on students' contribution to inferences (using graphic organisers) appeared to be helpful and may easily be implemented in ordinary comprehension instruction.
Adriana G. Bus (Leiden University, Netherlands); Anna Both-de Vries - It's all in the name
Purpose: From Levin's seminal research appears that name writing, a natural focus of most young children, is the first stable written form with meaning. Inspired by Levin's findings we investigated the role of name writing in children's acquisition of invented spelling. Method: Sixty 4- to 5½-year-old children from middle SES families with Dutch as their first language wrote their proper name and named letters. For each child we created unique sets of words with and without the child's first letter of the name to test (invented) spelling skills and phonemic sensitivity. Results: Results showed that the first letter of the name was among the first letters that are written phonetically when children begin to create invented spelling. Furthermore, name writing correlated with children's knowledge of the first letter of the name and phonemic sensitivity of the first letter of the name. Hierarchical regression analysis makes plausible that both knowledge of the first letter's name and phonemic sensitivity for this letter explain why name writing results in phonetic spelling with the name letter. Conclusions: Phonetic writing starts with the first letter from the name whichever it is. We guess that adults enable children to make the step towards phonetic writing with the first letter of the name by promoting knowledge of the first letter's name and phonetic sensitivity for this letter. They make such commends as: it's /p/ of peter.
Brian Byrne (University of New England, NSW, Australia)Dipti McGowan; William Coventry - Heritability of low- and high-end reading in the early years
Purpose: To compare high ability and low ability readers and prereaders in terms of proportions of variance in reading and print knowledge attributable to genes and aspects of the environment. Method: Monozygotic and same-sex dizygotic twin children were assessed for print knowledge in preschool and for reading in the first three school years. Using extremes analysis, low-end and high-end readers (and prereaders) were compared for the influence of genes and of shared and nonshared environment on the literacy measures. Results: Low-end reading tended to be less heritable and more influenced by the home/school ("shared") environment than did high-end reading. The difference was also seen for preschool print knowledge. Conclusions: Early in a child's development, low reading ability is more subject to environmental influence than is high reading ability. The same pattern is seen prior to the start of school, indicating that the effect is not limited to factors stemming from formal reading instruction. Reasons for this pattern are explored.
Kate Cain (Lancaster University); Hannah Nash - Making links: how do interclausal connectives influence young readers' on-line processing and comprehension of text?
Purpose. Connectives are cohesive devices, such as 'after', 'because', and 'but', that signal the semantic and conceptual links between clauses. Adults use connectives to guide the integration of clauses and their interpretation of a text's meaning. Although evident in speech from an early age, the comprehension of connectives has an extended period of development. We report the first systematic investigation of how connectives that signal different types of relation (temporal, causal, adversative) influence young readers' on-line processing and comprehension of text. Method. Readers aged 8 and 10 years completed two computer-based tasks. Experiment One: Children read two-clause sentences joined with an appropriate connective, an inappropriate connective, or 'and', e.g., "Amy wanted a dog but/after/and she was not allowed one." Each sentence was embedded in a short text. After each, readers judged whether or not the text made sense. Reading times and sense judgments were recorded. Experiment Two: Different groups read the same texts containing either the appropriate connective or no connective. Reading times and accuracy to comprehension questions were recorded. Results. Both age groups experienced disruption when reading texts containing two-clause sentences linked by 'and' or an inappropriate connective compared to an appropriate connective (E1) and facilitation when the two clauses were linked by an appropriate connective compared to no connective (two sentences) (E2). Conclusions. Connectives benefit young readers' processing and comprehension of text. Thus, a longer sentence in which the link between two clauses is explicitly signaled may sometimes be easier to understand than two short separate sentences.
Margarita Calderón (Johns Hopkins University) - Expediting reading comprehension for English language learners
Purpose: This paper highlights the components of instruction and professional development designs that math, science, social studies, and language arts teachers integrate language development and reading comprehension skills with their subject matter in order to help English-language learners (ELLs) succeed. Strickland and Alvermann (2004) suggest that much of the low achievement of language-minority students may be pedagogically induced or exacerbated. They remind us that the key contributing factor to the achievement gap is the differences in the quality of instruction students receive. Method: The five-year study was conducted in four middle and high schools and four matched schools. Data were collected for students with the Gates McGuinitie vocabulary and reading comprehension tests, the State language assessments, and the schools' yearly report card based on student attendance, AYP, and drop-out rates. Teachers were observed monthly using a digital pen and protocol to measure fidelity and quality of instructional components. The digital protocol generated monthly progress charts of the teachers' instructional practices and how these impacted their students' learning. Results: The students in the four experimental schools outperformed those in comparison schools and the schools went from low performing to the "Best School of the Year" after two years of these interventions. The digital protocol and professional development program is going to scale this year. Conclusions: Adolescent ELLs need instructional components that focus on the development of academic vocabulary, complex syntax understanding, and comprehension strategies (Calderón, 2007) taught by all teachers in the school, not just the English as a second-language teachers.
Lauren Capotosto (Harvard Graduate School of Education) - A meta-analysis of teaching comprehension-oriented reading strategies to less-skilled college readers
Purpose: Instruction in college developmental reading classes often involves teaching reading strategies, such as visualizing, summarizing, annotating, self-questioning, and activating prior knowledge. While the collective goal of teaching such strategies is to promote active engagement with texts, the strategies differ in the levels of comprehension they aim to achieve and the conditions under which they are enacted (Graesser, 2007). This quantitative research synthesis examines the effects of teaching various comprehension-oriented strategies on the reading comprehension of less-skilled college readers. Method: A comprehensive search of the literature retrieved 28 experimental and quasi-experimental studies, which yielded 39 independent comparisons. Using meta-analytic methods, this study explored whether heterogeneity in effects across studies can be explained by the types and characteristics of strategies. Results: Strategies differed in their effect on reading comprehension. While those that required self-explanations, such as relating text to prior knowledge (d = 0.60) and self-questioning (d = 0.24), had significant effects, there was no effect of teaching college students to visualize. Some strategies that help students organize information had no effect on reading comprehension when compared to alternative treatment (e.g., using graphic organizers), but significant and positive effects when compared to business as usual instruction. Furthermore, teaching students to apply strategies during breakdowns in reading comprehension had significant and positive effects (d = 0.31), whereas instructing students to apply strategies regardless of level of understanding had no significant impact. Conclusions: Reading comprehension strategies are not a homogeneous group. Results of this study have implications for college developmental reading instruction.
Markéta Caravolas (Bangor University, Bangor, Wales, UK)Brett Kessler, Washington University, St Louis, MO, USA;Charles Hulme, University of York, York, UK - The development of phonological spelling skill in alphabetic orthographies: The roles of syllable structure, orthographic depth and cognitive ability
Purpose: It is generally accepted that spelling acquisition is slower in opaque than in transparent alphabetic orthographies due to the greater learning challenge presented by orthographic inconsistency. Moreover, these differences seem to be present from the earliest stages of development when spellings primarily reflect the phonological information in words, and not the conventional orthography. Therefore it is important to determine the influence of orthographic depth, relative to oral language factors on the development of phonological accuracy of children's spelling productions. Method: We investigated: (1) the rate at which children learning orthographies with varying degrees of transparency (i.e., English (n = 180), French (n = 170), Spanish (n = 190), Czech (n = 150), Slovak (n = 200)) learned to produce phonologically accurate spellings; (2) the relative effects of orthographic and phonological stimulus attributes on patterns of growth, and (3) the roles of phoneme awareness and letter knowledge in the plausibility of children's spellings. Spellings of comparable word lists were collected at 3 six-monthly intervals (kindergarten - grade 1) from the 5 groups. Results: All spellings were evaluated with Ponto, an automated scoring tool indexing phonological accuracy. Preliminary analyses revealed that between-language differences in accuracy were initially mainly related to letter knowledge and phoneme awareness, regardless of orthographic depth. Moreover, growth in phonological spellings was mediated by the typicality of the length and syllabic complexity of the words being spelled. Additional analyses will examine in greater detail the effects of stimulus-based attributes versus participant-based variables, and how their effects change over time.
Cláudia Cardoso-Martins (Universidade Federal de Minas Gerais, Brazil); Tereza Cristina L. Mesquita; Linnea Ehri - Does phonological awareness taught to children with letter name knowledge boost their learning of letter sounds? Experimental evidence from Portuguese-Speaking children
Purpose: The present studies investigated the role played by letter-name knowledge and phonological awareness (PA) in the learning of letter sounds. Methods and Results: Three experimental studies with Portuguese-speaking preschoolers in Brazil were conducted. Study 1 involved two groups: Children in the experimental group learned the names of letters whose sound occurs either at the beginning (e.g., the letter /be/) or at the middle of the letter name (e.g., the letter /'eli/). The control group learned the shapes of the letters but not their names. Then both groups learned the sounds of the letters. Results showed an advantage for the experimental group, but only for beginning-sound letters. Studies 2 and 3 investigated whether training in PA can boost the learning of letter sounds, particularly of middle-sound letters. In addition to learning the names of beginning- and middle-sound letters, children in the experimental group were trained to categorize words according to rhyme and alliteration, while controls were trained to categorize the same words semantically. All children were then taught the sounds of the letters. Results showed that the children who had been successfully trained in PA found it easier to learn the sounds of both types of letters than controls. Conclusions: Children benefit from their letter name knowledge in learning about letter sounds, regardless of their level of PA. However, PA does contribute to the learning of letter sounds, especially for letters whose sound occurs in the middle of the letter name.
Joanne Carlisle (University of Michigan)Dan Berebitsky - Do elementary teachers in high poverty schools take advantage of opportunities to support students' word learning?
The purposes were to study the extent to which teachers take opportunities to help their students learn word meanings during literacy instruction and the extent to which such instruction is related to characteristics of the students in the class (e.g., % free/reduced price lunch (FRL), %LEP, %minority). The literacy block of 88 second and third grade teachers was observed four times during the year. Observers coded teachers' explanations of word meanings during phonics, fluency, writing, vocabulary, and comprehension lessons; coded were instances where (1) teacher defines word, (2) teacher explains meaning of a word in context, (3) teacher asks students to use a word in a sentence, (4) teacher asks students to explain word meaning, and (5) teacher fosters discussion of word meaning. Analysis showed that one or more of the five word meaning interactions appeared in only 37% of the lessons. The most frequent interaction was asking students to use a word in a sentence (22%), followed by teacher defining the word (19%). We ran a series of five HGLMs to determine lesson and classroom factors that accounted for the use of each word meaning interaction; % FRL and % minority were negatively related to word meaning interactions of different types. Similarly, word meaning interaction "density" was negatively correlated with percent FRL and minority and not significantly related to LEP. In short, the higher the percentages of students with FRL or minority, the less likely it was that teachers initiated support for word learning, contrary to our expectations and to desirable practice.
Julia Carroll (University of Warwick) - The links between phonological processing, language and literacy: do they vary according to different samples?
Purpose Previous research indicates that there are close links between language and literacy development, particularly in young and high risk samples. The present study examines the predictive role of language in 4-6 year old children with a range of risk factors for future literacy difficulties. Method 212 children were given a battery of tasks measuring speech, language, literacy and phonological processing. Their literacy skills were retested after a six month interval. 120 of the children were selected for inclusion because of a history of speech and language difficulties, family history of dyslexia, or both. Results Language and phonological processing are both major predictors of progress in literacy across both the typically developing and the high risk groups. Structural equation modelling will be employed to assess whether the relationships between the variables differ in the younger and older children, and in the high risk versus low risk groups. Conclusions This research elucidates the relationships between phonological processing, language development and literacy in children around school entry age.
Séverine Casalis (Universite de Lille 3 Charles de Gaulle)Pascale Colé - Morphological priming in visual word recognition in French
Purpose: The study aims at examining if and when morphological information is activated when French children read words. Although previous studies have indicated that children do use the morphological structure, there is a need for clarity regarding when this information is activated in the word reading process, and what kind of information is extracted. The priming paradigm, contrasting various prime duration associated with a lexical decision, allows to track activation of information. Method: 75 third- and 75 fifth-graders participated to the study. They had to perform a lexical decision task with visual priming. There were three prime durations: 55 ms, 80 ms, 250 ms. Stimuli were 45 4-to-5-letter word targets. Targets were short high frequency base words. Three kinds of primes were paired with each target item: morphological (collage-COLLE 'gluing-glue'), orthographic (college-COLLE 'college-glue'), control (seringue-COLLE, 'syringe-glue'). Three versions of the experiment were constructed, with items counterbalanced across conditions. Results: Facilitation effects were observed for both morphological and orthographic priming at 55 ms (grade 5 only) and 80 ms (grades 3 and 5) prime duration. At 250 ms, for both grades, only the morphological priming lead to facilitation. In addition, amplitude of priming effects was very similar notwithstanding frequency of the prime. Conclusion: This study testifies to early activation of formal information and persistence of morphological activation in child word reading. Absence of effect of prime frequency suggests a prelexical influence.
Anne Castles (Macquarie Centre for Cognitive Science); Kristy Jones; Erin Barriball - Response to phonics and sight-word training as a function of reading reliance
Purpose: Even when significant group effects are found, there is typically a large about of variability in the response of children with developmental dyslexia to reading training. The aim of this study was to determine whether children's initial pattern of strengths and weaknesses in reading sub-skills might be a factor that modulates the success of phonics and sight word training. Method: Ninety children with dyslexia were divided into different "reliance" groups, based on whether they had particular weaknesses in phonics skills, particular weaknesses in sight word skills or weaknesses in both skills. The groups then did 16 weeks of training, with half of the time spent on phonics instruction and the other half on sight-word reading instruction. Results: Preliminary results indicate that the children with particular weaknesses in phonics skills benefited more from the phonics training than the sight word training, while those with particular weaknesses in sight-word skills benefited more from the sight-word training than the phonics training. Those with weaknesses in both skills benefited most from the training overall, and did so equally for phonics and sight-word instruction. Conclusions: It appears that the effects of intervention are modulated by reading reliance, and that it is beneficial for reading interventions to focus on a child's relative weaknesses rather than on their strengths.
Yu-Jen Chang ();Shihjay Tzeng; Sujan Lin; Suli Chen; Hwawei Ko - A tracking study of the relationship between rapid naming and chinese reading ability
Purpose- The effectiveness of naming speed tests utilized to screen Chinese reading disabilities (RD) was explored in this study. Method- We administered a rapid digit naming test to 2207 kindergartners averagely aged 75 months in 2004. A group of 66 children whose digit naming speed ranked the last 5% were then selected as slow namers (SN). A group of average namers (AN), one-on-one matched with the SN children by age, IQ, and social economic status, served as the comparison group. In 2007, when children were in grade 3, scores of the Chinese Phonological Awareness Test, Chinese Character Recognition Scale, and Chinese Reading Comprehension Test were collected. Diagnostic procedures of RD were also carried out to check if the two groups had different possibilities being identified as having RD. Results- The results showed that: 1.The SN group was more likely being identified as having RD than the AN group. 2.Although both groups showed a substantial growth in naming speed, but the SN lagged behind their counterparts by two years. 3.For the SN children, their naming speed in each grade is a valid predictor of reading comprehension, character recognition, and phonological awareness (measuring the awareness of Chinese phonological features such as onset, glide-rime combination, tone, and general PA). However, for AN children, naming speed could only predict general PA, onset and glide-rime combination. Conclusions- These findings support the validity of using naming speed tests as tools for early screening of Chinese reading disabilities. Suggestions for future search are addressed.
Victoria Cheah (Faculty of Education, University of Cambridge, UK); Jarmo Hamalainen; Fruzsina Soltesz; Usha Goswami - Amplitude envelope perception and prosodic sensitivity to syllable stress in developmental dyslexia
Purpose: We have proposed that dyslexics have impaired perception of slower modulations of the amplitude envelope that are important for speech intelligibility. Rise times in the speech envelope are correlated with the onsets of syllables, and associated with the perceptual experience of speech rhythm and stress. Difficulties in rise time perception should affect sensitivity to syllable stress patterns in speech. Method: Dyslexic and non-dyslexic adult participants made same-different judgments about the position of syllable stress in pairs of 4-syllable words. In Experiment 1, participants heard the same word pronounced twice, either with correct or incorrect syllable stress placement (eg. "MIlitary" - "miLItary"). In Experiment 2, participants compared stress placement across two different words (eg. "MIlitary" - "SEcondary"). Participants also received non-speech psychoacoustic measures for amplitude envelope onset (rise time), frequency and intensity discrimination, and standardised tests of IQ, phonological awareness, reading and spelling. Results: Compared to controls, dyslexics had marked deficits in their ability to perceive syllable stress patterns both when auditory on-line processing was sufficient to answer correctly (eg. "MIlitary" vs "miLItary"), and when they had to abstract and compare syllable stress patterns across different words (eg. "MIlitary" vs "SEcondary"). Dyslexics had significantly higher auditory thresholds for rise time and frequency, but not for intensity. Prosodic sensitivity to syllable stress was significantly related to rise time and frequency thresholds, and to literacy and phonological measures. Conclusions: We conclude that auditory processing difficulties in dyslexia are associated with reduced sensitivity to syllable stress, impaired phonological awareness and poor outcomes in literacy.
Aleck Shih-wei Chen (National Cheng Kung University, Taiwan); Jensen Chia-sen Tsou - Measuring Phoneme Awareness in Chinese-speaking Children
Chinese-speaking children learning English as a foreign language (EFL) have been found to tend to subdivide an English syllable into a core syllable (CV) plus appendices, i.e., any consonants preceding or following the core syllable. This implies that not all items traditionally measuring phoneme awareness are equally valid for this population. More specifically, it may be only items requiring segmentation of the core syllable that are measuring true phoneme awareness in Chinese-speaking EFL children. Fifty-one Chinese-speaking EFL second graders were measured for their performance on various EFL literacy tasks at two time points four months apart. The results showed that T1 performance in core syllable items, which require removal or isolation of the initial consonant from a CVC syllable, is highly correlated with both T1 (concurrent) and T2 (longitudinal) measures of English nonword reading (respectively .71 and .70) and spelling (respectively .63 and .69), all ps < .001. Performance on non-core-syllable items requiring segmentation of non-core-syllable phonemes, in contrast, is not correlated either concurrently or longitudinally with either English nonword reading or spelling, all ps > .10. The results suggest that only core syllable items should be used in phoneme awareness tasks when Chinese-speaking subjects are involved.
Hsiu-Fen Chen (Special Education Center, National Taiwan Normal University); Li-Yu Hung; Hsin-Yi Chen; Yu-Huei Huang; Sheng-Min Cheng; Su-Jan Wong - Reading interventions outcomes for students with LD in Taiwan: A quantitative synthesis of single-subject researches
From a cross-cultural perspective, the purpose of this study was to enrich current understanding by synthesized the reading interventions conducted between 1995 and 2006 with elementary and secondary school students with LD in Taiwan. Interventions addressing decoding, comprehension, and writing were included if they measured the effects on reading comprehension. A total of 29 single-subject studies, which produced 206 effect sizes, were acquired by a through search of major journals and computerized databank, but not included the studies with A-B design. PND (percentage of non-overlapping data) nonparametric approaches proposed by Mastropieri & Scruggs (1985-86) were employed to compute effect sizes. The important findings of this study are as follows: (1) the grand average effect size were .83 (SD=.32), which was considered as a moderate effect; (2)in the decoding-level, the Stem-Deriving Instructions of Chinese characters were the most effective treatments, whereas the Phonological-Awareness Instructions were the least effective treatments; in the comprehension-level, the single-strategy interventions had higher effect sizes than multi-strategy ones; (3) from the students' characteristics, G3-4 were found to be the most benefitted grades from the treatments, which yield the highest effect sizes; (4) among the various instructional strategies collected, the most effective strategies were (a) adjust the teaching objectives (b) inspire the interests of the students (c)utilize classmate assistance (d) use the multimedia (e)conduct self-management. Implications for comprehension instruction for the struggling readers are described.
Xi "Becky" Chen (University of Toronto)Gloria Ramirez; Cathy Yang Luo; Esther Geva - Morphological awareness and vocabulary in English Language Learners (ELLs): Evidence from Spanish- and Chinese-Speaking children
Purpose: The present study compared the relationship between morphological awareness and vocabulary in English Language Learners (ELLs)who speak Chinese or Spanish as their first language, and English-speaking monolinguals. Method: Participants included 78 English-speaking monolinguals, 76 Chinese-speaking ELLs, and 90 Spanish-speaking ELLs from Grade Four and Grade Seven. A battery of tests was administered including derivational morphological awareness, vocabulary, word reading, phonological awareness, and nonverbal ability. Family background information was collected through a questionnaire. Results: The study produced two important findings. First, we compared the performance of Chinese- and Spanish-speaking ELLs on English words that had cognates in Spanish and English words that did not. The two types of words were matched on frequency. After controlling for nonverbal ability and maternal education, Spanish-speaking ELLs outperformed Chinese-speaking ELLs on English cognate vocabulary, whereas the two groups on non-cognate did not differ on non-cognate vocabulary. Second, we found that in Chinese- and Spanish-speaking ELLs, as well as English-speaking monolinguals, morphological awareness made a significant contribution to vocabulary knowledge after controlling for nonverbal ability, mother's education, and other reading related variables. Conclusions: The results indicate similarities and differences in English vocabulary learning in ELLs from different language backgrounds. Because Spanish and English share cognates, Spanish-speaking ELLs are able to use first language vocabulary knowledge to learn English cognate vocabulary. This strategy, however, is not available to Chinese-speaking ELLs. On the other hand, the results suggest that similar relationships between derivational morphological awareness and vocabulary exist in different ELL groups and English-speaking monolinguals.  
Wing-Yin Bonnie Chow (City University of Hong Kong); Connie S.-H. Ho; Mary Waye; Dorothy V. M. Bishop - Longitudinal twin study on Chinese reading development: Stability and change in genetic and environmental influences
Past twin studies have shown the genetic and environmental stability and change in reading alphabetic scripts, mainly English (e.g., Byrne, et al., 2009). Our study extends this line of research by investigating these developmental trends in Chinese, a language which has contrasting characteristics with alphabetic languages, with a longitudinal twin study design. This study included 301 typically developing Chinese twin pairs aged from 3 to 11 in Hong Kong. Children were given tasks of Chinese receptive vocabulary, rapid automatized naming, phonological awareness, morphological awareness, word reading and nonverbal IQ individually. They were tested on the same tasks again one year after the initial testing. Zygosity was determined by SNP testing, and 221 pairs of monozygotic twins and 80 pairs of same-sex dizygotic twins were identified. Data controlling for age and IQ were analyzed with the Mx statistics package. Findings showed that genetic influences were stable across the one-year period on all tasks, except receptive vocabulary, and there was a new genetic factor contributing to word reading at time 2. Stable shared environmental influences were demonstrated in receptive vocabulary and phonological awareness, and specific environmental factors were found in time 2 rapid automatized naming, morphological awareness, and word reading. This study is the first longitudinal twin study of learning Chinese. It will be key to understanding the underlying causes of stability and change in Chinese reading development, and will shed light on the universal or specific factors of reading acquisition across languages. (241 words)
Micaela Christopher (University of Colorado, Boulder); Barbara Wise - Can setting goals to read more help struggling readers increase their motivation and ability?
Purpose: While research has shown that struggling readers can benefit from explicit reading instruction, how to increase motivation for reading has not been explored extensively. Evidence exists that increased reading time correlates with ability and that higher ability relates to motivation. Given these interrelations, the present study hypothesized that increasing reading times may also increase motivation and ability. Method: The study developed four ten-minute exercises to help children set goals to increase time reading during school year 2009-2010. All exercises were presented on a computer. 107 1st to 5th grade students participated. Half the participants did goal-setting while the other half did control exercises matched on length and presentation. All children also participated in a reading intervention. Pre-intervention scores on standardized tests were obtained Fall 2009. Changes in scores over the year as well as motivation and time reading will be recorded. Results from a preliminary study from school year 2008-2009 found that setting goals increased intrinsic motivation and self-efficacy. In addition children who set goals had higher gains on some measures of reading. However limited power and problems with timing of standardized testing may have depressed the results. The present study uses a larger sample size and improved timing to correct for these problems. Results: Analyses will use reading times as well as gains on standardized reading tests and motivation questionnaires to test if setting goals increases motivation, reading times, and reading ability. Conclusions: This project could provide an additional method for interventions to increase participants' motivation for reading.
Hui-Kai Chuang (); R. Malatesha Joshi; L. Quentin Dixon - Is Mandarin Chinese knowledge transferred to English reading proficiency among Taiwanese junior high school students?
Purpose The present study examines whether Taiwanese ninth graders' Mandarin (L1) reading ability is transferred to their English (L2) reading competence. Method The thirty thousand Taiwanese adolescents included in this study were randomly selected from a pool of national standardized examination (Basic Competency Test - BCT) participants who took the exam in a consecutive six-year period. The BCT formats for Mandarin and English included multiple-choice questions with an emphasis on contextual reading competence. Results Using multiple regression analysis, we found that Mandarin reading knowledge had a positive and significant relationship with Taiwanese Grade Nine students' English reading proficiency, controlling for gender, school district, and number of years studying English. Both gender and school district had a certain level of influence on cross language reading ability; English exposure, by contrast, did not have statistically significant effects on reading competence in English. Thus, the results indicate that reading knowledge in L1 predicts Taiwanese adolescents' L2 reading comprehension. Conclusions Our findings support Cummins' linguistic interdependence hypothesis (1979) that proposed students with better-developed L1 reading skills will acquire reading abilities in L2 faster than someone with less well-developed L1 reading skills. However, our findings contrast with previous studies which indicated that there is a reading ability correlation across languages only when both languages have similar orthographic system (e.g., Spanish and English) (Bialystok, Luk, & Kwan, 2005; Bialystok, McBride-Chang, & Luk, 2005). The findings suggest educational policy makers that boost native-Chinese speaking students' Mandarin knowledge helps support their development of English reading ability as well.
Kevin Kien Hoa Chung (Department of Special Education and Counselling, The Hong Kong Institute of Education)Connie S H Ho: David W Chan: Tsang S M: Lee S H - Effects of kindergarten intervention for the prevention of Chinese children with reading difficulties
Purpose: This study was to evaluate the impact of an early reading program for kindergarteners at risk of reading difficulties. Method: One hundred and ten children in Kindergarten Grade 3 were identified as at risk of reading disability. They were divided into 2 groups, the at-risk group (AR) and the control-risk group (CR). The AR received a centre-based intervention program that contained 15 training sessions on the four domains: oral language, orthographic knowledge, morphological awareness and character recognition. The CR received self-learning materials related to their school curriculum and required to be completed them at home. A further group of 50 children (NAR), who were not at risk, were selected to act as controls. Pre-and post-testing included measures for character recognition, story retelling, morphological construction, semantic categorical judgment, left-right reversal and word repetition. Results: Before the intervention, both AR and CR performed at a similar level on the pretest measures but worse than NAR. After the intervention, though AR and CR did not catch up with NAR, AR showed greater improvement in the posttest measures than CR. Significant growth was found in the domains of orthographic awareness, morphological awareness and character recognition. Findings from this study showed that children at risk in the kindergarten years could achieve a higher reading ability by receiving an early intervention program. Conclusions: The results highlighted the important role that early reading intervention plays in the development of reading skills for children at risk of reading difficulties.
Cassandra Coddington (University of Alabama)John T. Guthrie - Effects of Adolescents' Motivations for Reading on Comprehension and Grades
Purpose The purpose of this study was to determine whether two affirming (intrinsic motivation and self-efficacy) and two undermining (avoidance and perceived difficulty) motivations for reading independently contributed to predicting reading comprehension and classroom grades. Method Participants included 257 adolescents in the 7th grade (125 males; 132 females) who completed a self-report measure of motivation for school reading and the Gates Mac-Ginitie Reading Comprehension subtest. Reading/Language Arts (LA) grades were obtained for all participants. Two multiple regression analyses were conducted with the affirming motivations entered in one block and the undermining motivations entered in a second block as the independent variables. The dependent variables were scores on the Gates-MacGinitie and reading/LA grades. Results In both regressions, adding the undermining variables explained significant additional variance in the dependent variables. The beta for perceived difficulty was negative and statistically significant when predicting reading comprehension scores, after taking into account the effects of the other three motivation variables. The betas for intrinsic motivation (positive) and perceived difficulty (negative) were statistically significant when predicting reading/LA grades, after taking into account the effects of the other two motivation variables. Conclusions The results indicate that undermining motivations contribute to predicting achievement, even when the effects of affirming motivations are taken into account. Perception of difficulty in reading was a significant predictor of reading comprehension and grades. These results have implications for teachers and emphasize the importance of developing strategies for counteracting motivations that undermine reading achievement.
Lucia Colombo (Department of Psychology, University of Padua, Italy) - The acquisition of distributional characteristics of stress assignment in children.
Italian word stress assignment must be learned, as most three- and four syllable Italian words are stressed on the penultimate syllable (dominant stress pattern), while a minority is stressed on the initial syllable. The acquisition of information about stress neighborhood (i.e., words that share the final phonemes and stress pattern with a target; Colombo, 1992; Burani & Arduino, 2004) and of the dominance of the stress pattern on word and nonword reading was investigated in children, to find how distributional characteristics of Italian stress assignment are acquired. Children should initially be more influenced by general characteristics, like relative frequency of each stress pattern. With the increase in age the effect of stress neighborhood should increase. Twenty-nine children aged seven and 22 children aged nine read aloud dominant and non-dominant stress words with different degrees of stress neighborhood consistency, and nonwords with a strong stress neighborhood. Latencies and errors were measured for words, while for nonwords the proportion of dominant stress assigned was calculated. Results of the ANOVAs showed faster latencies and fewer errors on stress consistent than stress inconsistent words, but the effect was larger in older, compared to younger children. Younger children were affected by stress dominance in assigning stress to nonwords, but the effect of stress neighborhood on nonwords increased with age. Consistently with the hypothesis, distributional characteristics of stress are gradually learned by children, with more general information being acquired earlier, compared to the acquisition of more specific information about phonological (suprasegmental) similarity among words (i.e., neighborhood).
Eva Commissaire (Laboratoire URECA)Lynne G. Duncan; Severine Casalis - Lexical and sublexical orthographic interactions across languages in young learners of a L2
The bilingual literature has provided clear evidence for a language non-selective lexical access in visual word recognition. To address this issue, the processing of cognate words has been under study and evidence has been found for a cognate facilitation effect in lexical decision tasks. Although interactions between both languages have been found at the lexical level, the issue of sublexical units - particularly at the graphemic level - in bilingual word recognition has received far less attention. Our purpose was to investigate L1-L2 lexical interactions in the developing orthographic lexicon by investigating cognate words processing, as well as interactions at the sublexical level by manipulating L1 bigram frequency in L2 words. A central aim was to examine the extent to which the lexical and sublexical orthographic representations that the learner is constructing are connected to L1 representations. Grade 8 French children learning English as a L2 had to perform a L2 lexical decision task. Two variables were manipulated: the status of the word - cognate word ("silence"), L1 typical English words ("house") and L1 atypical English words ("white") - and its frequency - low and medium frequency. Two main results were found (1) a cognate inhibition effect, cognate words being responded to slower than atypical L2 words and (2) a L1 typicality effect on L2 learning: L2 words with typical L1 bigrams being responded to faster than L2 words with L1 atypical bigrams These results shed some light on the lexical and sublexical interactions between the L1 and L2 during L2 word recognition acquisition.
Nadia Cone (Northwestern University); James Booth - Phono-orthographic and attentional control factors in intra- and cross-modal rhyme decision tasks in reading-disabled versus typically-achieving children
Purpose: In this study, we examine differences between children with reading disabilities (RD) and typically-achieving (TA) children in the interaction of phonological and orthographic information, as well as the relationship between attentional control and phono-orthographic interaction. Method: In rhyme decision tasks, word pairs are presented with manipulations of presentation modality (auditory, visual, or one word in each modality) and phono-orthographic conflict (non-conflicting pairs being either similar or dissimilar in both their phonological and orthographic endings, e.g., DIME-LIME or STAFF-GAIN, and conflicting pairs being similar in only one regard, e.g., PINT-MINT or JAZZ-HAS). In attention tasks, judgments are based on only one dimension of stimuli (volume of tones in an auditory task, or color saturation of squares in a visual task) despite variations in a task-irrelevant dimension (pitch or brightness), and interference effects of variations in irrelevant dimensions are measured. Results & Conclusions: In the rhyme tasks, RD children are expected to experience greater performance impairments in the conflicting versus the non-conflicting conditions than are TA children, particularly when words are presented visually. This might suggest phono-orthographic processing as a potential site of breakdown in reading disabilities. Stronger interference effects in attention tasks are also expected in the RD group, indicating that these children may be less capable of controlling attention to relevant versus irrelevant nonlinguistic information. Stronger correlations in the RD group between conflict effects in rhyme tasks and interference effects in attention tasks might indicate that children with reading disabilities are more susceptible to interference from task-irrelevant information, whether linguistic or nonlinguistic.
Carol Connor (Florida State University/FCRR)Elizabeth C. Crowe; Frederick J. Morrison - Refining dynamic forecasting intervention models: new child X instruction interactions for first graders
Purpose: This study examined child characteristic by instruction (CXI) interactions in first grade and then tested new models in a randomized control field trial (RCT). Dynamic forecasting intervention models compute recommended amounts and types of instruction based on students' assessed skills Methods: The longitudinal correlational study (n = 461 children, 47 classrooms, 10 schools) utilized video-taped classroom observation data of first grade classrooms (2005-2006) coupled with fall and spring assessment of children's vocabulary and reading (WJ-III Tests of Achievement). Using multiple dimensions of instruction (teacher/child- versus child-managed and code- versus meaning focused-), content and duration of instructional activities were coded for each child. Using HLM, CXI interactions were tested. From these models, we developed. An RCT with random assignment of teachers (n = 33 in 8 schools, 446 students) tested the models. Results: Correlational: More time in teacher/child managed meaning-focused instruction was associated with stronger reading outcomes (main effect). In general child-managed code-focused instruction was associated with stronger word reading gains only for children reading below grade expectations at the beginning of first grade. For children reading at or above grade level, more time in these activities was associated with less growth (CXI). RCT: Children in treatment classrooms demonstrated greater gains in word reading, passage comprehension, and reading comprehension (as assessed by the Gates-MacGinitie) with an effect size (d) .39 compared to peers in control classrooms. Conclusions: Results suggest that we can be more precise in predicting optimal patterns of instruction for students based on assessed language and literacy skills.
Nicole Conrad (Department of Psychology, Saint Mary's University); Wafa Saoud; Laura Scallion - A comparison of orthographic learning following spelling and reading practice
Purpose: The acquisition of orthographic knowledge is a key determinant of literacy. The self-teaching hypothesis (Share, 1995) suggests that orthographic learning occurs during independent reading; the decoding of words enables connections between letter patterns and sounds to be established in memory, thus providing the orthographic representations necessary for fluent reading. However, because spelling requires the production, rather than the recognition of letter sequences, spelling practice may produce better quality orthographic representations than reading. The present study examined whether orthographic learning occurs during spelling through a direct comparison of orthographic learning following reading and spelling practice. Method: Children in Grade 2 either read or spelled one alternate spelling of eight pairs of pseudowords (e.g., slirst - slurst). The target items were embedded within stories. Children first learned the eight new "words" using simple line-drawings of non-objects. For the readers, the target words were repeated four times in each story. For the spellers, rather than the words appearing, the line-drawings were repeated four times and children were to spell the name that corresponded with the drawing. Following the practice phase, two measures of orthographic learning were administered. Results: Orthographic learning occurred for both readers and spellers on a spelling production task, but not on a spelling recognition task. Orthographic learning following spelling was greater than following reading. Generalization to new non-words with practiced orthographic patterns occurred for both readers and spellers. Conclusion: Results indicate that orthographic learning occurs during spelling, and to a greater degree than during reading. Spelling practice may be very useful for developing reading skill.
Piers Cornelissen (University of York, UK); Katherine Wheat; Peter Hansen - Further MEG explorations of early activation in Broca's area during visual word recognition
Purpose. Debate surrounds the precise cortical location and timing of phonological access in visual word recognition. Timing estimates in particular vary widely from ~80ms (e.g. Ashby et al., 2009) to 200-300ms (e.g. Sereno et al., 1998). To address this debate, we used MEG to measure brain responses during a masked pseudohomophone priming task. Method. Participants rapidly and silently read five-letter target words preceded by masked primes. Prime-target pairs were divided equally between pseudohomophones (brein-BRAIN), orthographic controls (broin-BRAIN), and an unrelated condition in which neither letters nor sounds were shared (lopus-BRAIN). No overt response was required except for randomly interspersed catch trials. Neural sources of activity were reconstructed with a vectorized linearly-constrained minimum-variance beamformer. We compared the time-frequency domain responses between experimental conditions in 6 ROIs: left and right middle occipital gyrus (MOG), left and right inferior frontal gyrus (IFG), and the left and right mid-fusiform gyrus. Results. All three conditions generated activity ~100ms in the left IFG. The critical comparison between pseudohomophones and orthographic controls shows that shared phonology between prime and target, over and above shared orthography, results in significantly greater non-phase locked (i.e. induced) 30-40Hz activity in left IFG ~100ms after stimulus presentation. No such differences were found in MOG. Conclusions. Early involvement of IFG in pseudohomophone priming supports a role for this site in prelexical access to phonological information during visual word recognition. Moreover, this finding suggests that early word recognition may be achieved by a direct print-to-speech mapping mediated by a speech production code.
Emily Coyne (Nottingham Trent University); Lee Farrington-Flint; Jean Underwood; James Stiller - The influence of rime-unit frequency on children's lexical word-reading strategies
Purpose - Despite recent attempts to outline the development of lexical and non-lexical word reading strategies on a global-level (e.g., Castles et al., 2009; Farrington-Flint et al., 2008), there has been fewer attempts to monitor this shift in lexical reading strategies based on word-specific orthographic features (c.f., Share, 1999). Through the manipulation of word items based on their rime-unit frequencies, the current work specifically examines the development of lexical and non-lexical word-reading strategies in young primary school children. Method - An examination of lexical and non-lexical reading strategies were carried out with 61, 6-to-8 year-old's using retrospective verbal reports. All children completed an experimental word-reading task comprising 45 single word items that varied in their rime-unit frequency (comprising consistent, exception and unique word items). Results - The data shows how rime-unit frequency has a significant influence on the development of lexical and non-lexical word-reading strategies. While performance on the consistent word items remained fast and efficient (compared to performance on the unique or exception word items), lexical word-reading strategies were most frequent for words that has irregular GPC rules and fewer rime neighbours. The use of lexical word reading strategies was dependent on the child's orthographic representations of specific word items (namely, exception word items). Conclusions - The data suggests the importance of assessing acknowledging the development of stored orthographic representations when assessing the use of lexical (and non-lexical) word-reading strategies and informs educational practice of promoting self-reflection and self-explanation of strategy success.
Michael Coyne (University of Connecticut) - Building vocabulary and listening comprehension through storybook reading
Purpose: This presentation will present an overview of a school-based intervention research study evaluating a year-long implementation of vocabulary instruction with kindergarten students. Method: Participants included 128 students from three elementary schools that serve diverse populations. Students assigned to the treatment group (n = 81) were taught the meanings of 54 vocabulary words over 36 half-hour instructional lessons (two lessons per week over 18 weeks). During instruction, students listened to a storybook read aloud. When target words were encountered, students were provided with a simple definition which was then used in the context of the story. After each reading of the storybook, teachers engaged students in activities that provided them with extended opportunities to interact with and discuss target words in varied contexts beyond those offered in the story. Another 47 students served as a no-treatment control group. Posttest assessments included measures of target vocabulary knowledge, overall receptive vocabulary, and listening comprehension. Results: Results revealed significant differences favoring the treatment group on all measures. Secondary analyses indicated that students with lower levels of initial vocabulary knowledge did not respond as strongly to the intervention. English learners scored statistically significantly lower on all measures than their peers although this effect was mediated by initial English receptive vocabulary knowledge. Conclusions: Integrating vocabulary instruction into interactive read alouds appears to be a promising way to boost student vocabulary. Overall, findings will be discussed within a response to intervention (RtI) or multi-tier approach to supporting vocabulary development with young children.
Jennifer Cromley (Temple University); Theodore W. Wills; Ilyse Resnick; Ting Dai; Anthony C. Perez; Shannon Fitzhugh; Nora Newcombe; Natalie Ramos-Castillo - Reading comprehension--sequences of cognitive moves while reading scientific text
Purpose: Reading comprehension strategies researchers describe the proficient reader as using strategies-such as monitoring, re-reading, self-questioning, and summarizing-adaptively. Descriptions of adaptively flexible reading derive from qualitative descriptions in novice-expert studies. Only rarely (e.g., McNamara, 2004) have sequences of strategic activity during reading of authentic text been quantified. Method. We collected data from N = 11 students in an introductory Earth science course (64% female, mean age 23.4 yrs., 91% White, 45% first-generation college). Participants thought aloud while reading a chapter in their own textbook. We collected pre- and post-reading knowledge measures, and analyzed transitions among low-level and productive strategies, activation of background knowledge, inferences, vocabulary knowledge, and word reading accuracy. Results. Posttest scores were bimodal, with an expected high-knowledge/-inference/- comprehension group (HC; n = 5) and a low-knowledge/-inference/-comprehension group (LC). We coded 1167 utterances; transition analyses showed three patterns disproportionate to each participant's own utterances, significant by a Mann-Whitney U test: 1) LCs disproportionately showed transitions to a variety of variables following inferences (HCs invariably transitioned to productive strategies), 2) HCs showed these varying transitions following verbalization of background knowledge, and 3) LCs showed a larger number (Mdn = 4 vs 2) of variables with disproportionate patterns. Conclusions. Think-aloud protocols from Earth science majors reading from their own textbook suggest particular patterns in transitions among background knowledge, use of productive strategies, and inference. These patterns are consistent with both the "adaptively flexible" model of reading comprehension and with models such as Kintsch's Construction-Integration model in which strategies mediate the relationship between background knowledge and inference.
Virginia Cronin (George Washington University) Maha Khalid (maha@gwmail.gwu.edu) Membership status: Non-member Maha Khalid (maha@gwmail.gwu.edu) Membership status: Non-member Maha Khalid (maha@gwmail.gwu.edu) Membership status: Non-member Maha Khalid (Maha@gwmail.gwu.edu) - Grain size, RAN, and PA
Purpose - Ziegler & Goswami (2005; 2006) suggest that grain size, or the combination of graphemes and phonemes that make up words, develop differently in English speaking children than in children from the more regular European languages. In consistent languages the pattern is from small to large while in English grain size develops in both small and large sizes initially. This research examines grain size development and the influence RAN and PA has on the process. Method - 130 Children were followed from preschool through fifth grade when 84 remained in the study. There were tested 10 times and were given the Woodcock word identification, word attack, and passage comprehension tasks They were formed into groups varying in PA and RAN by preschool and kindergarten tests. Simplex modeling was used to analyze the mutual relationships between word, pseudoword, and passage comprehension development. HLM were used to analyze reading acquisition in the no-deficit, low-RAN, low-PA, and double deficit children. Results - It was found that at every age level knowing words and passages were causally related to pseudoword acquisition. The children were found to know between 20 and 40 sight words before acquiring simple pseudowords. There were significant differences in the growth trajectories of children varying in PA and RAN, but the low-RAN and low-PA groups did not differ. Conclusions - It was concluded that children learning to read in English develop grain size in a small to large fashion as in the more regular European languages, and that PA and RAN influence grain size development.
Meaghan Crosby ();Lesly Wade-Woolley - Word stress, reading skill and the missing letter effect
Purpose. Researchers have long been aware of the Missing Letter Effect (MLE), in which readers, when told to strike out a target letter from words in text, are more likely to fail to detect those letters in function or highly frequent words. Certain phonological factors have also been observed to influence the MLE, such as atypical pronunciation of the letter ("f" pronounced as "v" in OF) and silent letters ("e" in TYPE), and occurrence in unstressed syllables ("e" in conCERN vs PATtern). With few exceptions, however, work on the MLE has been conducted with adult, skilled readers. In children, the MLE appears to be more pronounced as reading skill increases. No published studies have yet examined the impact of word stress on the MLE in young readers. Method. Participants were 80 students in grade three and eight. Participants read a short text passage, in which appeared 12 target content words matched for frequency. A target vowel letter appeared in the stressed or unstressed syllable in each word. Participants were instructed to read the passage for comprehension and answer the set of questions at the end of the passage. They were also instructed to cross out each instance of the target letter as they read. Additional standardized tests of reading accuracy and fluency were also administered. Results. We expect a greater MLE for older children, and a greater MLE in unstressed syllables. New to the research, we expect to demonstrate an interaction, where younger, less-skilled readers show less difference as a result of word stress than do older, more-skilled readers Conclusions. Such results will be interpreted as evidence that increasing reading fluency entails the laying down of a prosodic track during silent reading, and that this is an important aspect of skilled reading.
Anna Cunningham (Aston University)Julia Carroll - Phonological learning as a function of age and exposure to reading instruction
Purpose: The paper examines two questions: First, what are the relative contributions of age and reading instruction in the development of phonological awareness (PA)? Second, does a dynamic measure of PA provide better information than a static measure in beginning readers? Method: The oldest and youngest children in their second year at school (matched on length of reading instruction) were compared with the oldest children in the first year (approximately matched on age to the young second years) on dynamic and static tests of phoneme segmentation, word reading and letter knowledge. Children were re-tested six months later for word reading and spelling. Results: The younger second years significantly outperformed the older first years on all tasks. In contrast there were no significant differences between the younger and older second years, although differences were in favour of the older children. The schooling effect was on average four times larger than the age effect. The dynamic measure of PA was a better predictor of subsequent reading than the static measure. It also eliminated floor effects and improved responsiveness. Conclusions: Exposure to reading instruction (above age) was the principal influence on the development of literacy and phonological skills in 5-6 year olds. Consequently, concerns about the reading development of summer-born children during early schooling may be unfounded. In addition, the dynamic measure of PA proved to be an effective assessment tool with potential for more widespread use.
Márcia da Mota (Universidade Salgado de Oliveira); Luciene Corrêa Miranda - Spelling of derivational morphology in Brazilian Portuguese
Purpose: Much research, mostly in English, has shown the importance of morphological knowledge for both reading and spelling. However, orthographies that follow strict letter-sound correspondences, such as Portuguese, may not require further levels of linguistic analysis. Despite the presence of many morphologically complex words in Portuguese, learners might not rely on morphology in such a system. This possibility seems most likely for prefixed derived words that do not follow morphologically-based spelling rules. This was the focus of our study. Method: Second and third grade Brazilian children were asked to spell two types of derived words: some had prefixes that were a syllable and others had prefixes that were only part of the syllable. Each type was matched to morphologically simple words with similar sounds and letter patterns (e.g. infinito-inferno and inútil-início). We hypothesized that morphological processing, should it occur, would lead to greater spelling accuracy for derived than for morphologically simple words. Result: No differences in spelling accuracy emerged between the derived words and morphologically simple words, in either condition. Conclusions: These results suggest that in transparent orthographies young writers do not rely on morphological processing in spelling words for which morphologically-based spelling rules are not explicit. Overall these results indicate that the relationship between spelling and morphology is a complex one that needs to consider features of the orthography and of the learners.
Daniel Daigle (Universite de Montreal)Rachel Berthiaume; Elisabeth Demont - Graphophonemic processing in deaf readers and in dyslexic readers of French
Considering the role of phonological processing in reading development and the phonological deficit characterizing both dyslexic readers and deaf readers (Colin et al., 2007; Demont & Botzung, 2003; Musselman, 2000; Ramus, 2003), the purpose of this study was to compare both populations for phonological processing. Dyslexic readers (n=16) and deaf readers (n=16) matched on reading scores and on age were compared to 21 normal reading development students of the same reading level (CL) and to 24 students of the same age (CA). We used two graphophonemic tasks evaluating grapheme-phoneme correspondences, but varying in terms of cognitive constraints (CC- and CC+). In the CC- task, subjects have to determine which of two pseudo-words (e.g., assidème or annidème) resembles the most a target item (e.g., acidème). In the CC+ task, subjects have to choose, out of four pseudo-words, the item that starts with a different sound (e.g., cékane cavène codile cutare). First, for both tasks and for all groups, results are above the threshold of chance. Secondly, results from ANOVAs indicate, that all groups had better scores on the CC- task than on the CC+ task and, that, whatever the task, deaf subjects, dyslexic subjects and CL have comparable results, but perform more poorly than CA. This study shows the importance of considering the cognitive constraints attached to the phonological task when investigating phonological processing in reading, especially with special populations. It also indicates that dyslexic readers and deaf readers are comparable and can process phonological information when reading.
Catherine Darrow (University of North Carolina at Chapel Hill); Jill Freiberg; Sarah S. DeLisle; David K. Dickinson - The connection between book type and vocabulary instruction in preschool classrooms
Purpose The objective of this study is to examine how book type and characteristics of preschool teachers' storybook reading is interconnected and related to children's gains in language and literacy. This study aims to answer the following questions: 1. Does book type (i.e. genre and format) and length of reading relate to the amount of vocabulary taught by teachers? 2. Do the amount of words taught and the length of reading relate to children's gains in vocabulary? 3. Does the curriculum intervention affect the way treatment teachers select and read storybooks? Method Research assistants coded video sessions of teachers reading storybooks to preschool children in 47 Head Start classrooms involved in a curriculum intervention. Results Preliminary analyses show that teachers defined more words when reading traditional books than when reading large format books (t(43) = -3.82, p < .01). Teachers in the treatment condition selected traditional books more frequently (t(45) = -5.49, p < .01) and defined more words when reading (t(46) = -6.53, p < .01), but did not differ in read lengths than the control teachers. The number of defined words and length of reading, however, were not correlated with gains in vocabulary in children. Conclusions These results suggest that the types of books read by teachers are connected to the ways and frequency in which children are taught the meanings of words. When teachers read traditional stories, as opposed to predictable big books, they exposed children to more vocabulary. Book selection, therefore, is a critical consideration.
Neeltje J. Davidse (Leiden University, The Netherlands)Maria T. de Jong - Are effects of book exposure on vocabulary and letter knowledge influenced by learning behavior?
Purpose: Some young children show immature learning behavior and due to that they may benefit less from book sharing. Book sharing is often considered one of the most important activities parents can do to promote young children's early literacy skills. If a child fails to concentrate on stories he or she may acquire less words and print knowledge than a child that is attentive during book sharing. Method: Participants were 228 native Dutch kindergartners, mean age 54.29 months (SD = 2.12 months). Parents filled in a HLE questionnaire and children completed tests that assess book exposure, vocabulary, letter knowledge and executive functioning. Results: Vocabulary and letter knowledge were predicted by book exposure (10.24% and 8.41% variance explained). The more children are read to, the more letters they know and the more their vocabulary expanded. Executive functions affected results of book exposure (2.56% and 3.61% respectively). Effects of book exposure on literacy outcomes were alike whatever the level of executive skills. Conclusion: Children with normal as well as children with weak executive skills benefit from book reading probably because book sharing is a highly structured activity and all children are rather attentive and focused during book sharing. EF's have a direct impact on vocabulary and letter knowledge (2.56% on letter knowledge and 3.61% on vocabulary) which suggests that learning behavior is important in addition to book sharing.
Marcia Davidson ();Marcia Davidson; Medina Korda; Ollie Collins White - Effective reading interventions in developing countries: Year one results from an RCT study in Liberia
Purpose: 1) to evaluate whether a reading intervention in grades 2-3 in randomly selected clusters of schools in Liberia results in significant improvement in reading skills for children in grades 2 and 3. 2) To examine the role of assessment feedback only by providing a "light intervention" treatment group in which students are assessed (no intervention) by teachers trained to administer EGRA and families are provided report cards during the year on reading progress. Method: A cluster method of sampling was employed and 180 schools participated. A total of 2957 students were randomly sampled from grades 2 and 3 classrooms . Assessments included a locally adapted measure based upon the validated assessment tool Early Grade Reading Assessment (EGRA). The reading intervention, taught as the primary reading curriculum, was developed by one of the authors and designed to require minimal resources to teach. Year One Results: There were significant increases in student skills in the full treatment group compared to the control group in grades 2-3 at the .001 level after 3.5 months of intervention on letter-naming fluency, phonemic awareness, nonword fluency, passage reading/comprehension, and listening comprehension. Results also indicated that students in the "light" intervention group also made significant gains when compared to the control group, but gains were not as strong as the full intervention group. Conclusions: The reading intervention model developed to improve reading skills for all children in grades 2-3 in Liberia is working after only 3.5 months of intervention. The intervention and assessment continues for one more academic year.
Rob Davies (University of Oxford Brookes, UK) - The development of word attribute effects in word naming: A multi-level model
Purpose The average reader's word naming performance is affected by word-level attributes like frequency and AoA. However, reading performance also varies in relation to person-level attributes like developmental typicality or relative reading experience (print exposure). We predicted interactions between word- and person-level factors so that AoA and frequency effects should be smaller for more experienced readers. Reading system performance will tend towards maximal efficiency with increased print exposure, as network connections are adapted to experience. With linear mixed-effects modelling, we are able to estimate such cross-level interactions more accurately than has previously been possible. Method We examined word-, person- and cross-level effects on reading in a sample of 19 typically developing and 20 dyslexic adults, plus 62 typically developing children (29 aged 10 years and 33 aged 12 years, on average). We observed word naming performance for 160 words (repeated measures) varying in spelling-sound regularity, frequency, AoA, length and other factors. We tested readers' print exposure, orthographic processing skill, phonological awareness and general oral reading ability using a battery of standardized tests. Results A linear mixed-effects model showed that reading latencies were predicted by AoA, frequency and length, among other factors, and that the size of these item-level effects interacted with person-level reading experience and skill. Conclusions Reading performance is driven by influences operating at multiple levels. A fuller understanding of reading development through the life span is enriched by a multilevel model of word-level influences and how such effects vary in relation to person-level attributes like reading experience.
Bronwen Davis (University of Guelph)Mary Ann Evans - ABC book reading and components of emergent literacy skills as predictors of first grade reading
Purpose - Emergent literacy conceptualizes reading as a developmental process beginning early in a child's life. As such, several foundational skills have been identified as important precursors to conventional forms of reading. We previously found that emergent readers draw on early literacy skills to 'read' an alphabet book displaying pictured scenes with accompanying text "letter is for object". Based on this and work highlighting early reading as a strong predictor of later reading, we aimed to explore the relation of alphabet book reading, as an integration and application of emergent literacy skills, with subsequent reading development. Method - Vocabulary, phonological awareness, and letter knowledge were assessed in 52 five-year-old children who had not yet achieved conventional word reading. Parent-child dyads were audiotaped reading an alphabet book. Sessions were coded for child labeling attempts. Reading achievement was assessed one year later. Results - Number of correctly labeled alphabet book pages significantly predicted word reading and reading comprehension. Vocabulary, phonological awareness and letter name and sound knowledge simultaneously explained a significant amount of variance in both reading measures, over and above what was predicted by the integrated and applied alphabet book task. Phonological awareness was the only significant unique predictor of reading comprehension, while both it and letter name knowledge were significant unique predictors of word reading. Conclusions - Findings suggest that children's integration and application of emergent literacy skills relates to subsequent reading development, though phonological awareness and letter knowledge remain significant unique predictors. Findings will be considered in line with theories of reading development.
Peter F. de Jong (University of Amsterdam)Eva Marinus - Variability of the dyslexic reading system
Purpose: Several theories of normal reading development have proposed a gradual refinement towards high quality orthographic representations. In dyslexic's such representations are generally assumed to remain less detailed, and of low quality. Several features of low quality representations have been suggested (Perfetti, 2007), one of these being an instability in the retrieval of a pronunciation from a written word. Whereas, the major focus of studies on dyslexia has been on mean differences between normal and dyslexic readers, instability implies a larger word-to-word or intra-individual variability in the reading speed of dyslexics. The main hypothesis of the current studies was therefore that intra-individual variability in reading speed is larger in dyslexic than in normal reading children. Method: Two studies were conducted. In both studies word and pseudoword naming tasks were administered to 24 dyslexic readers individually matched to chronological-age and reading-age controls. Words and nonwords varied in length from three to five letters. The second study also included a measure of rapid naming to explore whether differences in variability are specific to reading. Results: The results of the first study revealed that intra-individual variability in dyslexic's reading speed was larger than the variability in both groups of normal readers. Interestingly, the largest differences were found on the shorter words. The data of the second study are available, but have not yet been analyzed. Conclusion: The results thus far suggest that variability in reading speed is an important characteristic of the dyslexic reading system.
Peter de Lissa (Macquarie Centre for Cognitive Science- Macquarie University); Genevieve McArthur; Anne Castles - Orthographic learning: insights from fixation-related potentials
Purpose- The aim of the study was to track the neural changes that occur as pronounceable non-words form orthographic representations. Method- An eye-tracker was used in combination with electroencaphalography (EEG) to form fixation-related potentials (FRPs). 14 participants read passages of text that contained pronounceable non-words in place of nouns. FRPs taken from the first reading of the non-words was compared to the fourth reading of those non-words, effectively comparing reading involving phonological processing with reading involving orthographic processing. Results- When read for the first time, non-words produced larger activation over central areas at approximately 400ms than when they were read for the fourth time. Fixation durations on the non-words decreased as exposures increased. Conclusions- The greater activation at the central sites elicited by the first reading of the non-words reflects a larger recruitment of motor-cortex processes for phonological processing and sub-vocalisation, which is required to a lesser degree as orthographic representations are stored and consolidated as words become familiar.
Adriana de Souza Batista Kida ()Adriana de Souza Batista Kida;Cinthya Kawano; Carolina Alves Ferreira de Carvalho; Clara Regina Brandão de Ávila - Reading Characteristics of Brazilian Students Considered Good and Struggling Readers
Purpose: Both national and international assessments indicate a wide number of Brazilian students with reading disability. Teachers and clinicians search performance indications that characterize the reading disability. This study assessed reading characteristics of typical and struggling students, from error types to fluency parameters, and investigated the potential correlations. Method: Sixty students from the 3rd to the 5th grades of public schools were evaluated: 30 (10 from each grade) with reading and writing complaints (Research Group) and 30 indicated by their teachers as good readers, paired by age and grade (Control Group). The oral reading of text, words (orthographically balanced for Brazilian Portuguese) and pseudowords was recorded to assess the rate and accuracy, as well as the error types (Goulandris, 2004; Goikoetxea, 2006) classified according to the word reading. Mann-Whitney test, Kruskal-Wallis test and Spearman's Correlation with a 0.05 significance level were used. Results: The worst performances were found in the Research Group, for all the studied fluency variables (p<0.001). The errors of non-compliance with context-independent matching rule (p<0.001*), omissions and additions (p<0.001*), non-compliance with accentuation (p=0.036*), complex errors (p<0.001*) and refusals (p=0.001*) were more frequently found in this group. Diverse negative correlation patterns between fluency and different error types also differentiated the groups. Conclusion: The RG presented lower fluency values. The frequency and the type of errors reveal the influence of the error type on the reading rate and accuracy differently for each group.
S. Hélène Deacon (Dalhousie University) - Word wonders: the effects of awareness of letter, sound, and meaning patterns on learning to read
Purpose: This study was designed to examine whether children's awareness of sound, letter and meaning patterns in words (or, more formally, their phonological awareness, morphological awareness, and orthographic processing) each make a unique contribution to reading development. Method: English-speaking children in grades 1 and 3 completed measures of these three constructs, as well as standardised measures of real and pseudoword reading and of vocabulary. Results: Results of linear regression analyses showed that each of the three metalinguistic variables made an independent contribution to both real and pseudoword reading. These contributions were independent of the other reading-related skills and vocabulary and age. Contributions of phonological awareness and orthographic processing to real and pseudoword reading were consistently larger than those of morphological awareness. This pattern was remarkably stable across the two grades, with the exception of a greater contribution of orthographic processing to pseudoword reading at grade 3 than at grade 1. Conclusions: Results suggest that we need to consider the independent contributions of each of these three variables to early reading. These findings will be discussed in light of current models of reading development.
Hélène Deacon (Dalhousie University); Diane LeBlanc; Chantal Sabourin - Double it up: How children integrate information about orthographic and morphological regularities in English spelling.
Purpose: Learning to spell in English can be challenging because it is determined by several levels of regularities. There are orthographic regularities, for example, that long and short vowels in English are followed by a consonant single and doublet, respectively (e.g., diner and dinner). There are also morphological regularities, such that the spelling of root words is often retained in related forms (such as dine in diner). The present experiment is designed to determine how children integrate these two sources of information. Method: We asked English-speaking children in grades 1 to 4 to spelling two- and one-morpheme words that contain consonant singlets (skater and miter) and doublets (knitter and patter). Orthographic and morphological regularities reinforce the same and correct spelling for the singlet pairs, and they provide conflicting information for the doublet pairs. Results: Results of suggest that 7- to 9-year-old children are sensitive to both orthographic and morphological dimensions of words, and that this dual sensitivity sometimes leads to correct spelling and sometimes to incorrect spelling. Results of Experiment 2 suggests that orthographic information dominates young (age 6) children's spelling, at least in a case when there is a strong orthographic regularity. Conclusions: Taken together, these two experiments suggest that children are sensitive to the multiple dimensions of regularity in the English orthography and that this sensitivity can lead to mistakes.
Sylvia Defior (Universidad de Granada (C.I.F. Q-1818002F) Cuesta del Hospicio s/n)Sylvia Defior, Nuria Calet, Luciana Nigro & Eduardo Onochie - Emergent prosodic skills and literacy development in Spanish preschool children
The importance of phonological awareness as a foundation for reading development is well documented. Recently, one aspect of phonology has received more attention, namely prosody; prosody is the phonological subsystem that encompasses the tempo, rhythm and stress of language. The purpose of this study is to investigate the relationship between preschooler's prosodic awareness skills, phonological awareness skills and literacy development. Participants were 180 5 years old-preschool children. Several tasks assessing prosodic awareness (compound nouns, sensitivity to words and pseudowords stress), non-speech rhythm, phonological awareness (phoneme isolation and phoneme blending), reading (one minute reading and picture word matching), and spelling were carried out. Overall, results emphasize the well-known importance of phonological awareness in literacy development It is found a significant indirect contribution of non-speech rhythm to literacy skills, mediated by phonological awareness. Generally children showed low prosodic linguistic awareness. Moreover, a weak relationship between prosodic linguistic awareness and literacy skills is found. The emergency of prosodic skills and the relation between rhythm, phonological awareness and the development of reading skills are discussed.
Carolyn A. Denton (University of Texas Health Science Center - Houston)Amy E. Barth; Tammy D. Tolar; Jack M. Fletcher; Sharon Vaughn; David J. Francis - The effects of an individualized tier 3 reading intervention provided to primary grade students with low reponse to previous tier 2 intervention
The purpose of this experimental study was to evaluate the effects of an intensive Tier 3 supplemental reading intervention provided to second grade students who demonstrated inadequate response to less intensive Tier 2 intervention in grade 1. Method. Participants were randomly assigned to receive the experimental intervention (n=46) or Typical School Practice (TSP; intervention normally provided in their schools; n=26). The sample was 48% female; 74% economically disadvantaged; 43% Latino, 42% African-American, and 15% Caucasian; and 31% had identified disabilities. The treatment group received daily multi-component small-group intervention for 45 minutes over about 25 weeks. Intervention was not scripted; teachers individualized instruction by selecting from a menu of well-described instructional activities to construct daily lesson plans based on diagnostic assessments. Students read leveled text that was not phonetically decodable. Fidelity averaged 93%. Assessments measured decoding, reading fluency, mathematics calculations, math story problems, and reading comprehension. Results. At posttest, treatment students outperformed TSP students on all reading measures and math story problems (but not computation). Preliminary results based on ANCOVAs covarying for pretest scores indicated differences were statistically significant for decoding and approached significance for fluency, comprehension, and math story problems (p < .10). Effect sizes were small to moderate. We will reanalyze the data accounting for nesting but anticipate similar outcomes. Conclusions. The study provides evidence of effectiveness in improving reading and mathematics problem solving outcomes for students with reading difficulties that are resistant to remediation, through individualized reading intervention with a non-scripted program and primarily non-decodable text.
Joseph T. Devlin (Cognitive, Perceptual, & Brain Sciences and Institute for Cognitive Neuroscience, UCL, UK) - Investigating ventral occipito-temporal contributions to reading
Purpose. Despite considerable interest in the role of the ventral occipito-temporal cortex (vOTC) during reading, there is little consensus regarding the information processing properties of the region. Instead a debate primarily focuses on how this region relates to cognitive models of reading. Here we aimed to characterize several basic aspects of information processing in vOTC. Methods. Dynamic causal modelling was used with fMRI data to investigate the flow of information between vOTC and other left hemisphere areas engaged during reading. A second fMRI experiment then explicitly tested the prediction of the DCM that top-down, non-visual properties of words affect vOTC activation during visual word recognition. Finally a series of transcranial magnetic stimulation (TMS) experiments examined how temporarily disrupting vOTC processing affected performance on both reading and object recognition tasks. Results. The DCM analyses demonstrate that interactive models better explain the fMRI data than simpler feed-forward models. This then predicts that top-down, non-visual aspects of a stimulus will affect vOTC response profiles, and this was confirmed in a second experiment that systematically varied the visual and non-visual demands on visual word recognition. The TMS experiments demonstrated that i) vOTC stimulation disrupted processing of low, but not high, frequency words or pseudowords; ii) this disruption occurred as early as 120msec post-stimulus onset, and iii) affected both words and pictures. Conclusions. These findings provide a basic response profile for information processing in vOTC that places important constraints on understanding the functional role of this region in higher order visual processing including reading.
Peter Dewitz (Educational Consultant);Peter Dewitz; Lili Claman - Knowledge Development and Comprehension in Core Reading Programs
Purpose The linking of text information to prior knowledge is a given of both schema theory (Anderson & Pearson, 1984) and the construction integration model (Kintsch, 1998). We sought to study how core-reading programs develop students' knowledge with and across texts. 1) Do the teacher's manuals in core reading programs direct teachers to develop the knowledge that students need to comprehend the selections? 2) Does the knowledge that students acquire from reading one text help them comprehend the next text with knowledge building across selections? Method We conducted a propositional analysis for each passage in the third and fifth grade levels three core reading programs. Next, we gave the complete text and the listed propositions to classroom teachers and asked them to rate each proposition as being a familiar, a partially familiar or an unfamiliar concept to their students. Then we determined what proportion of the concepts the students needed to know (important but unfamiliar ideas) were taught in the teacher's manual. In the second analysis we examined if important proposition in one passage were repeated and necessary to understand the next passage and the next, looking for intertextual connections. Results The teacher's manuals developed less than 1/3 of the knowledge that classroom teachers judged to be important for their students. Core programs do a better job of developing knowledge when the selection is non-fiction than when it is fiction. Two of the three core programs we studied did little to build knowledge across selections within a theme. What is learned in the first selection is rarely germane to the next. Conclusions Core reading programs do little to develop students' knowledge in a systematic manner a concern of Walsh (2003; Hirsch, 2003; Newman, 2003). These findings also suggest that core programs' failure to build knowledge from one passage to another may contribute to students' overall comprehension problems (Willingham, 2006), just as core programs do an inadequate job of developing comprehension strategies (Dewitz, Jones & Leahy, 2009).
Eric Dion (Universite du Quebec a Montreal)Eric Dion; Léonie Lemire-Théberge, Marie-Hélène Guay; Isabelle Dubé; Monique Brodeur; Douglas Fuchs - Improving at-risk second-graders' reading comprehension through vocabulary or strategy instruction: a randomized field study
Purpose: There are few evidence-based interventions that help underprivileged beginning readers improve their reading comprehension. In this study, we compared the effectiveness of vocabulary or strategy instruction. Method: Second-grade classrooms (N = 33) were randomly assigned to one of three conditions: control (regular instruction), 2) vocabulary instruction or 3) strategy instruction. In each class, nine students were assessed: five low-achievers, two average-achievers, and two high-achievers (N = 297). In the two intervention conditions, teachers conducted peer-mediated activities for three 30 min sessions per week from October 2008 to April 2009, and the fidelity of implementation of these activities was closely monitored. In the vocabulary instruction condition, the meaning of five new words were presented at the beginning of each session. Students then took turn associating words with their definition and reading a text containing these words. In the strategy condition, students first took turn reading the text. They then reread each paragraph and identified its main idea by asking themselves "Who the paragraph is about?" and "What is happening?" Results: Large effect sizes were observed on post-test measures of proximal learning. In classrooms assigned to the vocabulary instruction, students had a much better knowledge of instructed vocabulary (e.s. = 0.96), while their peers in classrooms assigned to the strategy condition identified paragraphs main ideas more easily (e.s. = 1.00). Effects on reading comprehension measures were in the moderate range. Conclusions: Both versions of the low-cost, teacher friendly peer-mediated activities helped participating students move beyond word recognition and become better readers.
Christian Doabler (Pacific Institutes for Research); Hank Fien; Scott Baker; Benjamin S. Clarke - Measuring Explicit Instructional Interactions Through a Direct Observation Instrument
2010 SSSR Poster Proposal Purpose: Direct observation is an effective method for measuring the quality of classroom instruction. We developed a direct observation instrument to better understand the potential mediating variables hypothesized to influence student achievement. The purpose of this paper is to discuss the development of the instrument, its conceptual foundation, and the instructional interactions it purports to capture. Methods: We developed the Coding of Academic Teacher-Student interactions (CATS) direct observation instrument within the context of two, federally funded randomized control efficacy trials. These studies involve kindergarten and first grade classrooms. We've conducted over 150 observations during reading and mathematics instruction. The CATS instrument systematically measures a set of instructional behaviors, including teacher demonstrations, academic feedback, and student practice opportunities. A growing body of scientific evidence suggests these behaviors are an integral part of effective instructional practice. Consequently, we are able to consider the quantity and quality of instructional interactions that occur between teachers and students. Results: For the presentation, analyses will focus on the instrument's reliability and capacity to predict important student outcomes. We also compare data collected across the two instructional domains. We believe this juxtaposition will demonstrate the instrument's generalizability. Conclusion: The recent advent of Response to Intervention calls for the delivery of high quality instruction to occur in the general education classroom. As such, there is need to establish reliable and valid tools for determining whether students are receiving effective instruction across content areas. The CATS tool could potentially serve as such a mechanism.
Michelle Drouin (Indiana Purdue University Fort Wayne)Sherri Horner; Toni Sondergeld - Letter names and sounds in preschool: A Rasch model analysis
Purpose: Within the U.S., there is no unified method of teaching preschool children letter names and sounds. Instead, methods of letter name and sound learning usually vary by school and teacher. However, empirical data exists that could inform instruction (e.g., McBride-Chang, 1999), specifically as it relates to the sequence of presentation of letters. In this study, we expand upon the findings of McBride-Chang to explore the relative difficulty of upper- and lowercase letter names and sounds using Rasch modeling. Method: We administered upper- and lowercase name, sounds, and recognition tasks to 500+ children aged 3-5 in four different cities (Brooklyn and Queens, NY, Fort Wayne, IN, and Memphis, TN). All sites had some letter instruction, but letter instruction methods varied, and no teacher used a "letter of the week" system. We analyzed data with a Rasch model program, which gives separate information about the relative difficulty of items and the relative performance (ability) of individuals. Results: Preliminary analyses with a subsample show interesting trends, some expected (e.g., an item map showed uppercase A and O were among the easiest letter names while lowercase d, u, and q were among the most difficult) and some unexpected (e.g., a lower ability student correctly named both "b" and "B"). Results will include item and person maps separated by age groups. Conclusions: Information about the relative difficulty of letter names and sounds could be used to inform preschool and kindergarten letter instruction. This work extends upon McBride-Chang's findings with regard to the sequence of letter learning.
Lynne Duncan (University of Dundee)Elaine Gray - Do young readers show morphological sensitivity in visual word recognition? New evidence from masked priming
Purpose Adult masked priming techniques adapted from Rastle, Davis and New (2004) were used to explore whether evidence exists of a morphological organisation in the developing lexicon. Method Participants were English-speaking normal readers in Grades 4 (n=25) and 6 (n=15). To explore links with morphological awareness (MA), the production of real and novel derivations was assessed via sentence completion. A visual lexical decision task examined three (long duration) masked priming conditions: (1) morphologically transparent primes (e.g. worker-WORK); (2) morphologically opaque primes (e.g. corner-CORN); and (3) orthographic (form) primes (e.g. turnip-TURN). Suffix type frequency was manipulated. An identical task was administered to the older children using new stimuli and brief duration primes. Results With long duration primes, reliable priming was found for Grades 4 (accuracy) and 6 (RT) with high frequency suffixes in the morphologically transparent condition. Older children also showed priming with low frequency suffixes in this condition (accuracy, RT). Comparison across conditions revealed differences for Grade 6 only: facilitation was greater with transparent primes than with orthographic primes, and marginally greater than with opaque primes. MA correlated with Grade 4 reading and with vocabulary in both grades but nothing correlated with the Grade 6 priming effect. No priming was observed with brief duration primes. Conclusions The outcome suggests an emerging sensitivity to morphological complexity in the developing lexicon. This effect may be primarily semantically mediated in view of the absence of a relation with MA in either grade and the lack of evidence of pre-lexical morpho-orthographic decomposition in Grade 6.
Carsten Elbro (University of Copenhagen)Hanne T. Daugaard; Anna S. Gellert - Dyslexia in a second language? A dynamic test of reading acquisition may provide the answer
Purpose. Specific reading difficulties may be hard to diagnose in a second language. A poor performance on a test of reading may be caused by poor language proficiency. The study was concerned with the effects of second language proficiency on three types of reading tests. The aim was to evaluate a language independent test of individual potential to learn to read with an alphabetic writing system. Method. A dynamic test of literacy acquisition was developed. It involved learning of three novel letter shapes and their sounds and subsequent reading of non-words using the three letters. Instructions were non-verbal. The dynamic test was administered to 160 adults together with standard tests of non-word reading, word reading, and several language measures. The adults were selected from four groups: Diagnosed dyslexic and non-dyslexic native speakers of Danish, and non-native speakers of Danish some of whom were suspected by their teachers to be dyslexic while others were not. Results. A significant interaction between reading test and group indicated that the experimental, dynamic test was the relatively easiest for the second language learners. Several of the second language learners with suspected dyslexia appeared to be mis-classified by their teachers. Conclusions. Current measures of word and non-word reading may be confounded with language proficiency and hence under-estimate the potential for reading acquisition in some second language learners. A dynamic, language independent measure of acquisition of the basic alphabetic principle may provide a more valid indicator of specific reading difficulties in second language learners.
Stephen Elliott (Vanderbilt University); Peter Beddow - Designing accessible achievement tests for students with reading difficulties
Purpose: Students with reading difficulties experience challenges demonstrating what they have learned in math and science because their reading deficiencies interfere with the processing and responding to test items. Method: In this presentation, we examine procedures for developing, modifying, and evaluating math and science test items using an evolving modification paradigm, as well as a classic test reliability and validity framework. Theoretical influences, such as principles of universal design, cognitive load theory, and item development research, are discussed. The Test Accessibility and Modification Inventory (TAMI), a tool that provides systematic and comprehensive guidance to help educators modify grade-level tests, will also be introduced. Results: Cognitive lab methods and experimental field tests are then described, along with examples and key findings from each, relevant to the development of highly accessible achievement tests for students with reading difficulties. Conclusions: The presentation concludes with a discussion of precautions, lessons learned, and questions generated about the methods used to improve both achievement test access and score validity for all students.
Asa Elwer ()Stefan Samuelsson - Developmental trajectories distinguishing between dyslexics and poor comprehenders
Purpose The aim of the study was to compare dyslexics and poor comprehenders with children who follow normal growth in reading from 5 years of age to grade 4. Method A total of 600 American and Swedish twins engaged in the international longitudinal twin study (ILTS) took part in the study. Based on the simple view of reading, three groups with different reading profiles based on test of listening comprehension and word decoding in grade 4 were identified; normal readers, poor comprehenders and dyslexics. After group identification, test results from an extensive battery of cognitive and language tests from 4 previous test occasions were analyzed. Results The results indicate severe language related problems in the poor comprehenders at age 5. They exhibit poor vocabulary, morphological and grammatical skills. More surprisingly they exhibit very weak phonological awareness and lexical access. Vocabulary and grammatical skills continue to predict group membership when comparing poor comprehenders to dyslexics with logistic regression analysis through development. Since poor comprehenders follow the normal path of development in word decoding, poor word decoding discriminates the readers as dyslexics as does the alphanumeric RAN. Conclusions As poor comprehenders are often identified based on their reading profile we know very little of their cognitive and language skills prior to their manifest problem in reading. This study questions the view that poor comprehenders only show deficits in higher order processes necessary for complex reading. Poor comprehenders exhibit deficits often linked to dyslexic reading problems in addition to basic language problems.
Carmen L. Escribano (Universidad Complutense de Madrid) - Precursors of word reading from kindergarten to first grade in two groups of monolingual Spanish-speakers
Purpose: To assess the relative contribution of phonological awareness (PA), rapid naming (RN), and vocabulary to word reading accuracy in the developmental period from kindergarten to first grade in two groups of monolingual Spanish-speakers. Method: Fifty-four Spaniards and twenty-eight Latin American immigrant children were first tested in the winter of kindergarten and a second time in the spring of first grade. Measures such as receptive Vocabulary (PPVT Spanish version), RAN-Letters, RAN-Objects (Wolf & Denckla, 2005), and a measure of PA were administered in the winter term of kindergarten. Latin American could not recognize letters at this point and RAN-Letters was not administered to them. In the spring term of first grade children were tested again on the WMLSR-I, Word Reading Accuracy Subtest (Woodcock-Muñoz Language Survey-Revised, 2005). The relative contribution of PA, RAN-Letters, RAN-Objects, and vocabulary to word reading accuracy was evaluated by means of a series of stepwise multiple regression analyses. Results: The results showed that RAN-Letters was the best predictor of word reading accuracy for Spaniards, and PA for Latin Americans, no other predictors turned to be significant. Conclusions: PA was a significant independent predictor of prospective word reading ability for Latin Americans. Nonetheless, relative to PA, RAN-Letters seemed to play a more decisive role in word reading accuracy for Spaniards. Similar to the results of other studies (Bryant et al., 1990; Silvén et al., 2007) the relationship between PA and later reading accuracy disappeared when kindergarten reading ability or the amount of letter knowledge was taken into account.
Mary Ann Evans (Department of Psychology, University of Guelph, Guelph, Ontario, Canada); Jean Saint-Aubin - Instances of attention to print during independent reading of Alphabet books by pre-readers
Purpose. This presentation will examine children's behavior when reading alphabet books. It will compare attention to print across three studies using different books to determine how graphic design and children's alphabetic knowledge influences attention to print. Method. In each study prereaders completed a letter naming test and "read" an alphabet book while their eye movements were monitored. Each page of the Study 1 (n= 20) book had one picture, one word, one featured letter in the corner, and a bear. The Study II (n= 46) book had complex pictures, the featured letter in the corner and either one or three words per page depending on book condition. The Study III book (study in progress) was a modification of the book used in Study 1 with larger and centered featured letters. Results. Results revealed that with only one word per page across studies children spent between 1 and 1.3 second on it. In contrast, when a sentence was presented, fixation duration on the featured letter dropped to .7 second. Finally, children with greater letter knowledge were more likely to look at a featured letter and the first letter of the picture label, even if they did not know that letter. Conclusions. Logan (1988) proposed that every instance of attention lays a trace in memory for the item of that attention. The findings suggest that alphabet books elicit attention to print. Moreover, some letter knowledge encourages attention to print, suggesting that even at this early stage literacy development is partly self-propelling.
Fataneh Farnia (University of Toronto, Canada); Esther Geva - Reading comprehension in ELLs and monolinguals- Patterns of growth and challenges
Purpose: We examined the growth trajectories of reading comprehension (RC) from Grade 4 to 6 in 400 English language learners (ELLs) and 153 English monolingual (EL1) learners, and extent of stability over time in sub-groups. Method: Groups of poor vocabulary (n= 62), poor decoding (n= 64), and normal achievers (n = 427) were identified. The poor vocabulary and poor decoding groups consisted of children who scored one SD below their reference group means (i.e., ELL or EL1) from Grade 1 through Grade 6. Results: Growth curve analysis indicated that RC growth in normally achieving EL1 children was linear. RC trajectories of normally achieving ELLs, and poor vocabulary ELLs and EL1s slowed down over time (quadratic effect). In ELL and EL1 alike, Grade 1 vocabulary contributed to the rate of linear change in RC. Students with Poor vocabulary had better scores on Grade 6 RC than poor decoders. In ELL and EL1 poor word reading in Grade 1 was associated with slower growth rate in RC and subsequent lower score on Grade 6 RC. Normally achieving ELLs and poor decoders scored significantly lower on Grade 6 RC than EL1 normal achievers or poor decoder. There was a persistent gap between ELLs and EL1s' RC in all subgroups. Conclusions: Overall ELLs with limited vocabulary lagged behind their monolingual peers; ELLs with poor decoding skills underperformed their monolingual peers on RC even after 6 years of instruction in English, and their RC trajectories slowed down significantly indicating a deficit rather than a lag in RC.
Amber Farrington (Florida State University); Christopher J. Lonigan; JoAnn M. Farver (Department of Psychology, University of Southern California) - Development of phonological awareness skills in preschoolers who are English language learners: Evidence of bi-directional transfer
Purpose: Previous findings demonstrate that first and second language phonological awareness (PA) skills correlate over time and across languages (e.g., Gottardo & Mueller, 2009). Recently, Anthony et al. (2009) examined PA in Spanish and English using a prekindergarten sample and found that earlier skill levels in each language accounted for considerable variance in the other language at a later time. However, it remains unknown whether earlier Spanish PA skills would also influence later English PA skills beyond the influence of earlier English PA skills. In other words, are the connections across languages bi-directional. Method: Regression analyses were performed using a sample of 380 preschoolers who were English language learners (ELL). PA skills were assessed using parallel measures of blending and elision in both English and Spanish that were administered to the children at the beginning and end of the preschool year. Results: Analyses revealed that spring scores on both English and Spanish blending measures were uniquely predicted by fall scores on both English and Spanish blending measures. Similarly, spring scores on both English and Spanish elision measures were uniquely predicted by fall scores elision scores in both languages. Conclusion: These results provide some support for the notion that the early literacy skills of preschoolers who are ELL transfer across languages in a bi-direction fashion. Whereas previous studies have demonstrated concurrent correlations of PA skills across languages, this study is unique in demonstrating bi-directional, longitudinal relations across languages. These findings have implications regarding early academic interventions and optimal learning and testing conditions.
Argyro Fella (Department of Psychology University of Cyprus);Timothy C. Papadopoulos;George K. Georgiou - RAN components and reading development in Greek: what underlies their relationship?
Purpose: To examine: (a) how RAN components - articulation time and pause time - relate to their total times and to each other, (b) how they relate to reading fluency (word, nonword, and text-reading fluency), and (c) what processing skills mediate the relationship between the RAN components and reading. Method: 68 Greek-speaking children were followed from grade 2 until grade 6 and were assessed three times (grades 2, 4, and 6) on measures of rapid naming (digits and objects), speed of processing, phonological awareness, orthographic processing, articulation rate, and reading fluency. The responses of the children on the RAN tasks were recorded and analyzed with a sound editing software. Results: The results indicated, first, that RAN Digits total time was more strongly related to articulation time than to pause time. In contrast, RAN Objects total time was more strongly related to pause time than to articulation time. Second, the components correlated significantly with each other (more so in RAN Digits). Third, the RAN components correlated strongly with the reading fluency measures and the relationships were stable across time. Finally, the RAN Digits components continued to predict reading fluency even after controlling for all the other processing measures. Conclusion: The findings suggest that the relationship between RAN Digits and reading is driven by articulation time more so than pause time. In contrast, pause time is the critical component in RAN Objects. None of the current theoretical accounts adequately explains why RAN is related to reading.
Ruth Fielding-Barnsley (University of Tasmaia) - Resistance to literacy intervention
Purpose This study examines profiles of 20 children who did not make significant gains in a longitudinal literacy intervention program. Method. A total of 752, 5 year old children were screened on measures of phonological awareness and expressive language and of these the lowest 20% were included in an intervention program. The two approaches to intervention included a language based program and a phonological program. The language program was based on Marian Blank's four levels of language and the phonological program was based on Hatcher and Hulme's Sound Linkage program. Each child received a total of 16 hours intervention in small groups including 8 hours of phonological awareness and 8 hours of language enrichment, in addition the 20 resisters received 4 hours of alphabet instruction and 4 hours of building cvc words. Results. The results demonstrate that even though these 20 children encountered significant difficulty in attaining similar results to the other 438 children in the study they did make significant gains after an additional 8 hours of targeted intervention. Conclusion Response to intervention is the current method used to identify children at risk of developing learning difficulties therefore it is important to understand that even though these children may be slower to respond to intervention they are still able to make considerable gains given additional targeted intervention.
Jessica Folsom (); Stephanie Al Otaiba; Luana Greulich; Jane Meadows - Kindergarten resilience factors of reading difficulties
Purpose: The purpose of this presentation is to examine the predictive role of early reading skills, home literacy environment, and classroom instruction on kindergarten reading growth and achievement. In addition, we report factors associated with resilience among children who enter kindergarten with weak initial skills. Method: In this presentation we analyze data from two cohorts of kindergarten students (n = 778) from a larger NICHD funded study on the school based prevention and identification of learning disabilities. Data include student assessments, observation of classroom instruction, and parent and teacher surveys. Two types of analysis are presented. HLM is used investigate the individual reading growth and outcomes using home literacy, classroom instruction, and child characteristics as predictors. A series of ANOVAs are conducted on a subset of students (n = 192) at high risk for developing reading difficulties to identify if home literacy, classroom instruction, and child characteristics are different between students that were resilient (n = 96) and those students that remain at risk (n = 96). Results: Preliminary results from the HLM and ANOVAs suggest that IQ, vocabulary, print knowledge, PA, amount of time spent reading, severity of risk, behavior, and a variety of instructional practices can act as resilience factors. Conclusions: The results confirm the importance of home literacy environment, high quality classroom instruction, and initial child characteristics in the growth of reading skills in kindergarten and in the prevention of reading difficulties.
Jessica Folsom (); Luana Greulich; Stephanie Al Otaiba; Jane Meadows - A SEM on reading development from kindergarten to second Grade: The role of vocabulary
Purpose: The primary purpose of this poster is to share the results of a study investigating the contributions of vocabulary, print knowledge, phonological awareness, and word reading on reading achievement trajectories from kindergarten through second grade Method: This study is a longitudinal study from a larger NICHD funded study on the school based prevention and identification of learning disabilities. Participants include 224 students in the 2006-2007 school year. A variety of student measures of print knowledge, vocabulary, phonological awareness, word reading, and comprehension were administered in kindergarten, first and second grade. We conducted a SEM to model the direct and indirect effects of phonological awareness, vocabulary, and word reading on reading achievement at the end of kindergarten, first, and second grade. Results: Preliminary results suggest a moderately fitting model where: print knowledge significantly predicts vocabulary and phonological awareness; phonological awareness moderates the effects of vocabulary on word reading; vocabulary has a significant direct effect on reading achievement in kindergarten and second grade, and a significant indirect effect on word reading in kindergarten, first, second grade. Conclusions: The results confirm that kindergarten predictors of reading achievement can be combined into latent constructs that significantly predict later reading achievement. Data suggest that it may be important to assess vocabulary as it significantly predicts reading achievement above and beyond phonological awareness, and word reading and may be an important risk factor.
Liliana Fonseca (Universidad Nacional de San Martín); Barbara Gottheil; Adriana Aldrey; Maria Pujals; Ines Lagomarsino; Sylvia Defior Citoler; Gracia Jimenez Fernandez; Francisca Serrano Chica; Graciela Rosa - LEE, Reading and writing processes assessment battery in Spanish. Discrimination between good and poor readers.
Reading is a complex skill which involves word recognition and reading comprehension. Both are necessary for the mastery of this skill. In this work we compare the results obtained for the general population and a group of children with reading and writing disabilities using the LEE test, an assessment standardized tool for children from 1º to 4 º grades. The test evaluates the processes involved in word recognition, reading comprehension and spelling. It comprises the subtests of word and pseudoword reading, sentence comprehension, prosody, text comprehension and word and pseudoword spelling. It also includes two complementary sub-tests concerning letter identification and phonemic segmentation, they allow the psychologist to make a more accurate diagnosis of the learning disabled children regarding the Spanish written code acquisition. Accuracy and time standards are offered and also for fluent and non fluent reading. The sample consists of 395 Argentinean children and 70 children with reading and writing disabilities from 1 to 4º Grade It's observed that the LEE battery is valid to compare the general competition in reading and writing being able to suitably detect the difficulties some children have. The following Subtests: Word Reading, Pseudowords Decoding, Text Reading Comprehension, Word Spelling and Pseudowords Spelling, present significant differences comparing the general population sample and the clinical sample. Fluency and reading speed assessed by LEE, are indispensable information to consider when reading in Spanish is involved. It has been proved that this information discriminates properly between normal and poor readers.
Barbara Foorman (FL State University, Florida Center for Reading Research) - Discussion of the contribution of reading subskills to reading comprehension across age and language
Component skills have been theorized and researched in multiple disciplines - cognitive, linguistic, neuroscience (e.g., Breznitz, 2005; Caplan, 2004; Harm, McCandliss, & Seidenberg, 2003; Perfetti & Bolger, 2004; Rayner, Chace, Slattery & Ashby, 2006; Sandak, Mencl, Frost & Pugh, 2004). They are the underlying foundational skills that enable efficient processing of the surface level and to some extent of the textbase. Increasingly, components frameworks are being incorporated into reading comprehension frameworks (e.g., Baer, Kutner, & Sabatini, 2009; PIAAC Literacy Expert Group, 2009) to better understand the underlying literacy skills of readers across the proficiency continuum. However, on their own, component skills do not sum up to reading proficiency. That is, one could hypothetically be proficient in each subskill and still fail to adequately integrate them to achieve proficiency. The studies presented in this symposium will be discussed with respect to how reading comprehension is concpetualized and how predictors relate to that conceptualization across ages and languages.
Alon Fragman (Ben Gurion University, Beit Berl College)Susie Russak - Grapheme phoneme correspondence knowledge in the initial stages of acquisition of Arabic (FL) among native Hebrew speaking students
The present study examined the spelling errors of 8th grade native Hebrew speaking children (N=34) learning Arabic (FL) after their first year of literacy acquisition. Two experimental tasks were created which targeted grapheme phoneme correspondence knowledge in Arabic: dictation and word choice. Results indicated a high number of both phonological and orthographic errors with a slightly higher number of phonological errors. A qualitative analysis of errors indicated that phonological errors included difficulties representing novel phonemes that exist in Arabic but not in Hebrew, and difficulties distinguishing between long and short vowel sounds. Difficulties were also found in the representation of diphthong sounds that exist in Arabic but not in Hebrew. These findings are in line with earlier findings regarding the analysis of spelling errors among native Arabic speaking students (Abu-Rabia & Taha, 2004, 2006). Further, these findings are in line with the linguistic constraint hypothesis (Saeigh-Haddad, 2007) and extend the relevance of the hypothesis to include non-native Arabic speaking populations learning Arabic as a foreign language. In addition, orthographic errors included difficulties distinguishing between letters with similar graphic representations and difficulties representing letters with multiple forms. The present findings have significant pedagogical implications regarding beginning literacy instruction for Arabic as a foreign language for Hebrew native speakers in Israel.
Silke Fricke (University of Sheffield); Maggie Snowling; Claudine Bowyer-Crane; Charles Hulme - Promoting oral language skills to build a secure foundation for reading
Purpose: The current project aims to establish whether identifying oral language weaknesses in preschool children, and providing language intervention before and during transition to school helps children at-risk of literacy difficulties respond to literacy instruction. Specifically, the project explores the role of vocabulary and narrative skills in developing literacy. Method: The project is a randomised controlled trial. 12 children from 15 nursery schools (N=180, mean age 4;2) were selected based on poor performance on standardised language measures and were randomly allocated to either an intervention or waiting control group. A peer comparison group was also recruited consisting of 6 children per school (N=90, mean age 4;2) matched on age and gender to the intervention and waiting control groups. The Language4Reading (L4R) intervention programme is a 30-week-programme delivered by trained Teaching Assistants, designed to begin in UK nursery and to continue during Reception. Children take part in group as well as individual sessions and the programme aims to develop three key areas: vocabulary knowledge, narrative and listening skills. Results: We will present findings after the first 20 weeks of intervention for all three groups: - Average language and literacy skills pre- and post nursery (10 weeks) and Reception part 1 (10 weeks), and their correlations - Hierarchical models exploring the relationship between language and literacy skills Conclusions: The paper highlights the role of oral language skills as a foundation for literacy. Implications for the practice of supporting children at-risk of literacy difficulties during the early school years are presented.
Naama Friedman (Tel Aviv University); Naama Friedman; Lilach Shvimer; Noa Kerbel; Einav Rahamim; Aviah Gvion - Treatment directions for developmental letter position dyslexia and attentional dyslexia
Purpose: Letter position dyslexia (LPD) is a reading disorder that results from a selective impairment in letter-position encoding. Attentional dyslexia results from a deficit in letter-to-word binding. The main error type in LPD is letter migrations within-words (reading tried as tired), and the main error type in attentional dyslexia is migrations of letters between words (reading coat-goal as coal-goat). Our aim was to explore different treatments for children and adolescents with these dyslexias. Method: Participants were 10 individuals with LPD and 9 individuals with attentional dyslexia. Each participant with LPD was asked to read 1904 words. Each participant with attentional dyslexia was asked to read 720 word pairs. The text manipulations for LPD included tracking the letters with the index finger, inserting spaces between letters, using different colours for different letters, and inserting a sign between letters. For attentional dyslexia, we manipulated the size of spaces between words, the size of font, the type of font, the use of finger tracking, the use of a "reading window", and we presented words in a table. Results: For LPD, finger tracking was the most efficient technique. For attentional dyslexia, using a "reading window" was most efficient. Other effective methods included finger tracking and increasing the spacing between words. Conclusions: The findings indicate that LPD and attentional dyslexia can be treated with different techniques, and so accurate diagnosis of the dyslexia type is crucial for the choice of treatment.
Angela Friend (University of Denver) - Parental education does not consistently moderate heritability of reading ability across the normal range: evidence from three independent samples
Purpose: Environmental moderation of the level of genetic influence on children's reading ability was explored in three longitudinal samples of twins collected from the US, Australia, and U.K. Method: Parental education was utilized in these studies as a proxy measure of environmental support for reading because it is correlated with a broad range of environmental factors related to reading development and academic achievement. The interactions were tested using the method outlined by Purcell (2002). Results: Parental education significantly moderated the heritability of word reading in kindergarten in the U.S. sample such that heritability increased as parental education increased. However, no significant interactions were found in the U.S. samples for reading, spelling, and comprehension in kindergarten or in grades 1 and 2 in the US, and Australian samples. Although a significant interaction with mothers' qualifications was found in U.K. data, the effect sizes were very small and not consistent with the results reported for IQ by Turkheimer et al., (2003) and for reading by Kremen et al., (2005), where they report higher heritability for higher educated families. Conclusion: Although there are many factors that could influence the null results in each of these samples such as differences in educational policies, environmental range, and sample size, taken together, these studies fail to find support for the environmental moderation of heritability on individual differences in reading development across the normal range.
Benjamin Gagl (University of Salzburg); Stefan Hawelka; Heinz Wimmer - Acquisition of efficient visual word processing: Evidence from eye movement and naming latencies
The present study examined the development of efficient visual word processing and measured both eye movements and vocal reaction time in reaction to singly presented words. Participants were children learning to read the rather transparent German writing system with about 20 participants from grade levels two, three and four. The youngest group, tested in the first months of Grade 2, had experienced about one year of reading instruction, because there is no reading preparation in kindergarten. The pool of words presented for reading aloud varied along several dimensions of interest. Word length varied from 3 to 6 letters and, among 4-6 letter words, number of syllables (one vs. two) and presence of consonant clusters was varied. Critical eye movement measures were number, duration and position of fixations. Results show a massive gain of efficiency from second to third Grade resulting from diminishing effects of word length, number of syllables and consonant cluster density. Interestingly, the increasing efficiency was mainly reflected in shorter fixation durations, whereas the number of fixations decreased only slightly from second to fourth Grade. Additionally, all three groups fixated first at word beginnings and successive fixations were distributed serially from the left to right. In sum, the present study suggests that early reading strategies rely predominantly on serial decoding which becomes efficient within the first three years of learning to read the transparent German orthography.
Kelly Geier (); John R. Kirby; S. Hélène Deacon - Morphological awareness, reading ability, and the reading of multi-morphemic words
Purpose There is growing evidence that children refer to the morphological structure of the words that they read. We examined the effects of morphological awareness and reading ability on the accuracy and speed of reading of multi-morphemic words. Method Tests of word reading ability (Woodcock Word Identification) and morphological awareness (Word Analogy and Sentence Completion) were given to 134 Grade 5 students. They also completed a test of their ability to read morphologically complex words, in which accuracy and response time were recorded. These words varied in surface frequency (the observed frequency of the words the students read), base frequency (the frequency of the bases of those words, which the students did not see as separate words), and transparency (the clarity of the relationship between base and derived form. Results Multi-morphemic word reading accuracy and speed were related positively to reading ability, morphological awareness, transparency, base and surface frequency. Significant interactions showed that (a) high base frequency increased reading accuracy and speed for words with low surface frequency, particularly for students with low reading ability or low morphological awareness skills, and (b) the combination of high base frequency and transparency increased reading accuracy and speed for words with low surface frequency. Findings were similar for accuracy and response time. Conclusion The effects of morphological awareness and of base frequency indicate that morphology is involved in the reading of multi-morphemic words. These findings suggest that children would likely benefit from instruction in reading strategies that target the morphological characteristics of words.
Anna Gellert (University of Copenhagen); Carsten Elbro - Dynamic measures of acquisition of reading fluency and vocabulary. A study of their validity
Purpose: Very little is known about the predictors of reading fluency and the course of development of reading fluency. Hence, the aim of the study was to investigate the concurrent validity of dynamic measures of reading fluency and vocabulary. The dynamic measures were constructed to measure children's potential for developing reading fluency and vocabulary. Method: Three dynamic measures of reading fluency with repeated readings and two dynamic measures of learning of new vocabulary were developed. Static measures of reading fluency, vocabulary and reading comprehension were included. Information about teachers' perceptions of the children's skills and their potentials are collected in questionnaires. The concurrent validity is assessed in three ways: 1) by means of comparisons of the various dynamic measures of reading fluency and vocabulary, respectively, 2) by means of comparisons with teachers' perceptions of the children's potentials, and 3) by means of comparisons with static measures of reading and vocabulary. Data are being collected from 100 Danish children in grade 3. Results: The various versions of the dynamic measures in reading fluency and vocabulary, respectively, are expected to correlate strongly, whereas the dynamic measures are expected to show medium to high correlations with the corresponding static measures as well as with the teachers' perceptions of the children's potentials. Consequently, the dynamic measures are expected to identify a somewhat different group of students at risk for reading problems than the traditional, static measures do. Conclusions: The predictive validity of the dynamic measures should be further investigated in a longitudinal study.
Anja Gendt (University of Potsdam); Reinhold Kliegl - Eye movements are modulated by cognitive load and individual differences
A series of three experiments followed the question whether working memory (WM) load affects eye movements in the reading-span task (RST) and informs about the role of WM during reading comprehension. Readers must keep information active (i.e., the last words of sentences) while moving their eyes across several separately presented sentences with the goal of comprehending each sentence. Experiment 1 revealed a reading strategy that is very different from normal reading: The number of fixations on the to-be-remembered words was surprisingly high, and reading often started with the sentence-final word. Experiment 2 and 3, prevented the use of this strategy by masking the sentence until the initial word was fixated. That leads to shorter sentence processing times (SPT) compared to experiment 1. Although shorter SPTs yield more time for 'pure' memorizing, there was no significant difference in memory accuracy between the experiments. The last experiments included also young and old adults to increase the variance in memory span. Effects of age were obtained only in the condition without working memory load. The lack of clear evidence for age differences supports the idea that reading is an automated process preserved with age. Effects of load and span measures were observed in gaze durations, regressions and blink rates. With increasing WM load fixation durations increased dependent on individual resources. The upcoming idea is that a restriction of resources due to WM load may also restrict the perceptual span.
George Georgiou (University of Alberta)Rauno Parrila; Timothy Papadopoulos; Hollis Scarborough - Why is RAN related to reading? Manipulating the demands and the format of RAN tasks
Purpose: Scarborough and Domgaard (1998) suggested that if X is an aspect of the RAN that underlies the relationship of RAN with reading, then modifying the RAN task so as to increase or decrease the demand of X will result in an increase/decrease in the RAN-reading relationship. The purpose of this study was to examine why RAN is related to reading by modifying the demands and the format of RAN. Method: 65 Grade 2 and 65 Grade 6 Greek-speaking children were assessed on different RAN and reading measures. We manipulated the naming (naming the stimuli vs. cancelling out), the phonological encoding (5 items repeated 10 times, 2 items repeated 25 times, and 10 items repeated 5 times), the motor programming (co-articulation vs. no co-articulation), and the speech production (multisyllabic vs. two-syllabic stimuli) demands of the RAN tasks. Results: The results indicated that seriality, motor programming, and speech production do not influence the RAN-reading relationship. Higher correlations were observed when naming the stimuli as opposed to cancelling out stimuli or saying Yes/No and when the RAN tasks consisted of 10 items repeated 5 times as opposed to 2 items repeated 25 times. Equally strong correlations were observed between RAN and reading when the RAN task consisted of 4 rows of 4 stimuli as opposed to 5 rows of 10 stimuli. Conclusion: The findings of this study suggest that RAN is likely related to reading because of the unique demands of accessing and naming a set of stimuli.
Hope Gerde (Michigan State University); Tiffany Martoccio; Kelli Smith; Lori Skibbe; Ryan Bowles - The longitudinal relation between preschool children's name writing, letter knowledge, and self-regulation
Purpose: Several literacy skills predict name writing, including children's letter knowledge (Bloodgood, 1999; Molfese, Beswick, Molnar, & Jacobi-Vessels, 2006). However, less is known about how factors outside of the literacy domain, including social skills like self-regulation, relate to name writing development. And, although correlational knowledge has linked name writing to letter knowledge, the longitudinal relation between these variables and self-regulation remains unknown (e.g., Diamond, Gerde, & Powell, 2008). The present study moves beyond previous work to focus on the longitudinal associations between children's name writing, letter knowledge, and self-regulation over one year of preschool. Method: Preschoolers (N = 106; age M = 48 months in fall) were tested individually by trained research assistants in October, December, February and April. Children were assessed using a measure of behavioral regulation (Head-Toes-Knees-Shoulder, Ponitz et al., 2008), letter knowledge, and name writing (a writing sample in which the child was asked, "write your name"). Name writing was coded from 1 (scribbling) to 9 (name spelled correctly; Diamond et al., 2008). Results: Growth curve analyses (Raudenbush & Bryk, 2002) identified that 1) children's name writing, letter knowledge, and self-regulation skills increase linearly across the preschool year and 2) controlling for child age (centered), children's growth in self-regulation and letter knowledge uniquely predicted children's growth in writing sophistication. Conclusions: Results suggest that in preschool, name writing is facilitated not only by foundational concepts of literacy, such as letter knowledge, but also is supported by self-regulation, which may help children to maintain their focus on writing activities.
Astrid Geudens (Thomas More University College - Expertise centre Code)Kirsten Schraeyen - The nature of phonological representations in dyslexia: an in-depth analysis of segmental nonword repetition errors
Purpose: An important claim in current theories on dyslexia is that children and adults with dyslexia have difficulties in encoding, maintaining, and retrieving phonological representations. These characteristics have often been assessed with a nonword repetition task (NRT). Previous studies on nonword repetition skills in dyslexics suggested that dyslexics produce more errors than non-dyslexics on the NRT, due to poor quality of their phonological representations. However, "poor quality" in this context remains a rather vague description. To study the nature of dyslexics' phonological representations, the present study compared non-word repetition performance of young adult (16-years-olds) dyslexics and non-dyslexics focusing on the frequency and quality of their suprasegmental and segmental errors. Method: Nonword repetition (Scheltinga, 2003) data of 30 Dutch-speaking dyslexics and 30 non-dyslexics were analysed with respect to suprasegmental and segmental errors. Error analyses addressed the level of prosody, syllable retention and phoneme retention. Apart from substitution, omission, deletion errors we also focused on the difference between segment retrieval and representation by comparing the proportion of segments being correctly retrieved irrespective of syllable position versus the proportion of segments being correctly retrieved and correctly placed within the syllable (correct syllable encoding). Main finding Analysing the ratio of retrieval errors to encoding errors, we observed a significant difference between dyslexics and non-dyslexics. Interestingly, both groups did not differ with respect to the frequency of correctly retained segments in the NRT errors. Conclusion Our combined set of findings suggest that dyslexics' weak phonological representations in the NRT may not crucially depend on poor retrieval of segments but on the encoding of location in the syllable structure. A replication study with 20 Dutch-speaking dyslexics and 20 non-dyslexics is currently being undertaken.
Amelia Gill (Department of Women's and Children's Health (Paediatric Section))Elizabeth Schaughency; Barbara Galland; Andrew Gray; Carmen Lobb - Does snoring impair literacy development in pre-school children?
* Purpose Sleep disordered breathing (SDB) is associated with reduced learning in school-aged children, but little is known about learning in preschoolers with SDB. This study examined whether persistent snoring, a marker for SDB, negatively correlates with early literacy skills in preschool children. * Method This prospective case-control study compared 85 snoring and 85 non-snoring New Zealand 3-year olds matched on age (mean 3.8; range 3.2-4.0), gender (M/F = 1.4 ), SES, and body mass index (z-score ± 1.5). Snoring was determined through a validated parental questionnaire, and literacy measured using Early Growth Indicator Benchmark Assessments (EGIBA's): Letter Naming, First Sound Fluency, Picture Naming, Alliteration, and Rhyming. * Results In preschoolers, early literacy skills are just developing, and score distributions are often positively skewed, with zero scores and possible floor effects. Our data showed these characteristics, so for preliminary analyses data were normalised where necessary using log transformation. One-way ANOVAs were used to compare snoring and non-snoring groups, Mann-Whitney U tests were also performed on raw data to examine the validity of the data transformation. Average performance of the non-snoring group was higher than the snoring group for all early skills measured, with statistically significant differences found for Letter Naming (p<0.01) and Rhyming (p<0.01). * Conclusions Persistent snoring was associated with less well developed early literacy skills in otherwise healthy 3-year-olds, consistent with findings from school-age children. Further research should investigate the effect pre-school snoring has on later achievement and whether recovery from snoring during the pre-school years is associated with better academic outcomes later in childhood.
Jorge E. Gonzalez (Texas A&M University, VS)Sharolyn Pollard-Durodola; Deborah C. Simmons; Aaron jB. Taylor; Matthew J. Davis, Leslie Simmons - Developing Low-income Preschoolers' Social Studies and Science Vocabulary Knowledge through Content-Focused Shared Book Reading
Purpose This study evaluated the effects of integrating science and social studies vocabulary instruction into shared book reading with low-income preschool children. Specifically, we examined the efficacy of a shared book reading science and social studies intervention on the vocabulary development of low income preschool children. Further, we differential effects by learners' entry-level vocabulary. We were interested in the situated effects of intervention when delivered by preschool personnel as opposed to research teams. Method Twenty-one preschool teachers and 148 children were randomly assigned at the class level to either the Words of Oral Reading and Language Development (WORLD) intervention or a practice-as-usual condition. Children were screened to approximate three vocabulary levels (15th, 30th, and 50th). WORLD teachers implemented the intervention in small groups of 5-6 students, 5 days per week, 20 minutes per session for 18 weeks. Results Findings from multilevel models indicated statistically and practically significant effects of the WORLD intervention on standardized measures of receptive vocabulary (δT = 0.93) and on researcher-developed measures of expressive (δT = 1.01) and receptive vocabulary (δT = 1.41). Conclusions The WORLD intervention had an overall main effect, irrespective of entry-level vocabulary, a finding that speaks to its potential applicability in preschool classrooms. This study confirmed and extended the literature showing that children primarily from low-income homes can learn vocabulary when shared book reading focuses on high priority words, provides multiple opportunities to interact with vocabulary in small groups, relates vocabulary to prior knowledge and world concepts, extends over an 18 week period.
Debbie Gooch (Royal Holloway, University of London); Hannah Nash; Maggie Snowling; Charles Hulme - Executive function and motor skills in pre-school children at risk of dyslexia
Purpose. Dyslexia frequently co-occurs with other developmental disorders such as attention deficit/hyperactivity disorder (ADHD) and developmental co-ordination disorder (DCD). However, the nature of these comorbidities and their impact on the manifestation of dyslexia is not well understood. This paper reports preliminary findings from the first phase of a study of children at high risk of reading difficulties addressing the question of whether there are signs of comorbid difficulties in children at risk of dyslexia. Method. 202 children aged 3-4 years were recruited and classified as family risk for dyslexia (FR), language impaired (LI) or typically developing (TD). This resulted in 79 FR children (20 of which had an additional LI), 42 LI children and 81 TD children who completed measures of complex inhibition, selective attention and manual dexterity. Results. The three groups were matched for age although there were group differences in non-verbal IQ (TD>FR>LI). There was a clear step pattern on all the measures of inhibition, attention and motor skills (TD>FR>LI). Within the FR group there were moderate correlations between children's receptive and expressive language skills and their performance on the measures of executive function and motor skills. Conclusions. In comparison to TD children, pre-school children at risk of dyslexia performed worse on tasks tapping skills known to be impaired in children with ADHD and DCD. In interpreting these findings, it is important to consider the impact that current weaknesses in language skill may have on children's performance on these measures of executive function and manual dexterity.
John Goodrich (Department of Psychology, Florida State University); Christopher J. Lonigan; JoAnn M. Farver (Department of Psychology, University of Southern California) - Do skills in children's first language promote development of skills in their second language? An experimental evaluation of transfer
Purpose: Although a number of theories of bilingual education suggest that skills in children's first language (L1) "transfer" to their second language (L2), most evidence of transfer comes from studies demonstrating correlations between skills measured in both L1 and L2. It has been suggested that skills in L1 or L2 represent "conceptual knowledge" that promotes learning of skills in the other language. This study examined the role of L1 skills for acquisition of L2 skills in an experimental study. Method: In this study (see Farver et al., 2009, for impact report), 63 preschooler Spanish-speaking English language learners were randomly assigned to a control condition (children received only their classroom curriculum) or a group that received small-group pull-out instruction in oral language, phonological awareness, and print knowledge in addition to their classroom curriculum. All pull-out instruction was conducted in English and lasted an entire preschool year. Pretest and posttest measures of emergent literacy skills were administered to children in both English and Spanish. Results: Analyses examined whether L1 skills at pretest moderated the impact of the intervention on the measure of the same construct in L2 at posttest. Results demonstrated that pretest scores of L1 receptive vocabulary and elision moderated the effects of the intervention for L2 receptive vocabulary and elision. However, L2 pretest scores did not moderate the effects of the intervention on L2 skills. Conclusions: These findings partially support the hypothesis that L1 skills promote acquisition of L2 skills. Children with higher L1 skills benefited more from the intervention.
Amanda Goodwin () - Untangling Morphological Awareness from Phonological Processing in Predicting Reading Comprehension and Decoding for ELLs
This study explores the unique role of morphological awareness and phonological recoding in predicting word reading and reading comprehension in 197 Spanish speaking English language learners (ELLs). Structural equation modeling was used to isolate morphological awareness from phonological and orthographic confounds present in opaque morphological relationships by creating a latent variable stemming from shared variance of four morphological tasks with different levels of transparency. The constructs of morphological awareness, phonological recoding, and reading were assessed in this study through 8 indicators. Two measurement models were constructed representing 4th grade Phonological recoding, consisting of two nonword decoding tasks which assess the letter by letter processing when decoding pseudowords that embodies phonological recoding, and 4th grade Morphological Awareness, consisting of indicators containing morphological relationships with different levels of phonological and orthographic transparency. The structural model determined the unique contributions of phonological recoding and morphological awareness to standardized measures of word decoding and passage comprehension, showing that for these Spanish speaking ELLs, morphological awareness made a significant moderate contribution to passage comprehension but not decoding whereas phonological recoding made a significant large contribution to decoding but not passage comprehension. This study further untangles morphological from phonological abilities by clarifying how each contributes to word identification and reading comprehension for Spanish speaking ELLs. Educational implications of findings will be explored and related to classroom instruction.
Alexandra Gottardo (Wilfrid Laurier University)Esther Geva, Fataneh Farnia, Amy Grant, Mashid Azimi - Factors related to reading comprehension in second language learners in second grade.
The Simple View of Reading states that reading comprehension is related to word reading (decoding) and listening comprehension, RC = D X LC (Gough & Tunmer, 1986; Hoover & Gough, 1990). Other views of reading comprehension suggest that when word reading skill is well established, vocabulary knowledge alone predicts reading comprehension (Carver, 1997). The SVR has been tested with monolingual speakers (Catts, Adlof & Weismer, 2006) and is being tested with second language (L2) learners, mostly Spanish-English speakers in the United States (Gottardo & Mueller, 2009; Proctor, Carlo, August & Snow, 2005). Method: Data were collected from 157 children in second grade who were learning English as L2 in Canada. The children had a wide range of first languages, Portuguese, Chinese and Spanish. A variety of decoding and oral language measures including vocabulary and grammatical knowledge were administered. Results and conclusions: The analysis of the combined group data shows that OLP and decoding are related to reading comprehension. Subgroup data for the Spanish and Cantonese groups revealed somewhat different patterns of relationships. Similarities and differences based on subgroups will be discussed. Implications for existing models are included.
Stephanie Gottwald (Ctr for Reading and Language Research- Tufts University)Kathleen Spencer; Robin Morris; Maureen Lovett; Maryanne Wolf - Beyond the phoneme and the word: deficits in sentence level production in developmental dyslexia
Purpose: For two decades, the largest emphases in dyslexia research has concerned weaknesses in phoneme-level processes and word-level decoding. Cross-linguistic studies, however, have underscored the importance of understanding fluency or speed of processing issues at multiple levels. In English, weaknesses of rapid naming are strong predictors of text-level reading fluency. Many studies have verified the close relationship between general, non-specific language assessments and word-level reading accuracy, but little research has been done to identify the role of sentence-level oral production skills in reading text. This analysis investigates the relationship between a measure of sentence-level production and text reading fluency in elementary aged children with identified reading disabilities. Method: Readers with dyslexia (n= 265, mean age= 7.15 ) were tested on a text reading fluency task, a sentence production task and other tasks that assessed their abilities at varying levels of linguistic knowledge. Results: Results indicate that participants perform one standard deviation below the mean on the sentence production task. Sentence production skills contribute unique variance to text reading fluency after controlling for phonological awareness, vocabulary and naming speed. Conclusions: This analysis lends support to the claim that children with impairments in written language also exhibit weaknesses in specific aspects of linguistic knowledge. These results contribute to the ever-deepening view of reading disabilities which must account for weaknesses beyond phonological awareness and word-level decoding.
Jen Goudey (); Jan Frijters; Rauno Parrila; Maureen W. Lovett - A randomized controlled trial evaluating the efficacy of parent tutoring on reading.
Purpose - This study examined the efficacy of two 16-week parent-tutoring programs on reading achievement, immediately following the intervention and at one-year follow-up. Method - Fifty-eight children in Grades 2 to 4 were randomly assigned to one of three groups: (1) Paired Reading (PR; Topping, 2001); (2) modified PR that integrated the word recognition strategies of the Phonological and Strategy Training Reading Program (Lovett, Lacerenza, & Borden, 2000) (PR-PHAST); and (3) a wait-list control group. Reading measures were administered before the 16 week intervention, immediately following, and one year later. Parents' satisfaction with the intervention was assessed via questionnaire. Results - Repeated measures ANOVAs were conducted to assess change in reading achievement during the intervention phase as compared to the follow-up period. Immediate outcomes indicated superior gains by the PR-PHAST condition over the PR and control conditions on word, nonword reading, and cloze comprehension measures [partial eta squared effect sizes (ES) ranged from .11 (medium) to .32 (large)], but not for fluency and text comprehension (ES from .00 to .02). No differences between the PR and control conditions were found (ES from .00 to .02). Parents in the PR-PHAST condition reported their children were more motivated to read. At follow-up, a significant main effect of time was found, but no significant differences between conditions. Conclusions - Results suggest parents can provide effective coaching during oral reading at home with ongoing support, and that the observed gains in reading can be maintained a year following the intervention.
Jill Grifenhagen (Vanderbilt University)Catherine L. Darrow; Sarah S. DeLisle; David K. Dickinson - Teaching different word types during preschool book reading
Purpose. This study examined preschool teachers' vocabulary teaching practices during whole-class bookreading sessions: * Which word types are most often taught during bookreading? * Which strategies are most often used to teach these words? * Do the types of words taught relate to child gains in vocabulary? Method. Data was collected during an experimental study in 51 Head Start classrooms. In each classroom, bookreading was videotaped twice during the academic year. Sessions were coded for type of words taught and strategies used. Children's vocabulary gains were measured by the PPVT and EVT (n = 448). Descriptive data, correlational and regression analyses were examined. Results. In preliminary analysis, teachers were most likely to teach nouns during bookreading; 67% of words taught were nouns, 25% verbs, and only 8% adjectives, adverbs, or others. Nouns were most often taught through implicit definition (57% of nouns) while verbs and adjectives were most often defined explicitly (56% and 80%, respectively). Adjectives taught correlated with children's gains in expressive vocabulary (r = 0.42, p<.05). Teachers in the intervention condition taught more of each class, and defined more words during bookreading (p<.01). Overall, initial vocabulary knowledge predicted vocabulary gain, with children starting at a lower baseline gaining more (p<.001). Conclusions. While nouns were the predominate word class taught during bookreading, teaching rarer word types, specifically adjectives, may affect children's vocabulary learning. Instructional implications will be discussed during presentation. Further analysis will examine the relationship between vocabulary teaching during preschool bookreading and children's vocabulary through grade one.
Cherie Josefa Guerrero (Florida State University)Christopher J. Lonigan - Relations of English and Spanish oral language skills with bilingual preschool children's phonological awareness development
* Purpose - Research has found that phonological awareness (PA) in English and Spanish reflects independent abilities rather than a single cognitive ability. However, little study has been done on how PA in English and Spanish relate to other important correlates of reading acquisition. The purpose of this study was to examine the relations of oral language skills in both English and Spanish with PA skills as well as to confirm previous findings on the structure of PA in a sample of bilingual preschool children. * Method -176 children (44.9% female) were recruited from preschools in Miami, Florida. Children were administered the Definitional Vocabulary, Blending, and Elision subtests of the Preschool Comprehensive Test of Phonological and Print Processing (P-CTOPPP) and the P-CTOPPP-Spanish. * Results - An initial confirmatory factor analysis demonstrated that the PA measures were best represented as separate factors in each language. Structural equation modeling revealed significant paths between oral language skills and PA skills within each language. Additionally, English oral language contributed uniquely to Spanish PA. In contrast, Spanish oral language skills did not contribute uniquely to English PA. * Conclusions -These results suggest that English oral language skills are involved in the development of PA abilities in both English and Spanish for bilingual preschoolers. However, Spanish oral language skills appear to only be involved in the development of Spanish PA abilities in bilingual preschoolers. For bilingual children, oral language skills acquired in the home may not transfer to academic contexts where the child is primarily taught in English.
Ying Guo ( University of Cincinnati)Laura Justice;Joan Kaderavek - Relations between features of the classroom literacy environments and preschool children literacy skill gains
Purpose: Ecological theory emphasized the importance of classroom literacy environments in preschoolers' literacy learning. The classroom literacy environment includes the physical (access to literacy materials, range of literacy resources) and the psychological literacy environment (interactions between teachers and children). This study was to examine the relations among preschool classroom's physical or psychological literacy environment and children's literacy gains and also whether the relation between physical literacy environment and children literacy gains is moderated by psychological literacy environment. Method: Participants were 209 preschool children and their 30 classroom teachers in 38 centers. Children measures included the alphabet knowledge and name-writing ability. Classroom measures included the three dimensions of physical environment: book materials, literacy area and writing materials (observation checklist) and one dimension of psychological environment: instructional support (standardized observation tool). Results: Hierarchical linear modeling results showed that the quality of literacy area in the classroom is a positive and significant predictor of children's gains in alphabetic knowledge, but not name-writing ability. Moreover, writing materials is a positive and significant predictor of children' gains in alphabetic knowledge and name-writing ability only within the context of high quality, instructionally supportive classrooms. Conversely, children made less alphabetic knowledge and name-writing growth in classroom with low quality of instructional support, even when there is high quality of writing materials available in the classroom. Conclusions: findings confirm the value of excellent instructional support and the physical environment (as well as interplay) to promote children emergent literacy skills.
Nicolas Gutierrez-Palma (University of Jaen); Sylvia Defior; Nuria Calet - Prosodic Sensitivity and the Acquisition of Punctuation Marks in Spanish
This work aims to study the acquisition of punctuation marks, an aspect of the written script closely related to prosody and syntax. In particular, we focused on the role of prosodic sensitivity, i.e., the capacity to detect prosody in the oral language. Children from 4th and 6th grade participated in a cross-sectional study in which working memory and phonemic awareness were controlled for. They performed prosodic tasks (at the supra-lexical and lexical levels), a sentence-picture matching task (to measure syntax processing), a non-linguistic rhythm task, and a punctuation writing task. Results showed that prosodic sensitivity (either lexical or supra-lexical) accounted for a significant and unique variance of the punctuation task. Interestingly, as this effect was independent of the non-linguistic rhythm skills, it suggests that prosodic sensitivity is rather related to linguistic processing than to general non-linguistic skills. All these results are discussed in terms of how prosodic sensitivity may affect the acquisition of the prosodic aspects of reading, as the punctuation marks.
Thomas Günther (RWTH Aachen University Medical Center, Child Neuropsychology Division);Ralph Radach - The role of 'mindless reading' in the development of reading fluency
Purpose: We recently developed a string scanning task that has proven successful in approximating visual processing characteristics and eye movement patterns typically found in skilled adult readers. The purpose of the present study was to test and validate this paradigm with a sample of elementary school readers. Method: In the 'Landolt scanning task' strings of ooooo are arranged similar to a line of text and participants are asked to verify the presence of strings that contain an open target, as in oocoo. This task was compared to normal reading of age appropriate sentences. Twenty-four German elementary school students with reading experience between three and four years participated. Participants also completed a battery of tests including measures for reading fluency, spelling and comprehension. Results: Spatial components of oculomotor behavior like fixation probability and saccade landing positions were very similar those found in reading and local word fixation patterns appeared to be controlled by identical mechanisms. In contrast to prior results with adult readers, word viewing time measures were longer in the reading task, reflecting the higher cognitive workload associated with word processing for developing readers. Interestingly, the gaze durations in both reading and the Landolt scanning task correlated with performance in a comprehension test. Conclusion: In children, the Landolt string scanning paradigm can be used to examine in detail the visuomotor and attentional demands of reading without imposing any linguistic processing demands. This opens a new avenue to directly study key non-linguistic components of 'fluency' in a nearly natural task environment.
Jonathan Haenen (School of Psychology and CLS, University of Reading);Patricia Riddell; Tim Williams - Dual-task attention and the reading process
Purpose: Recent research into reading comprehension has ventured beyond looking at associated linguistic skills and considered the role of executive functions. Additionally, the Simple View of Reading theory prescribes that the two crucial processes of reading, decoding and comprehension, are dissociable. It is thought that certain executive functions might be common to both decoding and comprehension processes, creating resource competition. Thus skilled reading requires a degree of dual-task attention, especially if decoding processes have fail to become automatic. The study analyses the role and influence of dual-task attentional demands on comprehension and decoding outcomes. Methods: Reading and general cognitive abilities were measured in a group of 86 children (6.9-10.4 years). Tasks included reading comprehension and decoding (Neale Analysis of Reading Ability), working memory, dual-task attention, and established language measures. Results: Multiple hierarchical regressions indicate that dual-task attention marginally explains variance in reading comprehension outcomes after taking account of established linguistic and working memory predictors (β=.184, p=0.085). Splitting the group based on decoding ability indicates a much stronger relationship in children with poorer decoding skills (ΔR2=.105, p=0.002). Conclusion: There is limited evidence to suggest that dual-task attention plays a general role in comprehension outcomes of reading tasks. There appears to be a stronger "stress effect", where children who have general difficulty decoding (perhaps failing to automatise the decoding process) have a greater reliance on dual-task attention resources for successful comprehension.
Marleen Haentjens (University of Amsterdam); Peter F. de Jong - Does the length effect reflect only sublexical processing?
Purpose: In beginning readers an effect of word length, the increase of word reading time with each additional letter, is often observed. This effect is generally assumed to indicate the use of sublexical processing. However, this effect can also occur if lexical processing is predominantly used for shorter words whereas longer words are (still) processed sublexically. This alternative interpretation of the length effect is examined. We predicted that if the length effect reflects two types of processes then the reading speed of longer words cannot be entirely predicted from the reading speed of shorter words. If so, other abilities, such as phonological awareness and rapid naming, might have an extra contribution in the prediction of the reading speed of longer words. Method: Approximately 100 second grade children did a naming task consisting of words and pseudowords ranging from three to five letters. Tasks for vocabulary, phonological awareness, rapid naming and visual attention span were also administrated. Results: Preliminary results of 62 children (M=8.1 years) show that reading speed of 3-letter words could not entirely predict the reading speed of longer words. Additional effects of RAN and phonological awareness were found. Such additional effects were not found in the prediction of the reading speed of longer pseudowords after reading speed of 3-letter pseudowords was controlled. Conclusions: The results indicate that the length effect of words might reflect two processes: short words are read lexically, while longer words are read sublexically. In contrast, the length effect of pseudowords reflects only one (sublexical) process.
Bente Hagtvet (University of Oslo)Bente E.Hagtvet; Solveig A.H.Lyster - Language profiles in dyslexia: evidence from dyslexic parents and their offspring.
The study investigates the core problems of dyslexia via a comparative language study of dyslexic Norwegian speaking parents and their offspring. Two research questions were addressed: 1.What characterizes the language profiles of dyslexic adults/parents? 2. What relationships may be identified between the language profiles of the dyslexic parents and those of their developing children? One hundred and thirty three dyslexic parents and 140 children participated in the study. The parents' cognitive and oral and written language skills were assessed and their children's cognitive, phonological, semantic, syntactic and literacy skills were measured yearly from age five through nine. Despite their dyslexic condition many parents were well educated. Their mean non-verbal IQ was 115 while the children's mean nonverbal IQ was 108. The data analyses showed that the written language skills of the dyslexic parents were significantly below the norms of high school students, in particular for speeded word reading and reading comprehension. Weaknesses were also revealed in auditory short term memory and in manipulation of phonemes. These main results are in line with findings in deep orthographies. However, in a semi transparent orthography like the Norwegian the persisting core problems of dyslexia appear more strongly related to speed of information processing than in the precise manipulation of phonemes. Correlation analyses and ANOVAS showed that parents and offspring shared many similarities in both reading patterns and oral language profiles, but the language profiles of the children varied with developmental age.
Tuomo Haikio (University of Turku); Raymond Bertram; Jukka Hyona - To bold or not to bold? The effect of bolding as a syllabification cue on multisyllabic word recognition among Finnish 2nd graders
Purpose Studies in French and Spanish have shown that children recognize multisyllabic words via syllable units (e.g., Maïonchi-Pino et al., 2009; González & Valle, 2000). Häikiö, Hyönä and Bertram (in preparation) hypothesized that if this generalizes to Finnish, children should benefit from syllabification by hyphens (e.g., kah-vi 'cof-fee'). Surprisingly, they observed that Finnish 2nd graders were in fact disrupted by hyphenation at syllable level. On the other hand, post hoc analyses indicated that words with a bigram trough (i.e., less frequent bigram than the preceding and following bigrams) at syllable boundaries may be read faster than words without such trough. This suggests that Finnish 2nd graders do make use of syllables in multisyllabic word recognition. In the present study, we examined whether another syllable structure cue, syllable bolding (e.g., <b>kah</b>vi), is more beneficial for multisyllabic word recognition than hyphenation. Method Finnish 2nd graders read single sentences containing multisyllabic target words, while their eye movements were registered. We manipulated troughness at syllable boundaries alongside bolding as syllabification cues, generating four conditions. Half of the target words contained a bigram trough at the syllable boundary while half of them did not, and both trough conditions were presented with and without bolding as an additional visual syllabification cue. Lexical-statistical properties of target words were matched across conditions. Results and Conclusions We found that both bolding and troughness had an effect on several eye movement measures, suggesting that the syllable is a functional processing unit in multisyllabic word recognition for Finnish children as well.
Frances Hampson-Jones (Institute of Education, London)Jackie Masterson - Reciprocal teaching - an effective intervention to improve reading comprehension for children in the U.K. aged 9-10?
PURPOSE: Reciprocal teaching (Palinscar & Brown, 1984) has been successful in improving reading comprehension in the USA. Can this be replicated in the UK, in a whole class format, within the Literacy Hour? Can reciprocal teaching can be strengthened by the addition of another strategy visualisation? METHOD: Three Year 5 groups formed a reciprocal teaching group (RT), a reciprocal teaching plus visualization group (RTV), and a non-intervention control (NI). Both teaching groups received 10 hours of instruction over a period of 3 months. Using a pretest, posttest design, standardised measures of reading comprehension, accuracy and rate were obtained (using the NARA II), alongside qualitative measures of reading strategy use (MARSI, MRQ). RESULTS: There was a statistically significant improvement in reading comprehension and reading accuracy for the intervention groups over the control, but no difference in gains between the RT groups. The MARSI showed a significant increase in strategy use for the RT group over the control. There was a decrease in reading rate for both RT groups, whilst there was a slight increase in rate for the non-intervention group. CONCLUSIONS: The intervention improved comprehension and accuracy. However, there was no difference in the improvement made between the RT groups, showing that visualisation did not add significantly to the effect, on this time scale at least. There was a decrease in reading rate for the intervention groups, which would benefit from further examination as it runs contrary to the theory that increasing reading rate is a way to improve reading comprehension.
Gina Harrison (Educational Psychology & Leadership Studies, University of Victoria)Keira Ogle; Megan Keilty - L2 Component reading skills predict L2 writing in 5-6 year old ESL Canadian children
ABSTRACT Purpose: This study examined the contribution of oral English, phonological processing, and alphabet knowledge to reading, spelling, and writing in 5-6 year old ESL and native English speaking Canadian children. Method: Measures of nonverbal reasoning, oral language, phonological processing, alphabet knowledge, word level reading, spelling and early writing were individually administered in counterbalanced order to 123 children (67 ESL, 56 non-ESL) at the midpoint of their first year at school. Punjabi was the L1 of most (95%) of the ESL children. Results: ESL children achieved significantly lower scores than non-ESL on the English oral language measures for vocabulary, F(1,120) = 94.70, p < .0001) and syntax, F(1,120), = 60.72, p < .001 and on the non-word repetition measure of phonological processing, F(1,120), = 10.94, p < .001. ESL children achieved significantly higher scores on one of the alphabet knowledge measures assessing alphabet writing fluency, F(1,120), = 5.75, p < .01). Correlation analyses identified potential predictors of writing and hierarchical regression analyses were run separately for ESL and non-ESL groups including these predictors. For ESL, sound matching was a significant contributor to the model, F(1,66) = 38.26, p < .0001 accounting for 37% of the variance t(66) = 11.34, p < .0001 and alphabet identification explained an additional 16% of the variance in L2 writing (R2 change = .168) independent of oral English vocabulary or syntax skills. Conclusions: L2 phonological processing and alphabet measures can provide valid indications of ESL children's burgeoning L2 writing skills, irrespective of oral English proficiency.
Sara Hart (Department of Psychology, Florida Center for Reading Research); Stephen Petrill - Exploring the genetic effects underlying the relationship between general executive functioning and reading performance
Purpose. The present study examined the relationship of various executive functioning abilities on assorted components of reading performance. Method. Participants were drawn from the Western Reserve Reading and Math Project, an ongoing longitudinal twin project of 430 same-sex MZ and DZ twins from Ohio. The analyses are based on tester-administered batteries of choice Reaction Time (RT) and working memory, and psychometric measures of reading decoding and comprehension when the twins were 12yrs old. Results. Pearson correlations suggested that working memory (reverse coded) and choice RT measures were significantly associated with the decoding and comprehension reading performance measures (r=.15-.38). A measurement model containing four latent factors was created representing Choice RT, Working Memory, Decoding and Comprehension. These latent factors were then decomposed into common and unique genetic, shared environmental and nonshared environmental influences. These results suggested that there were unique genetic influences on Choice RT, which were separate from common genetic effects underlying WM, Decoding and Comprehension. Beyond those genetic influences associated with Choice RT and WM, decoding and comprehension also suggested separate overlapping genetic influences specific to reading. Conclusions. In total, these analyses will allow for a better understanding of the etiology of the relationships in question, mainly the genetic association of various executive functioning abilities with secondary abilities such as reading performance. These results would suggest that choice RT and working memory have differential genetic effects with reading outcomes. Moreover, there are distinct genetic influences that are related to reading above those related to general executive functioning.
Rebecca Hartmann (Free University of Berlin)Michael Becker; Nele McElvany; Camilla Rjosk - The relationship of reading acquisition to the incongruity between school and family language
Purpose: Text comprehension, vocabulary and reading self-concept are crucial components of reading literacy and are bidirectionally linked (Byalistok, 2002; Chapman & Tunmer, 1995). International studies show significant differences in reading skills between students with and without immigrant backgrounds (Stanat & Christensen, 2006). The degree to which these achievement gaps are caused not only by incongruity between school and family language but also by differences in socioeconomic status and cultural capital is disputed. Moreover, group-specific predictive power of socioeconomic status and cultural capital for reading achievement is discussed. Method: The analyses are based on data of the Berlin Longitudinal Reading Study. Data of N=712 students, tested in third, fourth, and sixth grade, were included. 32.4% of the participating students spoke a foreign language in their homes. Analyses were conducted using cross-lag panel- and multiple group models. Results: The incongruity between home and school language is a significant predictor of reading literacy in grade three and of the increase in reading skills from grades three to six. The successive introduction of socioeconomic status and cultural capital into the model increases the percentage of explained variance, but leaves home language a significant predictor. Multiple group analyses show group-specific differences in predictive power. Conclusions: Confirming the crucial role of home language for reading acquisition, the findings demonstrate the need for intensified reading training for L2 learners to prevent suboptimal reading acquisition by students with immigrant backgrounds.
Meghan Hauptli (Florida Center for Reading Research)Yaacov Petscher; Barbara Foorman; Laura B. Lang - Investigating the relation between reading fluency and word analysis with a success probability rating for adolescents on a state accountability reading assessment
This study investigates the relation between reading fluency and word analysis (spelling) with the FCAT Success Probability (FSP) rating. This year, Florida has implemented Florida Assessment for Instruction in Reading (FAIR), a computer-based state-wide progress monitoring system. The tool is intended to assist teachers in individualizing instruction thereby improving student performance on FCAT, the state's criterion-referenced accountability assessment. Specifically, we asked: are fluency and word analysis predictive of FSP? The FSP is derived from a computer-adaptive reading comprehension test with multiple choice questions written to FCAT item specifications. Fluency and word analysis assessments are components of a targeted diagnostic inventory intended to assist teachers in meeting struggling students' specific instructional needs. The fluency measure is a timed, maze task and measures gist-level comprehension. The word analysis task is a computer-adaptive spelling test. As of September 2009, one of three waves of data collection has been completed. Waves two and three are scheduled for January and April 2010. Participants include Florida public schools ninth, tenth, eleventh, and twelfth grade students (n ≈ 300,000). Multiple regression analyses will be used to investigate the strength of relation between the variables. All analyses will be completed in May 2010. Any limitation in this proposal due to ongoing data collection is offset by the substantial sample size and the relevance of this understudied population, adolescent readers. Investigating the broad screen's algorithm for FSP through the secondary assessments will enable us to validate their usefulness in predicting success on the state's reading accountability measure.
Jarkko Hautala (Researcher); Jukka Hyönä; Mikko Aro; Heikki Lyytinen - Eye movement evidence for sublexical effects of repeated reading
Purpose. The stability of sublexical and whole-word reading strategies is studied in the face of repeated reading among highly skilled and poor readers in transparent Finnish orthography. A previous study has demonstrated a lexical but not a sublexical effect of repeated reading (Martens & de Jong, 2008). Method. Fluent adults (N=17) and dysfluent children (N=16) read aloud word and nonword lists ten times while their eye movements were recorded. Within words and nonwords the number of letters (NoL) and syllables (NoS) were independently manipulated while controlling for orthographic properties. Fixation-based measures were log-transformed to attain equal variances between groups and to prevent proportionally equal effect sizes from producing significant interactions. Results. Repeated measures MANOVAs were conducted using a 2 (Group) x 2 (Block) x 2 (Lexicality) x 3 (NoL or NoS) a mixed-factor design. There was a significant 4-way interaction in NoL -inspection, F(2,30)=4.079, p<.050,η²p=.214 in total fixation times. In adults, the nonword length effect, whereas in children the word length effect (but only in non-transformed values), was more reduced by repetition. In NoS-inspection the 4-way interaction was significant in gaze duration, F(2,30)=4.039, p<.050,η² p=.212, as adults allocated extra attention to trisyllabic nonwords during Block 1. Conclusions. In both groups, sublexical reading (as indexed by the length and syllable effect) was reduced by repetition, with children improving more their word reading while adults improving their nonword reading. Adults can more rapidly form orthographic whole-word representations for nonwords, whereas repetition appears to encourage dysfluent children to utilize their existing whole-word representations.
Jelena Havelka (University of Leeds, UK) - Effects of AoA and frequency on picture and word naming in transparent orthography: Evidence from Serbian
Purpose Two experiments examined the effect of AoA, frequency, and word/picture name length on picture (Experiment 1) and word naming (Experiment 2) in Serbian. The aim of these experiments was to establish the relative contribution of these variables to picture and word naming in a perfectly regular orthography. Method The experiments were based on a correlational design. In Experiment 1 thirty-two native Serbian speakers named 268 pictures in a picture naming task and in Experiment 2 another 32 native Serbian speakers named 268 pictures in a word naming task. Results Our results reveal AoA as a significant predictor of naming latencies in both experiments, but the proportion of variance accounted for by AoA is significantly larger for pictures compared to words. Frequency had an equivalent and small effect in both experiments. Word/picture name length had a significantly larger effect on word naming than on picture naming. AoA emerged as the main predictor of reaction times in picture naming in Serbian, replicating significant effects of AoA from previous picture naming experiments (for a review see Juhasz, 2005). This implies that this effect is robust and independent of the language in which the pictures are named. In contrast, the effect of AoA on word naming in Serbian was small, and only marginally significant. Conclusions These results provide support for the models which propose that AoA effects are modified by the regularity of the mapping between the input and the output.
Stefan Hawelka (University of Salzburg, Austria); Benjamin Gagl; Heinz Wimmer - A dual-route perspective on eye movements of dyslexic readers
This study assessed eye movement abnormalities of adolescent dyslexic readers and interpreted the findings by linking the Dual-Route model of single word reading with the E-Z reader model of eye movement control during silent sentence reading. A dysfunction of the lexical route was assumed to account for a reduced number of words which received only a single fixation or which were skipped and for the increased number of words with multiple fixations and a marked effect of word length on gaze duration. This pattern was interpreted as a frequent failure of orthographic whole-word recognition (based on orthographic lexicon entries) and on reliance on serial sublexical processing instead. Inefficiency of the lexical route was inferred from prolonged gaze durations for singly fixated words. These findings were related to the E-Z reader model of eye movement control. Slow activation of word phonology accounted for the low skipping rate of dyslexic readers. Frequent reliance on sublexical decoding was inferred from a tendency to fixate word beginnings and from short forward saccades. Overall, the linkage of the dual-route model of single word reading and models of eye movement control lead to a useful framework for understanding eye movement abnormalities of dyslexic readers
Michael Hebert (University of Nebraska-Lincoln)Amy Gillespie; Steve Graham; Don Compton - Using writing to improve reading: A meta-analysis
Purpose. The purpose of this report was to collect, categorize and analyze research examining effects of various writing practices on reading outcomes. Specifically, the researchers asked: Is student comprehension enhanced through writing activities, including analyzing or interpreting text, summarizing, note-taking, and answering questions? Method. Meta-analysis was used to examine empirical evidence gathered from experiments, as well as quasi-experiments including pretests. Studies were included if: (a) participants were in grades 1 through 12, (b) the treatment group wrote or received writing instruction, (c) participants in the control group did not write or receive writing instruction, (d) reading measures were used as dependent variables, and (e) data needed for calculating effect sizes was provided. Effects based on standardized outcome measures (SOMs) were analyzed separately from effects based on researcher created measures (RCMs). An average weighted effect size (ES) was calculated for overall effects of writing on reading. Due to a lack of SOs in each category, effects were calculated for each writing type separately only for RCMs. Results. Overall effects of writing on reading were moderate for both SOMs (ES = 0.40) and RCMs (ES = 0.51). Significant effects were also found for each writing type: analyzing or interpreting text (ES = 0.77), summarizing text (ES = 0.52), taking notes (ES = 0.47), and answering questions about text (ES = 0.27). Conclusions. Evidence from this analysis identifies effective writing practices for improving reading. Emphasis should be placed on utilizing these practices for the enhancement of learning from reading across the content areas.
Lindsay Heggie (Queen's University), Lesly Wade-Woolley, Pam Briand - How do children read multisyllabic words?
PURPOSE. Prosody, the linguistic stress and intonation patterns evident over multiple syllables, has been shown to be related to reading ability (e.g., Goswami et al., 2002; Holliman, Wood, & Sheehy, 2008). Children are expected to read longer words even in early grades, and skilled readers encounter long and unfamiliar words on a daily basis. This study set out to examine the types of errors students made when reading multisyllabic words. METHOD. English-speaking Grade 3 and Grade 8 students (n=80) were recruited for this task. Participants were given a list of fifty low frequency words (ten each of 2- through 6-syllable words) to decode. Each group's words were matched for mean frequency and contained items that conformed both to typical and atypical stress placement. Participants also completed measures of prosodic sensitivity, reading ability, vocabulary, and nonverbal intelligence. RESULTS. It is expected that the types of errors that participants make will vary depending on reading ability; I expect that poor readers will make errors on larger-sized units (e.g., syllables), while better readers will be more likely to make errors on smaller units (e.g., phonemes). Further, the better the reader, the more likely he or she will be to put the word's main stress on the correct syllable when reading the unfamiliar word aloud. Words with both typical and atypical stress placement will be read more accurately by Grade 8 students than Grade 3 students, but the difference between groups will be greater for typical stress placement than atypical. CONCLUSIONS. Data collection is ongoing. Prosody is implicated in decoding.
Riikka Heikkilä (Niilo Mäki Institute, Jyväskylä);N. Aro; V. Närhi; T. Ahonen - Who benefits from syllable recognition? A computer-based fluency intervention with second and third grade poor readers.
Purpose: The basic feature in reading disabilities in regular orthographies is relatively accurate but slow reading. Recent intervention experiments with Finnish poor readers indicate that repeated reading of syllables increases reading speed also on the word level. This study aims to explore if the same effects can be achieved with computer-based repeated recognition of syllables and whether the length or frequency of the syllable or rapid naming interacts with the effect. Method: A sample of 152 second and third grade poor readers was divided in a control group and three intervention groups varying in length and frequency of trained syllables. Syllables were repeated 50 times during the ten intervention sessions. Results: A training effect was observed in all intervention groups but was stronger with longer and infrequent syllables. Transfer to the word level was significant only in four-letter infrequent syllables. Rapid naming had significant effect on the initial level of reading speed but not on the training effect. Instead, the initial reading speed explained the training effect. Conclusions: As the fluency problems are known to be permanent in nature these results show that they are not totally resistant to intervention. However, the items that students are already broadly exposed to may need more repetitions than the more infrequent ones.
Stephanie Herppich (University of Goettingen); Roel van Steensel; Nele McElvany; Jeanne Kurvers - Effects of family literacy programs: results of a meta-analysis
Purpose: Family literacy programs are seen as a means of promoting children's literacy development. However, current empirical evidence does not allow generalizable conclusions about effectiveness. Many reviews analyze the effects of parent involvement in general, and do not permit specific conclusions about literacy interventions. Moreover, literacy is often included in more general categories of effect measures, or seen as an undifferentiated ability. Recently published meta-analyses that do focus on family literacy include relatively small numbers of studies, that are sometimes outdated or have a very specific focus. The current meta-analysis was set up to solve at least part of these problems. Method: We analyzed the results of 26 recently conducted effect studies of family literacy programs (1990-2007), covering 41 samples. We categorized effect measures following the distinction between 'comprehension' and 'code' skills, to see whether effects differed. Additionally, we examined the effects of three types of moderators: program, sample, and study characteristics. Results: The mean effect size (ES) was .25 (a small, but significant effect). There were no differences between comprehension- and code-related measures (d=.28 and d=.24). The magnitude of effect sizes was influenced by three variables: content (larger ES for programs that provided more than only shared reading), age (larger ES for older children), and sample selection (smaller ES for randomized studies). In a combined analysis, randomization remained as the only significant predictor. Conclusion: The results call for more research into how these programs are incorporated in existing family literacy practices, and how they can positively affect those practices.
Elfrieda H. Hiebert (University of California, Santa Cruz);John Stewart; Masa Uzicanin - A comparison of word features affecting word recognition of at-risk beginning readers and their peers
Purpose: Understanding what it is that students of particular levels can and can't do is central to the design of appropriate and effective Response to Intervention (RtI) (Fuchs & Fuchs, 2006). Often, RtI has not been informed by data on the learning trajectories of students at different risk levels. This study analyzes the words read by students classified as at-risk, some-risk, and low-risk. Method: The database drew on oral reading fluency (ORF) performances of 2000 randomly selected 1st graders and 2000 randomly selected 2nd graders from a database of 3-million students. 1st graders were designated as: 12% (at-risk), 29% (some-risk), and 59% (low-risk) and second graders: 26%, 36, and 38%. Corpora consisted of 173 unique words from 1st-grade assessments and 368 from 2nd-grade assessments. Words were coded for length, frequency, vowel/syllable patterns, initial consonants, morphological family size, rime family size, and imagery. Results: Multivariate logistic regressions were conducted for 1st- and 2nd grades, where word recognition was regressed onto word features, students' ORF risk level, and interactions of risk level with word features. At-risk and some-risk readers were significantly more inhibited by word length than low-risk readers. At-risk readers made greater use of word imagery than other students but were less sensitive to word frequency. Conclusions: Understanding what it is that at-risk students know at particular points is critical to designing effective responses to interventions. This study draws on a substantial database to provide such information. References Fuchs, D., & Fuchs, L.S. (2006. Introduction to response to intervention: What, why, and how valid is it? Reading Research Quarterly, 41, 92-99.
Kerry Hofer (Vanderbilt University)Karen S. Anthony; Cathy Yun; Dale C. Farran; Mark W. Lipsey - The Renfrew Bus Story: An Investigation of the Elements of a Measure of Early Narrative Skills
Purpose This study examined a measure of children's early language skills that is growing in popularity: the Renfrew Bus Story. Authors examined the relationship among scoring categories of the Renfrew, and also looked at the predictive relationship between children's preschool language ability and their kindergarten achievement. Method This study was conducted as part of a larger randomized scale-up of a preschool mathematics curriculum. The sample included 617 children in 31 urban low-income preschool classrooms (public and Head Start). Quantitative analyses were used to examine the correlations among the five Renfrew skill components. Multilevel regressions were used to examine the relationship between children's Renfrew scores in preschool and their standardized achievement in preschool and kindergarten. Results Results indicated that children who took a longer period of time to retell the story had significantly lower utterance lengths, information scores, and independence scores. Additionally, children with higher initial language abilities had higher math scores at the end of preschool. This relationship was not seen for the other two math or literacy standardized outcomes. Additionally, children with higher language abilities at the end of preschool had higher achievement scores in all subjects measured at the end of kindergarten. Conclusions This study addresses important questions about an increasingly popular measure of early language. In particular, both the inter-relationships between various Bus Story score components, as well as the measure's predictive validity were examined. Findings suggest that children's retelling scores at the beginning of prekindergarten predict their later achievement through kindergarten in both literacy and math.
Tiffany Hogan (University of Nebraska - Lincoln); Hugh Catts - Early reading predicts later nonword repetition in children with dyslexia, language impairment, or both
Purpose: Phonological awareness and word reading are reciprocally related. In this study we predicted that another phonological processing task, spoken nonword repetition, would be reciprocally related to word reading. Moreover, we hypothesized that early and later word reading would predict increases in nonword repetition more so than early phonological processing in subgroups of children with dyslexia, language impairment (LI), or both. Method: Structural equation modeling was used to examine 2nd and 8th grade word reading and nonword repetition in a longitudinal dataset of 527 children. Subgroup analyses examined these relations in children with dyslexia only, dyslexia + LI, and LI only. Kindergarten letter identification and phonological awareness were used to predict later nonword repetition. Results: Second grade word reading predicted 8th grade nonword repetition (β= .190) and the reverse was true (β= .158). Further, in children with dyslexia with and without LI the influence of 2nd grade word reading on 8th grade nonword repetition was minimal (dyslexia: β = .002; dyslexia + LI: β = .001), whereas in those with LI only 2nd grade word reading was significantly associated with improved 8th grade nonword repetition (B = .362). In subgroups, kindergarten letter knowledge, not phonological awareness, predicted nonword repetition at both grades. Conclusions: Children's word reading and nonword repetition are reciprocally related. In children with language or reading deficits, early letter knowledge predicts later nonword repetition better than early phonological processing. Findings highlight inter-related developmental changes in orthographic and phonologic representations.
Andrew Holliman (Coventry University)Jane Hurry - The impact of Reading Recovery three years after intervention
Purpose: Reading Recovery (RR) is a school-based early intervention programme for children with the lowest literacy achievement in the first year of school. The RR programme is well known to have impressive effects in the shorter term, however less is known about its long-term effectiveness. Method: The present study followed up 73 pupils who received the RR intervention in the year 2005, along with 48 control children from RR schools who did not receive the intervention, and 119 control pupils from equivalent non-RR schools who did not receive the intervention. These groups were compared three years later in the year 2009 using their national assessments for reading, writing, and mathematics. Results: The results showed that RR children were achieving age appropriate levels in reading and writing which indicates being on track for reaching national target levels by the end of primary/elementary schooling. After controlling for scores at baseline, RR children significantly outperformed both control groups in reading and writing, but not mathematics. Moreover, RR children were significantly less likely than control children to be identified as having Special Educational Needs (SEN Code of Practice, 2008) at age 8 years. Conclusions: These findings indicate that effects of the Reading Recovery intervention are still apparent at age nine years. This suggests that RR children are still benefitting significantly three years post intervention.
Virginia M. Holmes (University of Melbourne)Sachiko Kinoshita - Unexpectedly poor spelling in English speakers
Purpose: The written system of English is basically alphabetic, but allows a variety of graphemes to represent a given phoneme. If enough phonemes in a word are spelt consistently, people can read a word containing less common phoneme-grapheme correspondences, such as somersault, even if they can't reproduce the exact spelling. This disparity between reading and spelling in some individuals becomes so large that they are deemed unexpectedly poor spellers. Why is this so? Method: From a sample of university undergraduates, 18 unexpectedly poor spellers and 18 good readers - good spellers were selected. They classified regularly spelt words (harvest), strangely spelt words (receipt), items with transposed letters (selifsh) and standard nonwords (platcher) in a printed lexical decision task. In a new orthographic processing task, an intact word (either short or long) was presented briefly, followed by either the same word (transfer-transfer) or a word with transposed letters, either adjacent (prospect-propsect) or two letters distant (dramatic-dratamic). Results: Unexpectedly poor spellers were less efficient than good readers - good spellers at classifying items in all conditions apart from standard nonwords in the lexical decision task, and less efficient at detecting transpositions in all conditions of the matching task. Conclusions: These results show that unexpectedly poor spellers access appropriate orthographic representations in their lexicon, but find them indistinct or difficult to scan. Moreover, they have trouble coding letter order even when a template of the orthographic representation is available in temporary memory. This effortful orthographic processing hinders the creation of precise orthographic representations necessary for accurate spelling.
Ingrid Hoonhorst (no member) - The development of speech perception in childhood : comparison between speech, colors and facial expression
Purpose: The aim of the study was to investigate two hypotheses proposed to account for the development of speech perception in childhood. Whereas for the reading hypothesis (Burnham, 2003) explicit reading instruction is responsible for the improvement of speech perception, the general cognitive hypothesis (Lalonde & Werker, 1995) postulates that the perception of speech sounds evolves in synchronicity with other abilities through the influence of cognitive maturation. Method: we compared the development of the categorical perception of voicing, colors and facial expressions in French-speaking children (aged from six to eight years) and adults. Both Relative Categorical Perception, i.e. the correspondence between identification and discrimination performances, and Boundary Precision, indexed by the steepness of the identification slope, were investigated. Results: none of the two hypotheses under scope was confirmed by our results. (1) Reading acquisition did not influence relative categorical perception but did influence boundary precision. (2) We did not evidence a single pattern of development but rather a differential development for the color continuum compared to the voicing and facial expression continua. Conclusion: the structural complexity of the categories might be one of the determinants responsible for the differential development of categorical perception.
Jane Hornickel (Auditory Neuroscience Lab www.brainvolts.northwestern.edu, Northwestern University, USA); Steve Zecker; Nina Kraus - Subcortical neural markers of reading and speech-in-noise impairments in children
Purpose: The spectral and temporal characteristics of speech syllables are maintained with extreme fidelity in the auditory brainstem and neural response characteristics can be quantified in individual subjects. Brainstem activity is also experience-dependent; malleable with online, short-term, and lifelong experience and related to cognitive functions such as reading and speech-in-noise perception (reviewed in Tzounoupolis and Kraus, Neuron, 2009). Here we review how brainstem function relates to reading and speech-in-noise perception in impaired and expert populations. Method: Brainstem responses were elicited by stop consonant syllables in children with developmental dyslexia, typically developing children, and auditory experts (musicians). Results: Neural measures of timing representing stop consonant stimulus differences relate to reading ability and speech-in-noise perception in children (Hornickel et al., PNAS, 2009). Additionally, typically developing children show enhanced brainstem representation of features important for "tagging" a target voice in background noise with stimulus repetition, while children with reading impairments do not (Chandrasekaran et al., Neuron, 2009). Importantly, the elements of the response that are deficient in children with reading impairments are enhanced in adults with lifelong musical training (Parbery-Clark et al., J Neurosci 2009). Conclusions: We have identified objective neural markers of reading and speech-in-noise perception, relevant to the management of developmental dyslexia and believe auditory training is a viable option for individuals with deficiencies in this aspect of auditory processing. We maintain that greater attention to and thoughtful processing of sound through auditory training could lead to improved neural encoding of speech in children with reading impairments in a top-down fashion.
Chung-Hui Hsuan (Asia University, Taiwan)Chen-Hui Su; Bi-Chin Chen; Yin-Juin Chen; Han Wang; Mong-Ju Yu - Longitudinal study of phonological processing, visual memory, RAN, morphological awareness, listening comprehension and reading acquisition upto Grade 1
Purpose: Reading comprehension is based on word decoding and listening comprehension (Gough & Tummer, 1986). Children in Taiwan start learning to read in Grade 1, after acquiring how to use Zhuyin, a phonetic system consisted of 37 symbols and 4 tones representing onset and rime, to transcribe Chinese syllable. This study aims to investigate the contribution of cognitive skills underlying Chinese character reading and listening comprehension during the first year of learning how to read. Method: 283 preschoolers (mean age 6;08) were recruited, among them 260 continued participating upto Grade 1. Phonological processing (tonal awareness, syllable deletion, decoding of Zhuyin in accuracy and fluency), RAN, visual memory of Chinese graphic units, morphological awareness, and listening comprehension were first assessed (April to May 2009, Time 1) in preschool after learning Zhuyin system. Character reading ability was assessed by the beginning of Grade 1 (September 2009, Time 2). The aforementioned skills, plus reading comprehension, will then be assessed by the end of Grade 1 (April to May 2010, Time 3). Results: Data analyses will be performed in June 2010 by correlation and regression. Conclusion: The results will give an insight to (1) concurrent (Time 1 and Time 3) and longitudinal (Time 1 to Time 3) relationships between listening comprehension and cognitive skills of character reading; (2) the significance of longitudinal reading predictors (Time 1) at Time 2 and Time 3; (3) the significance of concurrent predictors for character recognition and reading comprehension at Time 2.
Laura Hume (Florida State University, Florida Center for Reading Research)Christopher J. Lonigan - Relations between parent literacy-promoting practices, child literacy interest, and emergent literacy skills
Purpose: An increasing body of research has been devoted to the domain of emergent literacy--the skills, knowledge, and attitudes that are precursors to reading and writing success. There is some evidence linking children's early interest in literacy to emergent literacy skills. This study examined how one possible contributor to literacy interest--parent practices are causally related to children's developing emergent literacy skills through their impact on children's literacy interest. Determining the aspects of parent practices that encourage literacy interest and early literacy skills could lead to the development of methods to increase these skills and prevent further literacy skills deficits. Method: This study used Structural Equation Modeling with a sample of 789 3- to 5-year-old children. Parents completed surveys assessing home-literacy-related practices and the literacy interest their children displayed. Children were assessed on phonological awareness, print knowledge, and oral language skills. Results: Specific practices of parents such as teaching and exposure to literacy (the frequency children are read to and parents familiarity with children's books) had significant positive effects on children's literacy interest. However, while controlling for parent teaching and exposure children's literacy interest had limited effects on phonological awareness, print knowledge, and oral language. In contrast, parent exposure, was found to uniquely contribute to all three early literacy skills. Conclusions: Results suggest children's literacy interest has little impact on the development of early literacy skills. However, the degree to which parents provide children with exposure to literacy opportunities does seem to be an important component in the development of early literacy skills.
Jane Hurry (Institute of Education, Uni of London) - Conceptual awareness and its role in children's spelling
Purpose The importance of conceptual understanding has a long tradition (eg. Piaget, 1978; Karmiloff-Smith, 1992). In the context of literacy, the importance of understanding the alphabetic principle is a well-known example of the significance of conceptual understanding (eg. Frith, 1985). Less is known about the importance of a conceptual understanding of morphology. Conceptual knowledge may be either implicit or explicit. We wish to test the hypothesis that both forms of conceptual knowledge will be positively associated with spelling ability, and that children with explicit conceptual knowledge (both of phonetic and morphological principles) will be the best spellers. Method 256 7-8 year old children were assessed on word spelling to measure spelling ability. They were also assessed on pseudo-words to measure their knowledge of spelling rules. They were interviewed on their spelling strategies which were then coded for conceptual awareness. Finally, their performance on national assessments of writing/spelling were collected at the beginning and end of the school year. Results Preliminary analysis show a significant relationship between explicit conceptual knowledge of spelling articulated by the children and: word spelling and non-word spelling; national assessments of writing spelling at the end of the school year. Further analyses will explore the role of implicit knowledge of spelling rules. Conclusions These results suggest that explicit tuition of spelling rules, and an encouragement of explicit knowledge in children, will improve children's spelling. Whilst attention is currently given to some aspects of explicit tuition, eg. certain phonic rules, morphological rules, tend to be neglected.
Martina Huss (Faculty of Education, University of Cambridge, UK); Tim Fosker; Natasha Mead; Usha Goswami - Basic auditory processing skills and phonetic discrimination in developmental dyslexia
Purpose: Children with developmental dyslexia have a specific problem in representing the phonological structure of speech. A possible cause is a perceptual deficit in temporal auditory processing. Difficulties in the auditory perception of amplitude modulations and envelope onsets in speech, both basic acoustic features associated with speech rhythm and syllabic structure, may be causally related to impaired phonological representations and poor reading ability. Method: We tested if a group of 45 children with developmental dyslexia showed compromised auditory processing of amplitude envelope onsets (rise-times) and slow amplitude modulations compared to a sample of 27 age-matched typically-developing children. We measured auditory discrimination thresholds for both synthetic speech stimuli (ba/wa), contrasting fast formant transition cues with amplitude modulation cues and a series of simple sinusoidal stimuli, with varying rise-times, durations, intensities and frequencies. Results: Dyslexics had difficulties in processing rise-times and showed poor phonetic discrimination based on envelope cues; however, they displayed good phonetic discrimination based on formant transition cues. Multiple regressions showed that rise-time sensitivity was a unique predictor of reading acquisition, suggesting a special role for the perception of rhythmic timing in the development of phonological representations and literacy. Conclusions: We concluded that indeed dyslexic children have difficulties in the auditory perception of rise-time and amplitude envelope structure and that the deficit is related to poor phonological representations and poor reading ability. Furthermore, the results suggest that auditory processing of rise-time could be a useful sensory marker of developmental difficulties with phonology and literacy.
Florian Hutzler (University of Salzburg); Isabella Fuchs; Benjamin Gagl; Stefan Hawelka - Preview benefit revisited: Fixation-related brain potentials question eye-tracking evidence
Evidence on parafoveal preprocessing and on the question, which information is integrated across saccades is mainly based on eye tracking studies using the boundary paradigm. This technique allows to change stimulus displays dependent on a participant's gaze position: The preview that a reader has of a parafoveal target word can thereby be experimentally varied. In the present study, we used the new fixation-related brain potentials (FRP) technique to explore the electrophysiological correlates of parafoveal preprocessing: Brain potentials and eye movements were simultaneously recorded during the natural reading of lists of words. Parafoveal preprocessing of target words was either enabled or disabled using the boundary paradigm: In the latter condition, target words were masked by an "xxx"-mask. Results revealed (a) The typical acceleration in the recognition of the upcoming, parafoveal word: Target words were recognized around 67 ms faster when parafoveal preview was enabled than during the recognition of isolated words. (b) Parafoveal masking delayed word recognition beyond that during the recognition of isolated words. This delay was most probably caused by visual interference due to parafoveal masking. Such an interference of parafoveal masking does have serious theoretical implications: Our data suggests that parafoveal masking might result in an incorrect estimate of parafoveal preview benefits. As a consequence, the present data might challenge findings on parafoveal preprocessing in the literature as acquired by parafoveal masking.
Dimitra Ioannou (Center for Language, Reading and Writing)Margaret J. Snowling; Emma Hayiou-Thomas - Developmental links from early language and phonological skills to first grade literacy achievement in the transparent Greek orthography: evidence from typically developing and language impaired children.
Purpose: The current study sought to explore the longitudinal relations between kindergarten language status, phonological processing skills and phonological awareness and literacy outcome at the end of first grade. Method: One group of 39 typically developing (TD) children and one group of 20 language impaired (LI) children were assessed with tests of language functioning, speech processing, word and nonword repetition, naming speed, syllable and phoneme awareness and letter knowledge, over the course of two years. Tests of reading and spelling accuracy and reading speed were administered at the end of first grade. Results: In the TD group, hierarchical multiple regression analyses, controlling for age and nonverbal ability, indicated specific contributions of syllable awareness to reading speed only; phoneme awareness was a significant predictor of all literacy measures, while rapid naming independently predicted reading speed. Word and nonword repetition contributed independently to reading speed only. In the LI group, phoneme awareness was the unique predictor of both reading speed and accuracy, while word and nonword repetition skill was the unique significant predictor of spelling accuracy. Conclusions: It appears that regardless of language outcome, phonological awareness and output phonological processing skills play a crucial role in the development of early literacy skills. Thus, children with language impairments should be explicitly trained in phonological awareness skills at both the syllable and phoneme level, in order to enhance mastery of the alphabetic principle.
Shahrzad Irannejad (McGill University Health Center); Robert Savage - Performance of poor and average readers on cerebellar and phonological measures: A reading-level design
The present study investigated whether a group of children with dyslexia differed in their performance on any of reading, phonological, rapid naming, motor, and cerebellar-related measures as compared to two control groups that were matched to the group with dyslexia based on their reading level and chronological age. Participants attended mainstream English schools in Quebec. The dyslexia group included 17 children from grades three, four, and five whose single word reading performance were below a standard score of 90 (below 25th percentile). Reading-age match group included 17 normal reading children in grades one and two whose reading level were matched to the dyslexia group. Seventeen normal reading children attending grades three, four, and five were also matched to the dyslexia group on their chronological age. The groups were matched on intellectual functioning. All participants were administered IQ, word and non-word reading accuracy, reading speed, phonological awareness, rapid naming, bead threading, peg moving, toe tapping, postural stability and muscle tone measures. Parent Conners' was administered earlier. Group comparisons indicated that participants in the dyslexia group were poorer than both the RA- and CA- matched controls in their phonological awareness. There were no group differences in rapid naming or motor/cerebellar measures. Patterns of results persisted for most measures after attention was controlled. A similar pattern was found for most measures in a subsequent multiple case study analysis. Results are more consistent with a phonological core process account of both typical reading and reading difficulty but do not support accounts positing additional motor-cerebellar involvement in either typical or atypical reading acquisition.
Elena Ise (Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, Ludwigs-Maximilians-University, Munich, Germany); Gerd Schulte-Körne - Evaluation of an orthographic spelling training in German dyslexic students Grade 5-6
Purpose: Orthographic consistency influences the manifestation of dyslexia. After a few years of schooling, German-speaking children with dyslexia usually make few phoneme errors but show major difficulty in orthographic spelling. The aim of the present study is to evaluate the effectiveness of a rule-based spelling training in dyslexic children Grade 5 - 6. Method: In Experiment 1, ten dyslexic children (treatment group) received 15 individually administered once-weekly intervention sessions (60 minutes each). We expected both reading and spelling ability to improve during training. In Experiment 2, we aimed to replicate the results with a larger sample consisting of a treatment group (n=13) and a delayed treatment control group (n=14). We expected both spelling and orthographic knowledge to improve during training. Results: In Experiment 1, gains in reading were analysed using a paired samples t-test. The results showed significant improvements in reading. Gains in spelling were analyzed using an integrated data set from both experiments. A repeated-measure ANOVA showed a significant interaction, indicating that spelling improvement was greater in the treatment group than in the control group. A repeated-measure ANOVA on pre- and post-measures of orthographic knowledge (Experiment 2 data) also showed in a significant interaction, indicating that the treatment group showed greater improvements than the control group. Conclusions: The results demonstrate that difficulties with orthographic spelling can be alleviated by means of a rule-based spelling training. This finding is not only relevant for all dyslexic children learning to read and write a language with a transparent orthography.
Marion Janiot (Laboratoire URECA)Severine Casalis - Lexical Competition in Reading Acquisition: A Masked Priming Study in French Children
Developmental accounts of reading acquisition suggests that orthographic representations become progressively more and more specified (Perfetti, Ehri). The use of masked priming paradigm has been shown to be a fruitful approach for investigating how word recognition process develops in children (Castles). According to expert word recognition accounts, lexical competition is strongly associated to neighborhood of words (words sharing all but one letter). The question then arises whether lexical competition, as an indicator of highly structured lexicon, may be observed early in the construction of orthographic representation. The main objective of this study was to examine whether lexical competition may be observed in developing readers as early as grade 3 or 5. The masked priming procedure (prime duration = 57 ms) was used to explore word recognition process in children varying in age. Children (grade 3 and grade 5) had to perform a lexical decision task. 68 pairs of four and five-letter primes and targets. Primes were words orthographically related or unrelated to targets. The target was primed by the most frequent orthographic neighbour. For example, bulle ('bubble') was primed by vieux ('old') in unrelated condition and by belle ('beautiful') in related condition that is the most frequent orthographic neighbour of bulle. Priming effects differed according to grade, suggesting that orthographic representations change with reading development, increasing progressively lexical competition. These results were interpreted in terms of activation and selection processes operating in visual word recognition.
Christa Japel ()Delphine Vuattoux; Éric Dion; Monique Brodeur; Catherine Gosselin; Deborrah Simmons - Explicit vocabulary instruction to enhance at-risk preschoolers' vocabulary
Purpose: Upon school entry we observe an alarming disparity between children from different socioeconomic backgrounds with respect to the richness of their vocabulary. These findings highlight the necessity to provide more explicit vocabulary instruction for children at risk for later reading difficulties and to implement interventions during the preschool period that contain a significant vocabulary. Method: Two hundred at-risk 4-year-old preschoolers were randomly assigned to the intervention group (n = 100) or the control group (n = 100). Thirty storybooks were developed around target words representing the vocabulary gap identified between low-risk and high-risk children in the population. Each storybook introduced six new words and was accompanied by a detailed lesson plan with definitions and activities. Each story was read twice. Three of the six words were introduced at each reading followed by activities to discriminate, correctly use and generalize the target words. Children's receptive vocabulary and attention skills were assessed at the beginning of the study. Knowledge of target words was assessed at the beginning, during and at the end of the intervention. Parents provided information on sociodemographic characteristics of the family. Results: Children in the intervention group produced significantly more correct definitions of target words at the end of the intervention (p <.001). A multiple regression analysis reveals that net of family socioeconomic and child characteristics, the intervention succeeded in enhancing children's vocabulary (p <.001). Conclusion: These results suggest that early explicit vocabulary instruction is a promising intervention to reduce the gap between children from different socioeconomic backgrounds.
Annette R. Jenner (Syracuse University)Colin Poon; W. Einar Mencl; Stephen Frost; Kenneth Pugh - Neuronal activation for consistency effects in a sequential spelling task
Purpose: This study explored the pattern of cortical activation associated with a sequential spelling tasks in which words varied in phonologic-to-orthographic (P-O) and orthographic-to-phonologic (O-P) consistency. Method: Ten adult neurologically normal right-handed native English speakers participated in an event-related fMRI paradigm. During scanning the participants heard a spoken stimuli and saw a visual display consisting of placeholders for each of the letters in the stimuli. Participants made a series of yes/no decisions indicating whether each letter, that sequentially replaced the placeholders, correctly spelled the spoken stimuli. Results: The results suggest that the sequential letter-by-letter spelling task activated regions in the left hemisphere that previous investigators have suggested are important in reading and spelling including: bilateral occipital-temporal junction, left inferior frontal gyrus, and bilateral parietal lobe. A direct comparison of correctly spelled words that differed on orthographic consistency revealed more activation for consistent words in bilateral inferior frontal gyri, left superior temporal gyrus and left superior frontal gyrus. However a comparison of correctly spelled words that differ on phonological consistency revealed more activation for consistent words in bilateral occipital lobe as well as several frontal regions bilaterally. Conclusions: The comparison of the activation patterns produced when spelling words that vary on O-P and P-O consistency reveal alteration in cortical activation. This difference in neuronal pathways may reflect different cognitive strategies used to map sounds to letters when words differ in O-P and P-O consistency.
Gracia Jiménez-Fernández (University of Granada) Sylvia Defior; Francisca Serrano - Reading fluency development in Spanish normal and dyslexic readers
Reading fluency is commonly defined as reading accurately at a quick rate with appropriate prosody or expression. A lack of fluency is a common characteristic of poor readers and also a reliable predictor of poor reading comprehension. Crosslinguistic research has shown that reading fluency problems are more persistent than reading accuracy difficulties in a transparent orthography, like Spanish. This study aims to analyze reading fluency development in Spanish children in different levels of reading acquisition. Additionally, it is aimed to study how dyslexic readers struggle in this development. For those purposes, reading fluency skills were assessed in typically developed children in first, second, third and fourth grade (N=199) of primary school. The study sample was completed by 39 third grade children with dyslexia. Word and pseudoword reading tests were carried out and both accuracy and fluency measures were taken (fluent and non fluent indices). Additionally, a measure of reading speed based on time was used. The results showed that the difference between fluent and non fluent reading in normally developed children increased through grades. However, this difference was not found in dyslexic group, in particular in pseudoword reading. Reading fluency and speed difficulties were dramatically showed in dyslexic children, even when they were compared with a control group of first grade children (thereby younger than them). These finding are discussed regarding its implications with reading instruction, reading fluency assessment, and intervention.
Amber Johnston (University of Guelph); Marcia A. Barnes; Alain Desrochers - Development of bridging inferences and relation to working memory processes in school-aged children
Purpose: The goals of this study were to: 1) assess the development of bridging inferences which serve to link ideas in the text to maintain semantic coherence; 2) investigate the relation between bridging inferences and various working-memory processes; and 3) test the relation of bridging inferences to reading comprehension. Method: The Bridge-IT, a multiple-choice, group administered task, was developed to assess children's ability to make inferences between adjacent sentences (near inference) and sentences separated by text (far inference) format. Half the inferences made in each distance condition (near vs. far) required physical causal inferences and half required character motive/affective inferences. 129 children in grades 3 through 8 participated. Semantic updating and reactivation processes in working-memory as well as vocabulary, word reading skill, and reading comprehension were measured. Results: Bridging inferences increased with grade such that the oldest children (grade 8) were more accurate than the youngest children (grade 3). Across grades, near inferences were more accurate than far inferences and physical causal inferences were more accurate than affective inferences. Semantic updating predicted near inference performance while semantic updating and reactivation predicted far inference performance. Bridge-IT scores uniquely predicted reading comprehension on three standardized reading comprehension measures after controlling for vocabulary and word reading. Conclusion: Results are discussed in relation to: cognitive and developmental models of comprehension; the role of specific aspects of working memory in inferential comprehension; and implications for the development of tools to assess text-based processes important for reading comprehension.
Manon Jones (University of Edinburgh);Manon W. Jones; Holly P. Branigan; Anna Hatzidaki; Mateo Obregon - Is the 'naming' deficit in dyslexia a misnomer?
Purpose: We report a study that investigated the widely held belief that rapid automatized naming (RAN) speed deficits in developmental dyslexia reflect impaired access to lexical phonological codes. Method: To investigate this issue, we compared adult dyslexic and adult non-dyslexic readers' performance when naming and semantically categorizing 10x5 arrays of objects. Critically, for each participant, we obtained naming and categorization data for the same object items. We used eye-tracking and voice-recordings to obtain fine-grained - individual object level - information of gaze durations and production times. Results: Linear Mixed Effects (lmer) models were used to analyse the data. Dyslexic readers yielded slower production times than non-dyslexic readers when naming objects. However, a subsequent group-difference comparison on both object-naming and object-categorization tasks showed that the apparent 'naming' deficit in dyslexia could in fact be explained by the between-groups variance in categorizing the items. Conclusions: Our findings suggest that in an object-naming task, dyslexic readers are slower to make a response based on the item's lexical phonology. However, they are also slower to make a response based on the item's semantic properties. Shared variance between the two deficits suggests that they probably emerge from a single underlying problem. We suggest that although visual-phonological connections are important in explaining RAN performance these connections may not fully characterize dyslexic readers' RAN-related impairments.
Janina Kahn-Horwitz (Oranim College of Education; Israel); Mona Saba - Weak versus strong high school readers of English as an Additional Language: First language effects
Purpose: The aim of this study was to examine individual differences in English as an Additional Language (EAL) among Arabic first language (L1) high school female students from lower socioeconomic backgrounds. Method: Participants included 66, 11th grade female students from lower SES backgrounds whose L1 was Arabic. Students with a composite z score above 0.5 on reading comprehension in English comprised the strong EAL comprehension group (n = 22) and students with a composite z score less than 0.5 comprised the weak EAL comprehension group (n = 25). Students around the cut-off were excluded. The following variables were entered into a stepwise discriminant analysis: phonological awareness, spelling, and morphological awareness (judgment and production) in the L1 (Arabic), and EAL word recognition. English exposure out of school was entered as hours spent watching English T.V. per day. Results: Discriminating function analysis indicated that the strong and weak EAL groups were different from each other on a demanding aspect of L1 oral language, the morphological judgment task measured in L1 Arabic, and on he EAL word recognition task. Conclusions: The results support the central processing theory whereby L1 metalinguistic ability, specifically, morphological judgment in the L1 differentiated between strong and weak EAL readers. This finding is supported in L1 research (Koda, 2008). In addition, word recognition in English differentiated between the two groups, providing additional support to the relevance of the simple model of reading (Gough & Tunmer, 1986) whereby without efficient decoding, reading comprehension cannot take place.
Leonard Katz (Haskins Laboratories);Stephen Katz - Phonological processing in college students who are poor readers
Purpose: College students generally have average-to-above IQs and, as children, were exposed to adequate instruction in reading. Therefore, if they are poor readers, their problem is more likely to be the result of other factors. We hypothesized a phonological disability as a root cause and we examined the scope of this functioning, broadly defined. Method: Each student received 8 hours of testing, including standardized reading tests, IQ, ADHD, math, and spoken language abilities including McGurk, abstract phonological, tongue-twister, word-in-noise detection, variability in word production, rapid naming, artificial language learning, vocabulary,lexical decision, naming and others. Results: This poster presents a progress report of the results of approximately 90 participants (from an expected 150) at a point halfway through a five-year data collection period. Conclusions: The data provide a broad picture of phonological dysfunction and its relation to reading disability.
Tami Katzir (University of Haifa)Yamit Harush (NOTE, Tami Katzir's member's dues were paid on Dec 12 2009) - Expressive vs. receptive word learning from text: A comparison of three groups of readers in fourth grade
Previous research has shown that Poor Comprehenders exhibit difficulties inferring the meanings of new words from context (Cain, Oakhill & Brayent, 2004). Most of the studies however, examined expressive word learning, by asking children to define the new words. Purpose: The main purpose of this study was to examine whether Poor, Average, and Good Comprehenders differ in their expressive as well as receptive ability to learn new words from text. We also explored the contribution of vocabulary acquisition to reading comprehension, beyond that of linguistic, cognitive and decoding skills. Methods: Seventy children from fourth grade, with average decoding skills and a range of reading comprehension abilities, read twelve short stories, each containing a new, made-up word. After each story, participants were asked to recognize the meaning of the new word in a multiple-choice task, define it, and use it correctly in a sentence. Results showed that Poor Comprehenders did not differ from Average and Good Comprehenders in the recognition multiple choice task, but did significantly differ in the two expressive tasks (definition and using the word in a sentence). In addition, hierarchical regressions controlling for vocabulary and decoding showed that the ability to infer new word meanings explained independent variance in reading comprehension. Conclusions: These findings suggest that there are different levels of learning a word (Rosental & Ehri, 2008), poor comrehenders do not exhibit difficulty in receptively extracting meaning, their challenges lie in sentence formulation and abstraction. This ability is important for reading comprehension. Clinical implications will be discussed
Sibel Kaya (Kocaeli University, Turkey)Carol Connor - Comprehension Instruction in First Grade Classrooms
The current study examined types and characteristics of first grade comprehension instruction. The Individualizing Student Instruction (ISI) intervention aimed to determine types and amounts of instruction depending on children's language and literacy skills and help teachers deliver appropriate instruction. Sixty nine first grade classrooms in 17 schools in the South East United States were videotaped three times (fall, winter, spring) during 2006-2007 school year. For the purpose of this study, only literacy block which lasted approximately 120 minutes was coded. Detailed field notes were completed throughout the observation period. The classroom videos were coded using the Noldus Observer software package. Coders were trained research assistants, some of whom also conducted the video observations. In the ISI coding system, any event lasting 15 seconds or longer is captured across specific multiple dimensions: management (i.e., teacher/child managed, child self managed), grouping (i.e., whole class, small group), instruction (i.e., word decoding, comprehension), and duration of the activity. Instructional codes often had modifiers that detail the instruction and materials used (i.e., Comprehension > Questioning). Similar to previous research, results of this study showed that questioning was the most commonly used comprehension strategy during all three seasons at first grade level. Questioning was followed by previewing and predicting/inferring. Average amounts for all strategies were the highest during winter which shows most of the comprehension instruction occurred during this season. Teachers in the treatment group delivered significantly higher amounts of small group and individualized comprehension instruction compared to control teachers as intended by the ISI intervention.
Janice M. Keenan (University of Denver);Sarah J. Priebe; Amanda C. Miller;Chelsea Meenan; Anh Hua; Richard K. Olson - Speaking Up for Listening Comprehension
Purpose: A substantial portion of individual differences in reading comprehension are accounted for by differences in vocabulary knowledge (e.g., Braze et al., 2007). Because comprehension involves not just knowing concepts that are referred to in a text, but also maintaining them in working memory, making inferences to connect them, and building a mental model of the situation, we hypothesized that assessing such text processing skills through listening to discourse might explain additional variance in reading comprehension, above and beyond that of vocabulary. We examined four age groups to assess developmental differences, and examined the effects of the specific test used for listening comprehension. Method: 1064 children, ages 8 - 18, from the Colorado Learning Disabilities Research Center were given tests of word reading, reading comprehension, listening comprehension and vocabulary. Results: Across all ages, vocabulary accounted for an average 50% of the variance in composite scores from all reading comprehension tests. Listening comprehension, however, accounted for substantial additional variance - average12%. This held for all four age groups. Additional analyses show how these findings are affected by the type of discourse used in the listening task - whether listening to longer discourse accounts for additional variance beyond that accounted for by short texts such as those used in the Woodcock Johnson Oral Comprehension. Conclusion: Assessing text processing skills through listening to discourse explains substantial individual differences in reading comprehension that go beyond those captured by individual differences in vocabulary. These findings speak up for listening to discourse in comprehension assessment.
Cornelia A.T. Kegel (Leiden University, The Netherlands)Adriana G. Bus - Impulse control and oral feedback: Essential ingredients for kindergartners to learn from a computer intervention
Purpose: In this experimental study we evaluated the effects of a web-based intervention program for kindergartners. Children practice how their names look and they solve tasks as finding their name-letter among other letters and clicking on the picture that starts with the same sound as the name. The interactive version provides oral feedback cues (e.g., /p/ of peter is also /p/ of peer), which may be essential to advance alphabetic learning. Do young children benefit from computer programs and do they benefit as much when they are easily distracted and impulsive? Method: 312 four-year-old Dutch students, mainly with low social economic backgrounds, played with the target computer program or with another web-based program (control condition), in all 7 to 17 sessions of about 10 minutes. Half of the intervention group received oral feedback. In addition to pre- and post-testing of the first name-letter, phonemic sensitivity for this letter and early writing skills, we measured inhibition and working memory at pre-test. Results: After controlling for background variables both working memory and inhibition skills were significant predictors of dependent measures (β's varying between .10 and .20), as was the intervention condition with feedback (β's between .13 and .17). Furthermore, children with high inhibition scores benefited more from the intervention (with feedback). Conclusions: Practice of early alphabetic skills can be stimulated by computer programs on condition that they provide oral feedback that directs attention to letter-sound relationships. A certain degree of cognitive control, especially inhibition regulation, enhances independent learning from a computer program.
Nenagh Kemp (University of Tasmania); Lesli Hokanson - Multiple determiners of English spelling: An intervention study with university students
Purpose: Writers of English must learn to write words as they sound, but also to take into account other determiners of spelling, such as graphotactics (conventions about letter combinations) and morphology. This study investigated adults' spelling of words whose spelling is determined by several factors. Method and Results: At pre-test, 53 university students wrote familiar words with relatively transparent two-morpheme structure whose root changed spelling when an affix was added (e.g., changing, picnicked). Participants made about 20% errors, of which about half preserved the root word spelling and ignored overriding phonological (e.g., changeing, picniced) or graphotactic (e.g., merryment) constraints. Another 48 students wrote infrequent words containing sound sequences with a common Greek/Latin meaning (e.g., eu, "good", in euphemism, eugenics) and words whose matched sound sequence had a variety of sources (e.g., uvula, ubiquitous). Participants erred approximately equally on both word types (36% and 31%, respectively). After three one-hour training sessions for half of each group, the intervention subgroups' spelling of matched words improved significantly: by 8% for the transparently two-morpheme words, and by 20% for the Greek/Latin root words. Control subgroups improved on neither. Conclusions: These findings suggest that adults know that the spelling of morphological root words should often be preserved, but do not consistently obey overriding phonological or graphotactic constraints. Unsurprisingly, spellers also cannot make use of morphological roots with unfamiliar shared meanings. However, even a short intervention can enhance adults' use of these multiple influences, and thus improve their spelling of numerous English words.
Panayiota Kendeou (McGill University, Canada);Sandra Fulton - Reader individual differences during reading: Evidence from eye-tracking
Purpose In the present study we examine readers' systematic processing during reading comprehension taking into account reader individual differences in prior knowledge, epistemological beliefs, working memory, and need for cognition. Methods 48 undergraduate students participated who read two texts while their eye movements were recorded. Participants were asked to recall each text and to complete individual differences measures. Results Analysis of students' eye-fixations using Cluster analysis (Ward's method) revealed identical profiles for both texts and showed that the 3-cluster solution had the best description to the data. Processing of fast linear readers (n= 28) was characterized by the absence of fixations to previously read parts of the text. Processing of slow linear readers (n= 13) was characterized by the absence of fixations to previously read parts of the text and with significantly more and longer forward fixations than those of the other two clusters. Processing of selective reviewers (n=7) was characterized by significantly more and longer look backs to previous sentences. ANOVA examined the presence of individual differences among the three clusters. This analysis showed that readers only differed with respect to working memory capacity, F (2, 45) = 4.39, p= .018. Specifically, slow linear readers had significantly lower working memory span than fast linear readers (p= .048) and selective reviewers (p= .006). Conclusion The results demonstrate systematic differences in the ways readers process information during reading highlighting the presence of individual reading styles. The nature of the differences involves the ways in which readers reprocess information.
Sarah Kershaw (Florida State University, Florida Center for Reading Research)Chris Schatschneider - Factors that predict reading comprehension in 10th grade
Purpose The present study sought to examine the predictive ability of passage fluency, working memory, and IQ to reading comprehension after controlling for decoding and listening comprehension in a sample of 10th grade students (n = 180). IQ was measured using the vocabulary, similarities, matrix reasoning, and block design subtests from the WASI. Following this, another purpose for the current study was to examine the components of IQ individually to determine which subtests were most predictive of reading comprehension. Method The first set of analyses included a series of structural equation models to investigate the predictive ability of passage fluency, working memory, and IQ to reading comprehension after controlling for decoding and listening comprehension. The second set of analyses utilized multiple regression to investigate which subtests of IQ were most predictive of reading comprehension. Results Results revealed that passage fluency uniquely accounted for more variance in reading comprehension than decoding. Additionally, when IQ was included in the model of reading, results revealed that decoding and listening comprehension no longer significantly predicted reading comprehension. Given this, multiple regression analyses were conducted to determine which components of IQ play a role in reading comprehension. Results from these analyses showed that vocabulary, similarties, and matrix reasoning skills play a significant role in the reading comprehension ability of 10th grade students. Conclusions These results suggest that as children move from learning to read to reading to learn, some of the abilities tapped by IQ tests may play an important role in comprehending text.
Brett Kessler (Washington University in St. Louis); Tatiana Cury Pollo; Rebecca Treiman; Brian Byrne; Richard K. Olson - Predicting spelling accuracy to Grade 2 by computer analysis of erroneous spellings in kindergarten
[Purpose] Predicting children's later success in spelling from earlier accuracy rates is difficult when children still misspell everything. Measures of partial accuracy are required, the gold standard being Liberman et al.'s 5-point scale. To improve speed, accuracy, and flexibility, we developed a Web service, Ponto, to automatically score partial accuracy and help test which measures work best. [Method] Kindergartners (mean 6 years, 2 months) from Australia and the U.S. spelled 10 words to dictation. The productions of the 55 who spelled all words wrong were further analyzed. Responses were scored with a version of the Liberman measure and with several measures from Ponto. The latter differed in whether they aligned response letters to standard spellings (<tree>) or to pronunciations (/tri/), and in whether they deducted for sequencing errors (<rtee>) or for extraneous letters (<treeabc>). These measures were used to predict the number of words correctly spelled in WRAT subtests at Grades 1 and 2. [Results] All measures correlated significantly with Grade 1 accuracy. The strongest correlation (.43) was with the Ponto measure based on sound-letter plausibility, ignoring extraneous letters; sequencing was irrelevant. That measure accounted for 21% of variance in a hierarchical regression. It was the only measure still correlated (.27) with accuracy at Grade 2. [Conclusions] The fact that Ponto can be readily configured to automatically generate different scorings facilitates exploration of leading indicators of spelling achievement. For the present dataset, it already outperforms standard manual methods.
James Kim (Harvard University) - Does access to books during summer vacation increase home reading activities and improve reading ability? results from a longitudinal randomized experiment
Purpose Numerous studies indicate that low-income children fall behind in reading during summer vacation, causing the rich-poor gap in literacy to widen in the elementary grades. Voluntary reading of books and increased exposure to print play a key role in promoting reading achievement during the summer. This study was designed to examine the effects of increasing children's access to books on home reading activities and reading comprehension. Method A total of 1,028 children in a large US public school district were randomly assigned to receive 10 self-selected books for two consecutive summers. Both treatment and control students completed the Gates-MacGinitie Reading Test at four time points: the spring of Grade 4, the fall of Grade 5, the spring of Grade 5, and the fall of Grade 6. Using individual growth modeling, we estimated growth trajectories during two summers and a full school year. A survey of children's summer book reading activities was administered in the fall of Grade 6. Results There was no statistically significant difference in reading gains or at posttest between the treatment and control group. However, among low-income children, the treatment group reported owning more books, reading more books, and participating in more home literacy activities than the control group. Among middle-income children, there was no difference between the treatment and control group on measures of summer book reading. Conclusion Although the results suggest that increasing access to books can increase summer reading activities, additional scaffolding and better matching of books to readers are needed to enhance effects on reading comprehension.
Say Young Kim (Division of Psychology, Nanyang Technological University)Min Wang - The processing of derived words in Korean: a masked priming study
Purpose: The Korean orthography is an alphabetic syllabary designed to combine consonants and vowels into syllable blocks. Thus, a Korean word has a clear boundary between syllables. The present study aimed to address the question whether this syllable boundary plays a role in morphological decomposition. Method: Native Korean adults participated in one of two experiments (N=33 in each) using a masked priming lexical decision task with a short prime duration (47ms). In Experiment 1, participants were given prefixed words (e.g., 비현실, "unreal" in English), non-morphological pseudowords (e.g., 서현실 "tanreal"), and unrelated words (e.g., 몰이해 "misunderstand" in English) as the primes and the corresponding stem as the target (e.g., 현실 "real" in English). In Experiment 2, the primes were suffixed words (e.g., 현실감 "reality" in English) and non-morphological pseudowords (e.g., 현실녁 "realdal") and the target was their corresponding stem (i.e., 현실). The stems of the critical items were same in both experiments. Results: Reaction times for the targets were significantly faster when it was preceded by either prefixed words or suffixed words in comparison to the corresponding non-morphological pseudowords. There was no significant difference in the priming effects when comparing the prefixed and suffixed words. There was also no significant difference in priming effects of the non-morphological pseudowords on the targets as compared to the unrelated primes. Conclusions: These results suggest that it is the morphological structure, but not the syllable boundary that determines morphological decomposition in Korean. Furthermore, different types of affixed words are processed in a similar fashion.
John R. Kirby (Queen's University); Alain Desrochers; Glenn Thompson - The development of grammatical sensitivity in French and its relationship to phonological awareness, letter knowledge, and word reading
Purpose This study (a) investigated the development of sensitivity to grammatical violations in sentences among French-speaking children from Kindergarten to Grade 2, and (b) examined its relationship to other skills relevant to reading. The three main components of grammatical morphology in French are number agreement, gender agreement, and verb conjugation. Method Tests of spoken-sentence grammaticality judgment, phonological awareness, letter knowledge, and word / nonword oral reading were administered to 371 children from Kindergarten to Grade 2. Results (a) Children's sensitivity to grammatical violations increased significantly from K to Grade 2, as shown in conventional ANOVA on response accuracy and d' in a signal-detection analysis; (b) the bias estimate, Beta, in the signal-detection analysis decreased with grade level; (c) children's sensitivity was highest for gender-agreement violations and lowest for verb-tense agreement violations; (d) grammatical sensitivity was strongly correlated with phonological awareness, letter knowledge, and oral reading; (e) the relationship between grammatical sensitivity and oral reading was nonsignificant when letter knowledge and phonological awareness were included as predictors; and (f) letter knowledge and phonological awareness continued to predict oral reading above and beyond grammatical sensitivity. Conclusion The relationship of phonological awareness and letter knowledge to reading observed with English-speaking children is also found in French-speaking children. The absence of a significant link between morphological sensitivity and oral reading in this sample does not replicate the relationship observed with older English-speaking children. These results suggest that the latter relationship may appear at a later stage of reading development.
Ann-Christina Kjeldsen ()Katarina Witting, Åbo Academy University - Does preschool training of phonological awareness have an effect still in Grade 6?
Abstract The present study was a part of a longitudinal study which started in the Åland Islands in 1997 training phonological awareness in preschool children (Kjeldsen, Niemi & Olofsson, 2003). The intervention and the study was a replication of the study carried out on Bornholm, Denmark by Lundberg, Frost, & Petersen (RPQ 23, 1988). It is expected that training of phonological awareness promotes reading and writing skills at school. Previous results from this study reported training effects up to Grade 4 (Kjeldsen et al., in preparation). The longitudinal 10 year study ended up in Grade 9. The aim of the present study was to find out whether there was an effect of the intervention still in Grade 6. The study also reports results of rapid naming. Earlier studies have shown that children with double deficits in phonological processing as well as in the processing underlying rapid naming are poorer performers compared to children with only one of the deficits mentioned above. The training program consisted of metalinguistic games and exercises and were given in daily doses to preschool children during their last year at kindergarten and at the beginning of the first year at school.. A training dose 60% of the original was given to a smaller group. In the beginning of the study at preschool the experimental group consisted of 108 children and the comparison group of 101 children. In Grade 6 the experimental group consisted of 96 children and the comparison group of 95 children. The children at risk were defined as the children belonging to the lowest quartile in letter knowledge and by metaphonological tests in the beginning of preschool. Training effects were found on four of nine different variables and they were Sentence chains, Word comprehension, Reading comprehension and Essey writing. Corresponding effects were found for the children at risk. Both groups benefitted equally from the 100% and 60% training programmes. The children with double deficits performed worse throughout all the tests. By combining phonology and rapid naming a group of children with a double deficit could be identified. Gains of the intervention could be seen six years after the intervention. A structural, systematic and consequent preschool training of metaphonological awareness is a strong tool in the hands of preschool teachers in order to promote developement of reading and writing skills at school. References: Kjeldsen, A-C., Niemi, P., & Olofsson, Å. (2003). Training phonological awareness in kindergarten level children. Consistency is more important than quantity. Learning and Instruction, 13, 349-365. Kjeldsen, A-C., Niemi., & Olofsson, Å. (in preparation). Preschool training in phonological awareness promotes literacy skills of children-at-risk until Grade 4. Lundberg, I., Frost, J., & Petersen, O-P. (1988). Effects of an extensive program for stimulating phonological awareness in preschool children. Reading Research Quarterly, 23, 263-284.
Janette Klingner (University of Colorado at Boulder)Sharon Vaughn - Collaborative Strategic Reading with adolescent struggling readers
Purpose The purpose of this IES-funded mixed methods study is to conduct a randomized, controlled trial to test the efficacy of Collaborative Strategic Reading (CSR) for improving the reading comprehension of adolescent struggling readers and to examine contextual variables that influence teachers' CSR implementation. Method The study is taking place in low income middle schools in Colorado and Texas with five reading intervention teachers and eight Language Arts teachers and their students. Teachers are in their second year of the study. We asked them to implement CSR two to three times per week. Teachers serve as their own controls, providing students in control classrooms with "business as usual" instruction. We are applying a multi-site cluster randomized design-students were randomly assigned to classes and classes were randomly assigned to condition. We are estimating a series of hierarchical linear models to investigate the effect of CSR on comprehension using the Gates-MacGinitie (GMRT) and analyzing data from Implementation Validity Checklists, other classroom observations, interviews, and teachers' reflections. Results We are in the process of analyzing Year 1 data and still collecting Year 2 data. Preliminary analyses show statistically significant differences in favor of CSR on the GMRT (p = .05). Teachers report numerous facilitators as well as barriers to CSR implementation. Year two data demonstrate that teachers continue to feel more confident and improve their implementation of CSR. Conclusion This study adds to our knowledge about how to improve adolescent struggling readers' reading comprehension and how to help teachers teach comprehension strategies.
In Yeong Ko (University of Maryland, College Park)In Yeong Ko; Min Wang - The time course of morphological, semantic and orthographic sensitivity in visual word recognition in children and adults
Purpose Morphological decomposability has been shown to be a critical factor in early visual word recognition (Marslen-Wilson et al. 2007). The present study investigated the time-course of morphological, semantic and orthographic sensitivity in visual word recognition among grade 6 native Korean-speaking children in comparison to adults. Method A group of 28 native Korean-speaking children in grade 6 and 28 adults made visual lexical decisions on stem targets preceded by 4 different types of masked primes. The 4 conditions of prime-target pairs were: (1) orthographically overlapped only (-M-S+O: 도시락-도시, scandal-scan), (2) orthographically overlapped, morphologically decomposable but semantically unrelated (+M-S+O: 구두쇠-구두, archer-arch), (3) morphologically decomposable, semantically related, and orthographically overlapped (+M+S+O: 음악가-음악, bravely-brave), and (4) semantically related only (-M+S-O: 반대-이의, accuse-blame). 2(age: 6th vs. adults) x 4 (priming conditions: -M-S+O vs. +M-S+O vs. +M+S+O vs. -M+S-O) x 3 (SOA: 36ms vs. 48ms vs. 72ms) design was employed. Results Adults showed their morphological sensitivity when prime-target pairs were semantically related at short SOAs (36ms and 48ms) (i.e., a facilitative effect in +M+S+O condition), and the morphological priming effect was independent of the semantic factor at the longest SOA (72ms) (i.e., facilitative effects in both +M+S+O and +M-S+O conditions). 6th graders showed a facilitative effect in +M+S+O condition at the shortest SOA (36ms), and a facilitative effect in -M+S-O condition at long SOAs (48ms and 72ms). Conclusions There is a developmental shift in terms of sensitivity to the morphological information. Sensitivity to semantic information occurs earlier than the morphological information among adults. 6 graders showed sensitivity to semantic information only at the given SOAs.
Saskia Kohnen (MULTILIT Research Unit); Thushara Anandakumar - Treatment for different forms of surface dyslexia
Purpose: "Surface dyslexia" is a reading difficulty where irregular word reading is poor while rule-based (e.g., nonword) reading is normal. Surface dyslexia is a shorthand term for three possible deficits at the cognitive level: (1) poor orthographic representations; (2) poor connections between the orthographic and the phonological word-forms and (3) poor connections between an orthographic word-form and its meaning. These different impairments result in different forms of poor irregular word reading: poor recognition; poor reading aloud and poor comprehension. While these different forms of irregular word reading have been found in adults with acquired surface dyslexia and in Hebrew-reading children, we do not know at present whether these deficits dissociate in English-reading children with surface dyslexia. Given the emphasis on rule-based reading in mainstream and remedial classrooms, clinicians will encounter an increasing number of children who have poor irregular word reading. Hence, it is important that we investigate what specific form of surface dyslexia these children have and what kind of intervention may be most beneficial and efficient. Method: This study was a single-case series design. We screened participants' reading aloud, comprehension and visual lexical decision for 300 irregular words. The training study consisted of a multiple baseline design with three training phases. Training targeted each of the three different possible impairments in turn. Results: Preliminary results show that developmental surface dyslexia exists in three different forms and that each treatment responded to a different types of treatment. Conclusions: The results suggest that effective and efficient training depends on the correct diagnosis of the underlying cognitive impairment.
Ofra Korat (Bar-Ilan University)Ora Segal-Drori; Pnina S. Klein - The Effects of Reading E-books to Low SES Children on their Early Literacy Development
Purpose: We examined the effect of reading electronic books (e-books) and printed books on children's emergent literacy (EL) with and without adult support among LSES kindergarten children with high and low EL levels. Method: One hundred and twenty eight 5 to 6 years old children from low SES families were randomly assigned to one of four groups (32 children each): (1) independently reading the e-book; (2) reading the e-book with adult support; (3) reading the printed book with adult support; (4) receiving the regular kindergarten program (control). Intervention groups received four book-reading sessions each. Children's EL level was assessed before and after the intervention included letter-name recognition, letter-sound connection, phonological awareness, emergent word reading and print concept. Participants were divided into high and low initial EL levels (52 children each). Results: Results showed that reading the e-book with adult support supported both level groups; children with initial low EL level progressed in phonological awareness and emergent word reading, while children with high initial level progressed in emergent word reading. In addition, reading the printed book with adult's assistance supported the emergent word reading of high level children. Conclusions: Reading e-books with adult support to LSES children with high and low levels of literacy may enhance their emergent literacy skills. Reading printed book with adult assistance supports only the high level children.
Sebastian Korinth (University of Haifa); Olaf Dimigen; Werner Sommer; Zvia Breznitz - Reading speed enhanced? Effects of the Reading Acceleration Program on reading behavior and eye movements of young German adults
Purpose The study's aim was to test whether a training program based on Breznitz' Acceleration Phenomenon can improve silent reading speed. In particular, we examined whether the program can alter eye movement patterns and if so, in which way. Method 42 participants, most of them university students (mean age 22.4 years, mother tongue German) were tested with comprehensive reading, attention, memory and IQ tests before and after a six week reading training. Two training regimes employed text erasure as reading-pace regulators. In the accelerated reading condition (AR) erasure rate increased adaptively whereas in the continuous reading condition (CR) the erasure rate was fixed. Two reader groups completed the AR training: 17 regular readers (AR-RR) and 11 very slow readers (AR-VSR). 14 regular readers completed the CR training. Results All participants exhibited a gain in reading speed in post training tests. Eye tracking analyses identified a reduction of backward directed eye movements (regressions) as the main factor responsible for a stronger increase of reading speed in both AR groups. AR-RR and AR-VSR gained in equal measures. Conclusions Results suggest that the training program can improve silent reading speed for regular and very slow readers in equal measures. However, the comparison of AR training with CR training led to the conclusion that the essential element of the program may possibly be its pace-making function and not necessarily the acceleration. Eye tracking data suggest that reading speed gains in post training sessions were mostly caused by a reduction of regressions.
Emiko Koyama (Ontario Institute for Studies in Education);Guanglei Hong; Esther Geva - How reflective writing helps third grade children improve their reading abilities.
Purpose: This research examines whether reflective writing (writing about what they have read) enhances third grade children's vocabulary and comprehension when they are asked to use newly acquired language skills to produce ideas related to the reading. The influence of language minority (LM) status, home literacy, and English oral language proficiency is also examined. Method: The third grade data of the Early Childhood Longitudinal Study Kindergarten Class of 1998-1999 (ECLS-K) was used for investigation. The sample consisted of 14821 children. Teachers reported how often reflective writing was used in reading instruction. Children's home literacy environment was evaluated using a parental interview. Children's LM status and oral English proficiency were determined at kindergarten entry. Outcomes included total reading scores and 9 reading subscale scores (assessing different reading-related skills ranging in difficulty from letter recognition to evaluating non-fiction). Hierarchical Linear Modeling was used for data analysis. Results: Children from poor home literacy backgrounds (irrespective of LM status) were most likely to benefit from reflective writing. Reflective writing was helpful for LM children with low oral language proficiency in acquiring sight words, extrapolation skills, and comprehending expository texts. Native English speakers also benefited from reflective writing in comprehending expository texts. Conclusions: The results from this study point to the benefits of including reflective writing as part of reading instruction. This writing activity could potentially benefit disadvantaged children, especially those from poor home literacy backgrounds and those with limited initial oral English proficiency. Implications for instructional practices will be further discussed.
Melanie Kuhn (Boston University); Paula J. Schwanenflugel; Elizabeth B. Meisinger; Megan Gregory; Carolyn Groff - Reconsidering fluency theory: aligning assessment and practice
Purpose This literature review considers how definitions of reading fluency inform assessments and instruction; and how reconceptualizing fluency can help to improve comprehension as well as overall reading ability. Method Articles were obtained through a search of the ERIC and the PsychINFO databases. As articles were located, the abstract was read to confirm that the contents were appropriate. The bibliographies were also used as a means of cross-checking references to identify additional articles. Research that explored the theoretical bases of fluency development or the use and validity of fluency assessments were included for a total of 132 articles. Results Much of the focus on reading fluency has centered on a relatively narrow definition, one that emphasizes automatic word recognition. We would argue that increasing amounts of evidence not only indicate that there are multiple paths to automaticity, but also serve to support the important role of prosody in the overall reading process. It is critical that any assessment that measures fluency incorporates rate and prosody. Without such a focus, there is too great a likelihood that fluency instruction will continue to concentrate on rate and accuracy, leading to an overemphasis on fast, staccato reading as opposed to reading for understanding. Conclusions By looking at students' fluency as part of their overall reading development, instead of as a proxy for it, educators are likely to develop the kind of nuanced understanding of learners' reading ability that ensures literacy instruction assists learners in becoming truly fluent readers rather than just fast ones.
Victor Kuperman (Stanford University);Julie A. Van Dyke; - Test of reading skills predict individual differences in eye-movements among non-college-bound adolescent readers.
Purpose: This study explores the influence of individual differences in reading skills on reading behavior registered in the eye-movement record. The goals are a) to establish what domains and tests of reading abilities are predictive of eye-movement behavior in sentence reading; b) to gauge the magnitude and the fine-grained time-course of those effects; and c) to consider reading strategies and individual variability in the understudied population of non-college-bound young adults. Method: Seventy non-college-bound native speakers of English (age 16-25) completed a battery of 17 tests covering a wide range of reading-related skills (e.g., phonological awareness, naming, decoding, memory, etc.) and read a series of sentences while having their eyes tracked. We utilized linear mixed regression models to correlate individual performance in reading tests with a broad range of eye-movement measures. Results: Analyses revealed a number of novel findings: 1) Only a small subset of the considered tests was predictive of the online eye-movement behavior; 2) Individual differences in rapid automatized naming, decoding and comprehension tests elicited staggering differences across the entire time-course of reading; 3) Individual differences modulated the influence of word length and frequency on fixation times, as well as affected the quality of parafoveal preview; 4) Differences in reading strategies of non-college-bound readers as compared with earlier reports on university students were observed. Conclusions: Participant variables can have a substantial effect on eye-movements. Hence, eye-movement research would be improved by considering such variables, as well as broadening the range of skill levels being investigated.
Sarah Kuppen (Anglia Ruskin University)Sarah Kuppen & Usha Goswami - Longitudinal patterns in basic auditory processing skills, language and phonological awareness in low IQ children and controls
Purpose Outcomes presented here represent the second phase of a longitudinal project investigating phonological development in low IQ poor and good readers (PR-LIQ vs GR-LIQ). Specifically, the relationship between auditory discrimination of rhythmic timing cues and phonological awareness was investigated longitudinally. Method We re-tested 103 of 127 low IQ poor readers, low IQ good readers, reading age and chronological age controls (mean age = 9) with auditory (amplitude rise time, intensity, rhythmic timing, duration and frequency), reading (BAS, TOWRE - word & nonword) and phonological processing (Oddity, PSTM) tasks in addition to new language (BPVS,CELF, TROG, sentence correction, past tense elicitation), working memory (digit span and letter-number sequencing) and reading comprehension (WIAT II UK) measures. Results Overall, GR-LIQs performed at the level of CA controls for phonology but not language measures. Reading comprehension, picture vocabulary, recalling sentences (CELF) and formulated sentences (CELF) for GR-LIQs was at the level of PR-LIQs. GR-LIQs also demonstrated consistently lower thresholds than PR-LIQs on all auditory tasks. Fixed-step regressions controlling for group membership found Phase 1 auditory measures (with the exception of rhythmic timing) predicted significant variance in decoding (TOWRE nonword reading) for all children. Conclusion Similar to dyslexics, PR-LIQs have poor auditory processing related to poor phonological awareness and auditory ability predicts decoding development over time. Age-appropriate auditory processing for GR-LIQ suggests that low IQ or poor language are not proximal causes of this relationship.
Alexander Kurtz (Vanderbilt University) - Opportunity to Learn (OTL) in the context of special education
Purpose: In this presentation, I (a) synthesize four decades of research on OTL into a theoretical framework; (b) provide a meta-analytic review of empirical research within this framework; (c) clarify the relation between the concept of alignment, access to general curriculum, and OTL; (d) highlight the results of a published study that compared students' OTL in the context of general and special education; and (d) report on the status of an ongoing federally-funded research project that utilizes the presented OTL framework. Method: The meta-analytic review follows the structure established by Lipsey and Wilson (2000) and the highlighted study is a forthcoming peer-reviewed publication by Kurz, Elliott, Wehby, and Smithson (2009) in the Journal of Special Education. The latter study provides data on the content of instruction for 18 general and special education teachers and their alignment to state standards via the SEC alignment methodology. In addition, the relation between alignment and student achievement on three formative assessments and the corresponding state test is examined via correlation data. Results: The meta-analytic review supports the triune conceptualization of OTL and its relation to student achievement. The results of the highlighted study indicate low alignment and no significant differences between general and special education. Correlations between student achievement averages and teacher alignment indices ranged between .48 and .77. Conclusions: Documentation of OTL is feasible and needed to ensure compliance with federal laws, validity of test score inferences, and educational equity. Researchers are provided with a framework for subsequent investigations of OTL in the context of special and general education.
Fiona Kyle (City University London);Ulla Richardson; Catherine Khan; Heikki Lyytinen; Usha Goswami - The effect of reading motivation when using a computer-assisted intervention (GraphoGame-Rime) with poor readers
Purpose - This study evaluated the effectiveness of a computer-assisted reading intervention programme - GraphoGame-Rime. GraphoGame-Rime is a child-friendly, computerised reading intervention programme that trains children on letter-sound knowledge both at the level of the single phoneme-grapheme correspondence and at the larger level of the rime unit. We compared the gains in reading and spelling performance made by children who played GraphoGame-Rime with those made by a control group that played a mathematics-based computer-assisted intervention (GraphoGame-Maths) and a second control group that received no intervention. We also explored the relationship between reading motivation and intervention efficacy. Method - 64 children aged between 6 and 7 years participated in the study. The children were identified as poor readers by their class teachers. They were randomly assigned to one of three groups: intervention, treated control or non-treated control. Children in the intervention group played GraphoGame-Rime and children in the treated control group played GraphoGame-Maths. Both groups played the games daily for 10 minutes over a period of 12 weeks. All children completed a large range of cognitive, language, mathematics, literacy and motivational tasks both before and after the intervention. Results - The children who played GraphoGame-Rime made more improvement in reading, spelling and phonological awareness than those who played GraphoGame-Maths and the non-treated control group. However, there were large individual differences within the groups in terms of training effects, motivation to read and their progression through the intervention programme. Conclusions - The efficacy of GraphoGame Rime will be discussed in terms of computer-assisted intervention and reading motivation.
Katie Lam (OISE/University of Toronto); Adrian Pasquarella; Xi Chen - Examining cross-linguistic transfer of word reading, vocabulary knowledge and morphological awareness to reading comprehension in Chinese-English Language Learners
Purpose: Structural equation modeling was used to examine the relationships of word reading, vocabulary knowledge and morphological awareness on reading comprehension in Chinese and English, controlling for non-verbal intelligence and age. Method: One-hundred and twenty-five children in 1st to 4th grade completed comparable measures of word reading, vocabulary knowledge, derivational and compound morphology and reading comprehension in Chinese and English. Results: Overall, the cross-linguistic model of Chinese-English reading comprehension fit very well. Word reading measures significantly explained comprehension performance within languages only, but word reading residuals were correlated across languages. Vocabulary knowledge measures did not significantly predict reading comprehension performance in Chinese or English, but the vocabulary knowledge residuals were correlated across languages. Vocabulary knowledge measures predicted morphological awareness within but not between languages. English derivational and compound morphology measures were significantly related to English reading comprehension. No measures of Chinese morphology were significantly related to Chinese comprehension. However, the English compound morphology measure significantly predicted Chinese reading comprehension, suggesting cross-linguistic transfer. Conclusions: The results suggest similar relationships among component measures of comprehension within and between languages. Furthermore, compound morphology can be applied to reading comprehension across orthographies. These findings are explored through current theories of second-language reading acquisition and language and literacy skill development.
Karin Landerl (University of Tübingen)Franck Ramus; NeuroDys Consortium - Dyslexia in six different orthographies: Associations with phonological awareness, phonological working memory and naming speed
Purpose: Alphabetic orthographies vary with respect to grapheme-phoneme consistency and it has been argued that this factor has a significant impact on reading acquisition and the manifestation of dyslexia. More specifically, earlier studies suggest that the high relevance of phonological awareness as a predictor of literacy deficits that has been found for English may not generalize to more consistent orthographies. In contrast, naming speed was identified as a relevant predictor in consistent orthographies, where reading skills are typically assessed by fluency measures. Other studies indicate that predictive patterns are relatively universal across alphabetic languages. However, most research was carried out in only one specific orthography. The purpose of the current study was to gain further information on this controversy by investigating associative patterns of phonological awareness, phonological short term memory and naming speed on word and nonword reading and spelling in six orthographies representing different levels of phonological consistency (English, French, German, Dutch, Hungarian, Finnish). Method: 1493 eight to twelve year old dyslexic and 1302 grade level matched control children in six different languages were given tests of phoneme awareness (phoneme deletion), phonological working memory (digit span) and naming speed (digit and picture RAN). Concurrent associations of these constructs with word reading, nonword reading and spelling are analysed separately for each orthography in order to identify similarities and differences in associative patterns. Results: Data analysis is currently under way. Conclusions: Findings will be discussed in the context of reading models that consider cross-linguistic differences (orthographic depth hypothesis, grain size theory)
Kai-Yan, Dustin Lau ()Man-Tak Leung Him Cheung - Morphological awareness and Chinese reading development: an explanatory study
Purpose: The current study investigated how morphological awareness is related to reading development in Chinese through using the reading-level-match (RL), chronological-age-match (CA) design. Method: Two lexical decision tasks were administered to poor readers (n = 10; mean age = 12;03, S.D. = 5 months) with their CA controls (n = 10, mean age = 11;09, S.D. 5 months) and RL controls (n = 10; mean age = 9;06, S.D. = 4 months) in reading two-character Chinese compound words with different morphemic properties. The first task compared the effect of large vs small morphemic family size. The second task compared the effect of semantically transparent morphemes vs opague morphemes. Words with contrasting morphemic properties were in pairs that were matched for word frequencies, character frequencies and morphemic boundedness. Results: The analysis of accuracy and latency results obtained indicated that morphemic family size effects were significant in all three groups but semantic transparency effects were significant only in the CA controls and the RL controls but not in the poor readers. Conclusions: The positive results obtained in both the CA and the RL controls compared with the negative results obtained in the poor readers provide supportive evidence to the claim that morphological awareness is the causal factor of reading development. Educational implications will also be discussed.
Mark Lauterbach (CUNY Graduate Center);Mark D. Lauterbach; Alison Puliatte - Does the one-minute timing of the DIBELS phoneme segmentation fluency task introduce unintended variance into the test?
Purpose The purpose of this study is to examine whether the one-minute timing of the DIBELS Phoneme Segmentation Fluency (PSF) task introduces unintended variance into the score. Methods: A preliminary analysis of the relationship between early literacy measures revealed DIBELS PSF had anomalously low correlations with the other measures literacy (.040-.209). To investigate this 49 kindergarten and 1st grade students were administered the PSF task along with five other measures of early literacy-letter naming, invented spelling, word reading (Boder sight words and decodable words) and non-word reading. However, the PSF task administration and scoring was modified. The initial part of the administration was consistent with DIBALS protocols; however after the one minute expired students who had not completed the first 12 items were allowed to finish them. This provided two scores a timed score (range 0-75) and an untimed score (range 0-35). These two score scores were entered into regression models to predict the other literacy test scores. Results In all comparisons the untimed segmentation score predicted additional significant variance (7%-22%), whereas the timed official timed scoring accounted for no unique variance beyond the untimed score. Conclusion The timing of the task decreases the predicative quality even though the variance is higher for the timed task; clearly much of that additional variance is unrelated to other literacy skills. Whether the timing introduces problems into the administration, scoring or student response is unclear, but we put forth hypothesis how each aspect is rendered problematic by the timing of this assessment.
Joshua Lawrence (Harvard Graduate School of Education); Claire White; Catherine Snow - Improving reading comprehension through academic vocabulary instruction
Purpose Our purpose was to understand: 1) if participation in a vocabulary intervention resulted in improved word learning; 2) if improved vocabulary predicted scores on a state-mandated standardized assessment; 3) if the relationship between student vocabulary improvement and reading achievement changes as students age. Methods 1016 students in sixth (n = 398), seventh (n = 323) and eighth (n = 295) grade completed a pre and post test on 40 target vocabulary words. Students in the five treatment schools (n = 697) received vocabulary instruction for around fifteen minutes each day in different content-area classes, while students in comparison schools (n = 319) did not. Standardized test results were provided by the school district. Results Participation in the vocabulary program predicted word learning for all students (β = 0.125, p < .001). Pre to post vocabulary improvement did not predict standardized test scores for comparison students (F = 1.18, p = .308) but did for students in the intervention (F = 24.8, p < .01). Program participants who learned new words also did better on standardized measures of reading comprehension (β = 0.216, p < .001), although these effects were stronger in sixth- (β = 0.250, p < .001) and seventh-grade (β = 0.266, p < .001) students than for eighth-grade students (β = 0.220, p < .0132). Conclusions Improved vocabulary is related to better reading comprehension, and the relationship between the improved reading and comprehension are mediated by student age. Developmental and task specific interpretations are considered.
Julia Lee (Universiti Malaysia Sarawak);Stephanie Al Otaiba; Cynthia S. Puranik; Luana Gruelich; Jessica Folsom - End of kindergarten spelling performance: Are spelling errors associated with response to instruction?
Despite National Early Literacy Panel's (NELP) findings that spelling has a strong correlation with later decoding skills among kindergarteners (2008), limited studies investigate spelling as an outcome. This poster compares the end of kindergarten spelling performance of good and poor responders to Individualized Student Instruction (ISI-K; Connor et al., 2007; Al Otaiba, et al., 2009). Students (n = 566) attended 10 schools (n = 32 classrooms) serving a diverse socioeconomic population in a mid-sized southern city. Students' mean age in fall was 5.22 (SD = .37). About 55% of students were male. The research design was quasi-experimental; ISI-K was provided to two groups of teachers: half had received ISI-K the prior year and half had been wait-list-controls. Language and literacy assessments were administered in fall, winter, and spring of kindergarten to determine students' response to ISI-K. At the end of kindergarten, we administered a spelling task comprising 14 words (Bryne & Fielding-Barnsley, 1993). The students' spelling scores (ranging from 0 to 13; M = 4.55; SD = 3.36) had moderate correlations to reading performance (ranging .30 to .60). Detailed spelling error analysis (Tangel & Blachman, 1992) (e.g., initial, medial, and final phonemes; initial and final blends) is in progress. On average, students demonstrated good reading gains, 6-10% were poor responders (high risk on DIBELS letter naming fluency or word reading standard scores below 85). Based upon our preliminary findings, we anticipate finding weaker spelling scores and more errors among poor responders. Implications and directions for future research will be discussed.
Christopher Lemons (University of Pittsburgh) - An observational study of reading instruction for students with intellectual disabilities
Purpose - We know very little about how teachers teach reading to students with significant cognitive disabilities. This study addresses this knowledge gap. The purpose of this study was to provide a description of the reading instruction provided to children with intellectual disabilities who take the state's alternate academic assessment. Method - Observations are currently being conducted of the reading instruction provided to children with intellectual disabilities in grades 5 and 8 (approximately 20 children at each grade level). Participants are participating in one in-person observation and two videotaped observations. The data are being coded using the Instructional Content Emphasis (ICE) observation tool (Edmonds & Briggs, 2003). Additionally, teacher interviews and student assessment are being conducted. Results - Data will be analyzed to describe the type of instruction the child is receiving, the skills focused on (e.g., phonics, fluency, phonological awareness) in instruction, the match between student reading level and instruction, the engagement of the child, and overall instruction quality. Additionally, teacher's opinions on the challenges of aligning reading instruction for children with intellectual disabilities to academic content standards will be discussed. Conclusions - Implications for reading instruction and for teacher training will be discussed.
Paavo H.T. Leppänen (Finnish Center of Excellence in Learning and Motivation Research, Department of Psychology, University of Jyväskylä)Kaisa Lohvansuu; Jürgen Bartling; Jennifer Bruder; Ferenc Honbolygó; Jarmo Hämäläinen; Stephanie Iannuzzi; Rodolphe Nenert; Nina Neuhoff; Silke Streiftau; Annika Tanskanen; Dénes Tóth; Jean-Francois Demonet; Gerd Schulte-Körne; Valéria Csépe - Cross-linguistic study of brain responses to vowel differences in children with dyslexia in four European countries
Purpose: The role of speech perception problems in relation to dyslexia has been debated for several decades. Here we studied whether dyslexic children would show atypical speech processing for vowels representing native and non-native exemplars presented in a cross-linguistic mismatch negativity (MMN) paradigm in four different European countries and orthographies. Method: The same paradigm was used in all countries. Good representatives of /y/-vowels in three languages as rated by native listeners, Finnish/Hungarian, French, and German, were presented as the deviant stimuli (18 % each) in separate blocks among a common standard European /i/ stimulus (82%). Brain event-related potentials (ERPs) were measured from altogether 211 dyslexic and 198 typically reading school-aged children at 2-4 grades. Results: The dyslexic readers showed atypical processing of deviant /y/s as compared to standard /i/s across all four countries. This difference was found both at the typical N1 latency related to basic processing of speech sounds and at MMN latency reflecting change detection of contrasting speech sounds. For MMN and N250, group differences also varied between countries. Conclusions: Dyslexic readers show atypical vowel processing. Findings also suggest language specific modulation of top-down processing affected by long term speech sound representations. Speech perception problems may thus underlie, at least in a sub-group of dyslexics, underspecified phonetic representations.
Matt Lerner (Florida State University);Christopher J. Lonigan; Nicholas Allan - Contributions of internalizing, externalizing, and social behaviors to the development of literacy-related skills from preschool to kindergarten
Purpose: Considerable evidence suggests a correlation between behavior and literacy growth. This study sought to determine which domains of behavior contribute to growth in literacy-related skills as children transition from preschool to elementary school. Although studies of school-age children have demonstrated stronger links between later literacy skills and behaviors associated with attention problems than social behaviors, few studies have examined links between literacy and other behavioral domains, such as internalizing and externalizing problems. Method: At the end of preschool, children (N = 221; M age = 4.88 years, SD = 0.53; 42% girls) completed phonological awareness (PA) measures (blending and elision), and teacher ratings of children's behaviors were obtained using the Social Competence and Behavior Evaluation-Short Form, which provides ratings of children's internalizing, externalizing, and social behaviors. At the end of kindergarten, measures of PA and reading were obtained. Results: Preliminary analyses indicated significant relations between preschool behavior and kindergarten literacy skills. Partial correlations controlling for preschool phonological awareness skills revealed that prosocial, internalizing, and externalizing behaviors each added uniquely to the prediction of later literacy skills. Simultaneous regressions using measures of all three behavioral domains, however, indicated that only internalizing behaviors uniquely predicted later literacy skills. Conclusions: Previous studies of older children have found concurrent and longitudinal connections between children's behaviors and their reading skills. Whereas previous evidence has highlighted the relative importance of behaviors associated with attention problems, the present findings suggest a connection between preschool children's internalizing behaviors and later literacy-related abilities.
Arne Lervag (University of Oslo) - The structure and development of beginning reading comprehension skills.
Purpose: The current study examined the foundation and development of beginning reading comprehension skills. According to the simple view of reading, reading comprehension is the product of decoding and oral language skills. This study seeks to investigate the foundations of the oral language part and how this part, together with decoding, is able to predict the development of reading comprehension skills. Method: One hundred and ninety-eight Norwegian speaking children were measured four times from second through third grade. Latent-variable modeling was used in order to: 1. test the structure of the linguistic (listening comprehension, vocabulary breadth, vocabulary definition, vocabulary inference skills, grammar), cognitive (working memory, non-verbal abilities) and decoding skills in second grade, and 2. see how these constructs were able to predict the further development of reading comprehension skills. Results: The linguistic measures were best seen as one common factor that was almost isomorphic with listening comprehension. These linguistic skills explained (both concurrent and longitudinally), together with decoding skills (concurrent and to some degree longitudinally), a substantial part of reading comprehension. Working memory and non-verbal abilities were not able to explain any unique variance in reading comprehension skills. Conclusion: The study supports the simple view by confirming that decoding and oral language skills are important components of reading comprehension. Further, it extends the simple view by clarifying the foundations of the oral language component and how these linguistic skills, together with decoding, are able to explain the development of early reading comprehension skills.
Nonie K. Lesaux (Harvard University, US); Jeannette Mancilla-Martinez - Sources of Reading Comprehension Difficulties among Adolescent Spanish-speaking Language Minority Learners
Purpose: This study investigates the development of English reading comprehension and sources of reading difficulties for a sample of approx. 200 Spanish-speaking language minority (LM) learners in the U.S., followed from age 4 to age 13. The study advances research in this area by generating model-based reader profiles to understand sources of comprehension breakdown and thus inform effective instruction to meet LM learners' needs. Method: The current dataset includes seven waves of data on 178 Spanish-speaking LM learners collected annually from age 4 through age 8 and from age 11 to 12. The eighth wave will be collected from January to March 2010. Given the multi-faceted nature of reading comprehension, the study employs standardized measures tapping many constructs: word-level reading, oral language skills, metacognitive skills, reader self-perception, reading comprehension, and self-esteem. Results: Findings to date demonstrate: 1) in both languages, word-level skills develop within the normal range whereas oral language skills are well below average at each wave and 2) home literacy practices are stable over time, while English use increases. Structural equation modeling demonstrated that word reading exerted a greater influence on comprehension outcomes, compared to vocabulary, likely because of students' low levels. Next, we will generate model-based reader profiles using latent class analysis to investigate heterogeneity of reading difficulties for this population. Conclusion: This study contributes to a small research base to inform theory in this domain and in turn, data on the sources of difficulty will inform our understanding of effective instruction to meet LM learners' needs.
Nonie Lesaux (Harvard University); Michael J. Kieffer; Joan G. Kelley; Andrea Anushko - Academic vocabulary instruction for early adolescents from linguistically diverse backgrounds: findings from a large-scale experiment
Purpose: This study aims to evaluate the effects of an academic vocabulary program designed to promote the reading comprehension of learners in mainstream urban sixth-grade classrooms with high proportions of language minority learners. Method: The study was conducted in 14 middle schools in a large, urban district in the U.S. The intervention was randomly assigned at the classroom level (n=51), with 26 treatment teachers (886 students) and 25 control teachers (960 students). Seventy percent of students were language minority learners. At pre-test, the sample mean for reading comprehension was the 36th percentile. Pre-testing was conducted in fall 2008, followed by implementation of the 20-week program (45 min/day), and post-testing in spring 2009. A delayed post-test is scheduled for winter 2009. Key outcomes include aspects of vocabulary knowledge (depth of vocabulary, mastery of words taught, morphological skills) and reading comprehension, assessed using standardized and researcher-designed measures. Results: Preliminary results from multi-level modeling suggest the program had a positive, significant impact on several measures of vocabulary (words taught, depth of vocabulary knowledge, morphological awareness) and on a researcher-developed measure of reading comprehension (with effect sizes ranging from .20 to .60), but not on the standardized comprehension measure. Further analyses are planned to analyze variation in treatment effects as a function of student (e.g., achievement, language status) and classroom variables (e.g., implementation, classroom achievement level, and concentration of language minority learners). Conclusions: The present study seeks to advance research on vocabulary instruction for older readers who need intervention to promote their reading comprehension skills, particularly language minority learners.
Paul P.M. Leseman (Utrecht University, Department of Education, Utrecht, The Netherlands)Johannes E.H. van Luit; Marielle H. Messer - Verbal and visuo-spatial working memory predicting vocabulary and discourse comprehension in monolingual native Dutch and bilingual Turkish-Dutch 6-year-olds
Purpose: Vocabulary and oral discourse comprehension are among the strongest predictors of (later) reading comprehension. The present paper examined the role of working memory in the development of vocabulary and narrative discourse comprehension in pre-reading monolingual Dutch and bilingual Turkish-Dutch 6-year-old children. Based on Kintsch's model of discourse comprehension, we expected both verbal and visuo-spatial working memory to be involved in discourse understanding in all children. Given the delays in Dutch as second language, we expected visuo-spatial memory to be a stronger predictor for the Turkish-Dutch children, pointing to a compensatory route to circumvent lacunas in language knowledge. Method: 68 monolingual native Dutch and 70 bilingual Turkish-Dutch children were followed over three waves of measurements from age 4 till age 6. In addition to Dutch vocabulary and narrative discourse comprehension, we assessed children's verbal and visuo-spatial working memory. Results: Turkish-Dutch 6-year-olds were considerably behind their Dutch peers in Dutch vocabulary, Dutch grammar, and Dutch discourse comprehension, and also in Dutch verbal working memory, but there were no statistically significant differences in visuo-spatial working memory. Using Latent Growth Modeling, development of children's Dutch vocabulary and discourse skills was predicted by the intercept (mean level) and slope (growth) of verbal and visuo-spatial working memory in both groups. Moreover, the effects of level and growth of visuo-spatial working memory on Dutch vocabulary and discourse skills were significantly stronger in the Turkish-Dutch group. Conclusion: results confirm the text comprehension model and indicate that visuo-spatial working memory can compensate for less-well developed second language skills.
Paul Leseman (Utrecht University)Rian Aarts; Lotte Henrichs - Academic language input and early literacy in native Dutch and Dutch-as-second-language learning Moroccan-Dutch and Turkish-Dutch children
Purpose: The persistent disadvantages in reading of low income children and immigrant children with a minority language are a matter of concern. We propose that the core problem is insufficient experience with the 'academic language register', a register that contains many features of written language, such as specialized vocabularies, rare verb tenses, explicit space-time references, syntactic means to condense information, and strategies to structure discourse. This paper examines the impact of emergent academic language on emergent literacy and first grade reading achievement. Method: 162 Dutch, Moroccan-Dutch and Turkish-Dutch children were followed from 3-7 years. Receptive and productive use of academic language in first and second language was evaluated, using the Systemic Functional Linguistics framework of Schleppegrell and Halliday. Measurements also included receptive vocabulary, emergent literacy, first grade reading achievement, and language input at home and in preschool. Subsamples of each group were involved in observation studies of shared reading and talking in these contexts. Results: Children become increasingly skilled in using academic language. Specific language input mediates effects of family background. The observation studies reveal strong input-outcome contingencies regarding features of academic language. Compared to parents, preschool teachers initiate more academic discourse, but in addressing Dutch-as-second-language-learning children, they simplify the lexical and grammatical forms. For these children, moreover, there is a competition between first and second language for scarce input time, but also transfer of skills from first to second language. Conclusion: Emergent academic language predicts first grade reading. Input matters. Academic language skill in L1 transfers to L2.
Olivia Leung (); Ronald Stringer - The continued search for a double dissociation: relationships between rapid automatized naming (RAN), reading, and attention
Purpose: Serial rapid naming of letters and digits (alphanumeric RAN) is consistently and reliably associated with reading ability, yet converging evidence suggests rapid naming of colours and objects (non-alphanumeric RAN) is related more to attention, especially executive function and behaviour ratings linked to ADHD. The present study aimed to provide further support for this double dissociation with an adult population, by examining the correlations between serial RAN performance, attention/behaviour ratings, and reading ability. It was predicted that in older readers, alphanumeric RAN performance would correlate higher with reading measures than attention measures, whereas non-alphanumeric would correlate higher with attention measures than reading. Furthermore, differential performance on two levels of serial colour RAN (saturated and unsaturated) may provide further clues to the double dissociation between RAN task type and attention or reading measures. Method: Participants were 55 adults. Measures were selected for serial RAN (of letters, digits, saturated and unsaturated colours, and objects), list reading fluency of words and non-words, and attention/hyperactivity (Conners Adult ADHD Rating Scales). Results: A clear pattern of correlations were found related to RAN task stimuli type and naming speed. Participants had faster average performance on letters/digits versus colours versus objects. Generally, the pattern of correlations was as predicted. Conclusion: Results of the study tend to support the hypothesis that exposure to overpracticed stimuli such as letters and digits is related to reading speed in adults, whereas exposure to less common, more complex stimuli (colours and objects) is consistently slower, and more significantly related to measures of attention.
Iris Levin (Tel Aviv University, Israel) - Aleph, Bet, Gimel ...: Long Hebrew letter names bridge the gap between oral and written words
Purpose - Letter knowledge has been found related to early literacy skills like inventing spelling or learning to recognize printed words. English-speaking preschoolers spell letters correctly more often when their names are heard in the word (e.g., B in beach vs. bone). In Hebrew, letter names have been claimed to be less useful, because they are multi-phonemic, of 3-5 phonemes, and consequently words rarely include the full sounds of a letter name. The goals of the study were to examine if (1) Hebrew letter names are useful in spelling words orally and in writing, and if (2) their usefulness increases with the length of cue - full letter names helping more than partial letter names, which help more than phonemic cues. Method - Israeli kindergartners were asked to report and spell initial and final letters in words, including full (CVC), partial (CV), and phonemic (C) cues derived from these letter names (e.g., kaftor, kartis, kibel, spelled with Kaf). Results - (1) Correct responses increased with length of cues. (2) When the words included the first 3-1 phonemes of the letter's name the effect was stronger for monosyllabic than disyllabic letters' names (e.g., /kaf/ and /samex/), probably because the cue covered a larger ratio of the letter name. Conclusions - Letter knowledge is important for enhancing literacy among Hebrew speaking preschoolers. The help provided by Hebrew letter names appears stronger than in English, probably because the former names are acrophonic. The significance of letter knowledge should be studied in alphabets with different features.
M.Chiara Levorato (University of Padua)Chiara Levorato; Maja Roch; Elena Florit - The Simple View of Reading in Down syndrome: Concurrent and longitudinal predictors
Purpose. Two studies were carried out with children and adolescents with Down syndrome (DS) in order to test the validity of the 'Simple View of Reading' (Hoover and Gough, 1990), according to which individual differences in reading comprehension are accounted for by reading skills and listening comprehension. Method and Results Study 1. In Study 1, twenty-three individuals with DS and twenty-three typically developing children were compared as to their reading skills, listening and reading comprehension. Individuals with DS showed an atypical profile, similar to that of 'poor comprehenders' (Cain and Oakhill, 2006): they performed well in word reading and poorly in both reading and listening comprehension. Method and Results Study 2. Study 2 tested the hypothesis that poor listening comprehension might explain the scarce performance in reading comprehension and therefore be a cause of the asynchrony between reading skills and reading comprehension in DS. Ten individuals with DS were assessed in two points in time as to their reading skills, listening and reading comprehension: reading skills played a marginal role on the development of reading comprehension which was, instead, predicted by listening comprehension. Conclusions. The results of the two studies are discussed according to the assumptions of the "Simple View of Reading".
Chen-Huei Liao (National Taichung University); Bor-Chen Kuo; George Georgiou - Task-focused versus Task-avoidance Behavior and Chinese Literacy Development
Studies have shown that phonological awareness skills and rapid naming skills correlate and predict Chinese reading acquisition across grade levels (Liao, Georgiou, & Parrila, 2008; McBride-Chang, Shu, Zhou, Wat, & Wagner, 2003). However, reading-related cognitive skills did not account for all the variance in Chinese literacy development. The purpose of the study was to examine the contribution of the non-cognitive factor, children's classroom activity, defined as task-focused versus task-avoidance behaviour, on various literacy measures in Chinese. Eighty grade 2 children and One-hundred and three grade 4 children were assessed by a Task-focused questionnaire (filled out by homeroom teachers), literacy measures (character recognition, dictation, word reading fluency, text reading fluency), phonological awareness skills, and rapid naming skills. An initial analysis has shown that (a) task-focused behaviour correlated significantly with all literacy measures in grade 2 but not in grade 4, and (b) task-focused behaviour predicted unique variance in character recognition, text reading fluency and dictation in grade 2 even after the effects of age, phonological awareness, and rapid naming were accounted for, however, the same was not found in grade 4. The results suggest the importance of non-cognitive factor for Chinese beginning readers.
Candise Lin (University of Maryland, College Park)Min Wang; Hua Shu; Han Wu; Chu Chu Li - Development of tone sensitivity in young Chinese children and its relation to reading
Purpose - This study examined the development of tone sensitivity in Chinese children and its relation to reading acquisition. There are three types of tones in Mandarin. The four lexical tones (LT) attach to the same syllable but carry different semantic information (/ma/1 "mother", /ma/2 "linen"). Tone sandhi (TS) is tonal alteration in connected speech in which Tone 3 becomes Tone 2 when followed by another Tone 3 (/hen/3 "very" + /hao/3 "good" = /hen/2/hao/3 "very good"). The neutral tone (NT) is tonal alternation on the second syllable of a disyllabic word and pronounced with lower pitch and shorter duration (/xue/2 "study" + /guo/4 "-ed" = /xue/2/guo/ "studied"). Method - Participants were 19 3-year-old, 43 5-year-old, and 40 7-year-old native speakers of Beijing Mandarin. Measures included an expressive vocabulary test, onset, rime, and LT oddity, and TS and NT awareness tasks. Reading assessments, administered to only 7-year-olds, included single character recognition, two-character word recognition, and sentence comprehension. Results - Three-year-olds performed above chance only in TS. Five and 7-year-olds performed above chance in all phonology tasks. Vocabulary and onset and rime (composite z-scores) predicted character reading; onset and rime and LT predicted word recognition; onset and rime, LT, and TS and NT (composite) predicted sentence comprehension. Conclusion - Three-year-olds have developed sensitivity to TS. Awareness to onset, rime, LT, and NT emerges at age 5. Early development of TS may be due to phonotactic properties of the Tone 3 sandhi. NT involves tense marking auxiliaries which may require morphosyntactic knowledge that develops later. Onset, rime and tone awareness contribute to reading achievements in 7-year-olds.
Su-Jan Lin (professor of National Kaohsiung Normal University, Taiwan)Shih-Jay Tzeng;Yu-Jen Chang;Su-Li Chen - The Longitudinal Study of Reading Fluency Development with Six grades in Taiwan
The purpose of this study was to examine the development of oral reading fluency in Chinese from the first to the sixth grade in Taiwan. This study had been conducted for two years. The curriculum-based oral reading measurement (CBM) was used to test participants' reading fluency. The participants were 1,578 elementary students from Taiwan. These participants included the first, third and fifth graders in the first year study. They continued to participant this study when they became the second, forth and the sixth graders in the second year. There were four times measurement during one school year such as fall, winter, spring and summer time. The assessment time was one minute except 20 seconds in the winter time for the first graders. This study recorded participants' correct number when they read aloud Chinese characters. Results showed the growing growth from the first grade with 96 characters to the spring of the third grade with 193 characters correctly. The above 200 characters correctly started appearance from the summer of the third grade. This score continually stayed until the spring time of the sixth grade. The last summer test on the sixth grade presented another little progress of reading fluency with 221 Chinese characters. This longitudinal research indicated that end of the third grade, eight to nine years old, Taiwan's students had developed their confirmation and fluency ability of Chinese decoding. This finding is similar to the English reading development stage published by the J. S. Chall in the 1983 and 1996.
Pei-Ying Lin (University of Saskatchewan)Yu-Cheng Lin - The development of visual-spatial orthographic skill in Chinese children
*Purpose - The purpose of the present study is to examine Chinese children's development of visual-spatial orthographic skill. *Method - A total of 173 children in Grade 2-5 were recruited in Taiwan (n = 33, 43, 45, 52, respectively). They received a computer-based Chinese character structure task, which reflects the seven Chinese character structures (Horizontal structure type 1 and 2, Vertical structure type 1 and 2, P-shape, L-shape, and Enclosure) and three levels of stroke density (high, medium, low levels). The accuracy and response time of each item were recorded by E-Prime 2.0. *Results - We found that characters with Horizontal type 2 and Vertical type 2 structures which consist of complex components and strokes are the easiest for 4th and 5th graders whereas 3rd graders respond to Vertical type 2 structure most accurately and 2nd graders are most skilled at Enclosure structure. Moreover, characters with Vertical type 1 structure which is made of simple components and strokes are relatively difficult to differentiate for all participants. Children in grade 5 outperform the rest of children in term of accuracy; 2nd graders, on the other hand, respond significantly slower than other children in term of response time. *Conclusions - Skilled Chinese readers are highly responsive to complex visual-orthographic structures which have more visual cues embedded in characters such as Horizontal type 2 and Vertical type 2. Furthermore, older children develop faster visual-orthographic processing speed in grade 3-5.
Miriam Lipsky (University of Miami) - The relationship of non-verbal pattern recognition to morphological awareness
Purpose Structural equation modeling was used to examine the relationship between pattern recognition ability and morphological awareness (MA), after controlling for vocabulary knowledge, reading comprehension, and decoding. Method Participants were 282 children who were part of a longitudinal study on reading development of Spanish-speaking English language learners (ELLs) from grades 2 - 5. Verbal skills of participants were assessed using sub-tests of the Woodcock Language Proficiency Battery -Revised (WLPB), and pattern recognition was assessed using Raven's Coloured Progressive Matrices (RCPM). MA was measured by a set of four tasks in which participants were required to provide the base word of a given derived word. A Multiple Indicator Multiple Causes (MIMIC) model was fit to the data, where the MA latent variable was the outcome variable. In the structural part of this model, the outcome of MA was determined by the scores on four observed variables: RCPM, WLPB Word Attack test (decoding), WLPB Reading Vocabulary test, and WLPB Passage Comprehension test (reading comprehension). Results Non-verbal pattern recognition ability, measured by RCPM, provides a unique contribution to MA over what would be expected from verbal abilities (decoding, reading comprehension, and reading vocabulary) alone. Conclusions In our continuing efforts to understand the complex process of reading, it is clear that both linguistic and non-linguistic factors should be considered. Although verbal skills have a strong relationship to MA, this study shows that non-verbal pattern recognition skills may also play a part in this metalinguistic ability, and this relationship merits further investigation.
Duo Liu (The Hong Kong Institute of Education)Catherine McBride-Chang - Morphological sensitivity and its associations with vocabulary acquisition in Chinese children
Purpose: Very few studies have explored children's implicit morphological processing during lexical access (morphological sensitivity), compared to the exploration of morphological awareness. In the present study, third grade Chinese children's morphological structure sensitivity and its relationship with morphological awareness and Chinese vocabulary knowledge were explored. Method: Using visual priming lexical decision paradigm, morphological structure sensitivity was tested. There were four priming conditions: 1) primes and targets had related meanings and the same compounding structures (RMSS); 2) primes and targets had related meanings but different structures (RMDS); 3) primes and targets had unrelated meanings but the same structures (UMSS); and 4) primes and targets had unrelated meanings and different compounding structures (UMDS). Two types of Chinese two-character compound words were involved - subordinate and coordinative words. Results: A significant positive priming effect was found in the UMSS condition for subordinate but not coordinative compounds, besides of the semantic priming effects on both RMSS and RMDS conditions. Then, in the regression analyses, after controlling for children's age, non-verbal IQ, and phonological awareness, a full-mediation regression model was found for the relationship between morphological sensitivity and children's vocabulary knowledge, which was mediated by morphological awareness. Conclusions: The findings indicated that third grade Chinese children have developed the morphological connections in their mental lexicon (even though the semantic ones are still predominant), which directly facilitate the development of morphological awareness, and indirectly help them build up vocabulary.
Simon P. Liversedge (University of Southampton);Julie A. Kirkby; Hazel I. Blythe; Denis Drieghe; - Reading causes increased binocular disparity in dyslexic children.
Purpose: The role of visual deficits in dyslexia remains controversial. We examined the possibility that reading difficulties associated with dyslexia are linked to poor binocular coordination. Methods: We carried out two sets of experimentation testing dyslexic children, typically developing children and adults to examine binocular eye movements in non-reading tasks and in a reading task. In the non-reading tasks participants' eyes were tracked as they scanned dot strings, made successive saccades between two lateralised dots, and make target onset elicited saccades to four retinally eccentric locations. Results: While there were differences between adults and children in these tasks, performance in both groups of children was the same. In the reading task participants silently read single sentences as their eyes were tracked. Here we found a significantly greater magnitude of fixation disparity for dyslexic children compared to typically developing children and adults during reading. Conclusion: The existence of a language specific, binocular disparity deficit in relation to dyslexia is a novel finding, and our data indicate a causative relation between reading and increased binocular disparity.
Jessica Logan (Ohio State University)Steven Petrill - Examining differences in genetic and environmental effects across levels of reading
Purpose: Studies suggest that both genetic and environmental influences are important to reading development. However, is unknown whether genetic and environmental influences vary as a function of reading ability. This issue has been examined in the context of general cognitive ability, but has not been systematically studied in reading. One might hypothesize that shared environmental influences may vary as a function of children's ability as they come into increasing contact with direct instruction. Method: The present study examined the reading-related skills of 283 pairs of monozygotic (MZ) and dizygotic (DZ) twins drawn from the Western Reserve Reading Project. Measures of decoding, phonological awareness, fluency, and reading comprehension were analyzed. Twins were initially assessed at school entry (6.1 years) and were assessed annually in their homes four times. Results: We examined differences in the correlation between twins at multiple points in the distribution of reading using quantile regression. Results varied as a function of the type of reading skill and the age of the twin. For example, in younger twins, decoding skills showed large shared environment estimates in the higher and lower tails, whereas phonological awareness showed increasing genetic effects at the extremes of the distribution. In subsequent assessments, these differences attenuated and the genetic and environmental effects became more stable across the distribution. Conclusions: Results suggested that broad generalizations concerning the importance of genetic and environmental effects may mask complex gene-environment effects that vary as a function of age, as well as type and level of reading skill.
Jessica Logan (Ohio State University, Human Development and Family Science);Chris Schatschneider; Ralph Radach - Visuo-motor control and the use of parafoveal information in rapid letter naming
Purpose. This study examined the role of eye-movement control and parafoveal letter-level processing in standard rapid serial letter naming, one of the major tasks routinely used as a context - and comprehension-free assessment of reading fluency. Method. Based on eye-movement contingent display manipulations, the role of parafoveal information was examined by varying the amount of information available to the right of fixation. Eye movement data were collected for forty-one first-grade students and related to reading outcomes at three levels of processing: low-level visual processing, initial lexical access, and higher-order cognition. Results. Results suggested that visual processing occurs in a two-letter processing window; information was useful to participants for both the letter directly fixated and the parafoveal letter immediately following it. For cognitive outcomes, children who were better at rapid naming were differentially impaired by the lack of parafoveal preview than the children poorest at rapid naming. In addition, variability in performance was significantly related to all reading outcomes. Conclusions: The results suggest that the fluency of letter reading was enhanced by the availability of parafoveal information, particularly for good readers. Further, the strong relation of eye-movement control and consistency to reading outcomes suggests that visuo-motor skill may be an important contributor to reading fluency in first grade.
Christopher J. Lonigan (Florida Center for Reading Research, Florida State University);Beth M. Phillips - Preschool precursors to reading success: Joint and unique contributions to early word-reading accuracy and fluency
Purpose. Although there is increasing recognition of the importance of the preschool period for children's ultimate reading success, to date there has been only limited examination of how preliteracy skills impact the development of reading fluency. Method. In this study, two cohorts of 4-year-old children (combined N = 625) were followed longitudinally from preschool to the end of first grade. All children completed a battery of preliteracy assessments in preschool that included measures of cognitive abilities, print knowledge, phonological awareness, phonological memory, rapid naming, and oral language. In kindergarten and first grade, these children completed assessments of reading accuracy and fluency of words in isolation, and in first grade, fluency of reading connected text was assessed. Results. Preliminary analyses examined the joint and unique contributions of skills measured in preschool to measures of word- and non-word-reading accuracy and fluency in kindergarten and first grade. Across outcomes, preschool variables accounted for between 30 and 40 percent of the variance. Children's print knowledge, phonological awareness, and rapid naming skills account for unique variance in all outcomes. Additionally, children's phonological memory (nonword repetition) contributed unique variance to scores on the fluency measures. After controlling for concurrent measures of word-reading accuracy, children's print knowledge and phonological memory continued to contribute unique variance to fluency. Conclusions. Findings from this longitudinal study indicate that the foundations for both accurate and fluent word-reading skills are being built prior to the onset of formal instruction in reading. A similar set of skills contributes to word-reading accuracy and fluency.
Christopher J. Lonigan (Florida Center for Reading Research and Department of Psychology, Florida State University) - This spoken paper will be a discussant.
This spoken paper will be a discussant.
Susan Loveall (University of Alabama); Marie Moore; Frances Conners; Laura Hume; Dale Maddox - An individual difference analysis of the self teaching hypothesis
Purpose - The self-teaching hypothesis suggests that children learn orthographic structure of words through the process of phonological decoding (or sounding out) new words. Extending this hypothesis to individual differences, more successful decoders should acquire more orthographic knowledge about words, and become even better readers. We tested this idea by examining whether the relation between phonological decoding and word reading is partially or fully mediated by orthographic knowledge. Method - Forty 2nd and 3rd grade children completed tasks measuring IQ (KBIT-2), word identification (WRMT-R Word Identification), phonological decoding ability (WRMT-R Word Attack), and orthographic knowledge (orthographic choice, homophone choice, and orthographic awareness). Results - Hierarchical regression showed that after controlling for IQ, phonological decoding accounted for 27% of the variance in word identification, yet this dropped to 15%-18% when orthographic knowledge was entered into the equation first. Despite this drop, phonological decoding still accounted for a significant portion of variance in word identification, suggesting partial mediation. The bootstrapping method of multiple mediation (Preacher & Hayes, 2008) indicated that both word-specific orthographic knowledge (point estimate = .15, 95% CI = .04 to .39) and general orthographic knowledge (point estimate = .10, 95% CI = .01 to .24) independently mediated the relation between phonological decoding and word identification. Conclusions - The results support an extension of the self-teaching hypothesis to individual differences. Further, they suggest that phonological decoding may influence general as well as word-specific orthographic knowledge, which in turn have separate effects on word reading. Experimental research would be needed, however, to test this idea.
Catherine Lowe (University of Essex); Jackie Masterson; Yvonne Griffiths - Predictors of irregular word and non-word spelling in 11-15 year old students
Purpose The aim of this study was to explore the predictors of irregular word and non-word spelling ability in secondary school students. It was predicted that performance in a visual attention span task, used by Bosse, Tainturier and Valdois (2007), might be most strongly associated with irregular word spelling while Spoonerisms accuracy, a measure of phonological ability, was predicted to be most strongly associated with non-word spelling. Method Sixty-six students aged 11-15 years participated in a spelling to dictation task using items from the FRLL test (Funnell et al., 2009). Assessments of visual attention span (VAS), vocabulary and nonverbal ability were also included, as well as a Spoonerisms task. Results Stepwise regression analyses, where chronological age and vocabulary were entered on the first step, revealed that visual attention span task performance was a significant predictor of irregular word spelling, while for non-word spelling Spoonerisms accuracy was the significant predictor. Conclusion Visual attention span was shown to affect irregular word spelling, and a measure of phonological ability to influence non-word spelling. The findings reinforce those found for reading by Bosse and Valdois (2009), who argued that VAS has an important role in the development of whole word orthographic representations, and, for English, this should be reflected in a selective influence on irregular word reading. References: Bosse, M-L., Tainturier, M.J. & Valdois, S. (2007). Developmental dyslexia: the visual attention span deficit hypothesis. Cognition, 104(2), 198-230. Bosse, M-L. & Valdois, S. (2009). Influence of the visual attention span on child reading performance: a cross-sectional study. Journal of Research in Reading, 32, 230-253. Funnell, E. et al. (2009). FRLL test of regular words, irregular words and nonwords. Research version
Sarah Sau Ha Luk (sarahluk@cuhk.edu.hk)Pui-wan Cheng - Intervention for Children with Dyslexia in Chinese: A Comparison of Two Instructional Approaches
Purpose: This study evaluated the effectiveness of two supplemental intervention programs for children with Dyslexia in Chinese. Method: Twenty-two children (mean age: eight years and four months) who were formally diagnosed with dyslexia were randomly assigned to one of the two theoretically different reading interventions. Children in the Structured Reading Program (SRP) received explicit and systematic instructions in lexical knowledge of high-frequency Chinese characters and reading fluency. Children in the Guided Reading Program (GRP) received traditional instructions through integrated language experience, which focused on reading comprehension, language games and creative writing. Both groups received 25 two-hour sessions (50 hours) of intervention led by the researcher over five consecutive weeks in the summer when the children did not attend regular classes. Pre- and post- intervention reading achievements were measured in terms of (1) lexical knowledge (Timed and Untimed Copying, Ortho-Phonological Knowledge, Ortho- morphological Knowledge and Dictation), (2) reading fluency development (Character Reading Fluency and Passage Reading Fluency), and (3) reading comprehension. In the course of the intervention, students' progress was monitored eight times by measuring the students' character and passage reading fluencies. Results and Conclusions: Results indicate that (a) children who received either the SRP or GRP intervention scored higher on measures of character reading fluency, passage reading fluency and dictation, than children who received no intervention; (b) SRP students demonstrated greater gains over the GRP group at the end of the 50-hour intervention on measures of character reading fluency and reading comprehension; (c) GRP students showed an advantage over the SRP group on passage reading fluency; and (d) there was no significant difference in reading motivation between the two intervention groups. Implications for future research and practice of supplementary literacy interventions in Chinese in Hong Kong for children with dyslexia are discussed.
Yang Luo (OISE/University of Toronto)Xi Chen, OISE/University of Toronto - visual perception and short-term memory in learning to read Chinese characters
Purpose This study investigated the roles of visual perceptual skills and short-term memory (STM) in learning to read Chinese characters among both Chinese monolingual and Chinese-English bilingual beginning readers. Method We tested 116 monolingual Chinese children (Mean = 80.78 months, SD = 14.34) in China, and 86 Chinese-English bilinguals (Mean = 77.6 months, SD =13.28) in Canada, on character reading, general-visual and visual-orthographic perception, visual and verbal STM, vocabulary, phonological awareness, and digit naming. General-visual and visual-orthographic processing were measured, respectively, with TVPS (Gardner, 1996) and the experimental tasks developed by the authors with character/pseudocharacter stimuli. Results Based on the data, two groups were indentified with low and high reading abilities. Both general-visual and visual-orthographic skills were a unique predictor of reading for the low-level group after controlling for other cognitive skills, whereas only visual-orthographic skills uniquely predicted the high-level group's reading. The results of factor analyses showed that the two types of measures were loaded on one factor for the low-level reading group, but on two separate factors for the high-level group. Moreover, visual memory was a better unique predictor of reading for the low-level group, but verbal memory became a better predictor for the more advanced readers. Conclusion Our findings suggested that general-visual and visual-orthographic play different roles in the development of Chinese reading ability, and so do verbal and short-term memory. They also urged us to reconsider the importance of visual and memory skills in our conceptualization of models for reading Chinese.
George Manolitsis ();George K. Georgiou; Rauno Parrila - Early phonological predictors of spelling achievement in Greek
Purpose: To examine how early phonological skills - measured in kindergarten, grade 1, and grade 2 predict spelling achievement in grades 3 and 4. Phonological processing skills (phonological awareness, rapid naming, and phonological memory) have been shown to be important predictors of early spelling development in several studies (e.g., Caravolas, Hulme, & Snowling, 2001; Plaza & Cohen, 2004; Porpodas, 1999). However, very little is known about how they may affect later spelling achievement in mid school grades and particularly in Greek, a consistent orthography in reading, but inconsistent in spelling. Method: Seventy Greek children were followed from kindergarten until grade 4. In kindergarten, and in grades 1 and 2, they were assessed on measures of phonological awareness, phonological memory, and rapid naming. In grades 3 and 4, they were assessed on measures of spelling achievement. Results: Hierarchical regression analyses with non verbal ability, and letter-sound knowledge in kindergarten as control variables showed that (a) none of the phonological processing measures in kindergarten or grade 1 accounted for unique variance in spelling, (b) phonological awareness and rapid naming in grade 2 predicted uniquely spelling, and (c) phonological memory did not predict spelling at any measurement point. Conclusions: These findings suggest that both phonological awareness and rapid naming are critical for spelling in Greek, but their effect appears to take place after the beginning of formal reading instruction.
Eva Marinus (Macquarie University/ ARC Centre of Excellence in Cognition and its Disorders (CCD));Peter F de Jong; Kate Nation - Comparing the development of word-recognition processes in English and Dutch children
Purpose: English and Dutch are both Germanic languages and highly comparable in terms of syllable complexity. However, in contrast to Dutch, English is highly inconsistent with regard to letter-to-sound mappings. It has been suggested that English children learn to use additional word-recognition strategies, like the use of larger letter units, in order to resolve these inconsistencies, (Ziegler & Goswami, 2005). In this study, three different experiments were conducted focussing on cross-linguistic differences in developmental reading strategies. Method: Participants were 54 English and 54 Dutch children (grade 3 and 4). The groups were individually matched on word-reading performance, vocabulary, non-verbal reasoning, gender and years of reading instruction. Three naming tasks were administered comparing a) the use of a letter-by-letter decoding strategy, b) sensitivity to neighbour words, c) the use of rime units. Results: In the English children, the use of a letter-by-letter reading strategy diminished as a function of reading proficiency. In addition, they showed sensitivity to neighbourhood effects, but only for pseudowords. Finally, they did not seem to use rime units. The data of the Dutch sample has just been collected. Conclusions: The results are not in line with the idea that English children learn to use additional decoding strategies. Just like children learning to read in more consistent languages, English children seem to proceed from a serial decoding strategy to direct word recognition as their word-recognition skills improve. There were no indications that the letter-by-letter decoding strategy was supplemented by the use of rime units.
Rhonda Martinussen (University of Toronto);Martinussen, Rhonda; Judith Wiener - Examining the contribution of spelling, reading comprehension, and working memory to written expression skills in adolescents with and without ADHD
Purpose: In this study we examined the contribution of spelling, reading comprehension, working memory (WM), and attention-deficit hyperactivity disorder (ADHD) diagnostic status to writing ability in a sample of adolescents (aged 13 to 18 years). Methods: The participants included 26 adolescents (13 males, 13 females) without ADHD and 19 adolescents (9 females, 10 males) with a prior diagnosis of ADHD. The participants completed the Spelling, Passage Comprehension, and Writing Samples subtests from the Woodcock Johnson Tests of Achievement-Third Edition (Woodcock, McGrew, & Mather, 2001). The Digits Forward and Digits Backward subtests from the Test of Memory and Learning- Second Edition (Reynolds & Voress, 2007) were used to index verbal storage and verbal WM respectively. Standard scores (adjusted for age) were used in the analyses. Results: Adolescents with ADHD scored significantly lower their peers without ADHD on each academic measure and the verbal storage and WM subtests. The results of a multiple regression analysis with the total sample indicated that spelling, reading comprehension, and WM, but not ADHD diagnostic status, accounted for unique variance in writing achievement. When the analyses were repeated within each group, the findings indicated that WM and reading comprehension were independent predictors of writing ability within the ADHD group, whereas spelling was the only significant predictor within the comparison group. Conclusions: Adolescents with ADHD are at risk for low achievement in multiple domains of literacy. These findings suggest that weaknesses in reading ability and WM constrain written expression skills in adolescents with ADHD.
Linda H. Mason (The Pennsylvania State University) - Writing instruction for rural students in seven U.S. states
Purpose Students in rural schools often have poor academic outcomes. Two randomized control studies were conducted to investigate the effects of Self-Regulated Strategy Development instruction (SRSD) for writing with 6th grade rural students across seven U.S. states. Method Classrooms were randomly assigned to SRSD for persuasive writing. Teacher training in SRSD was conducted in a 1-day workshop with on-line consultation. Student essays were collected prior to instruction, immediately following instruction, and at three maintenance time points. Results Results of ANOVA testing indicated SRSD students (N=338 year1; N=17 year 2) had significant improvement in number of persuasive elements written (see effect sizes below). Testing condition Year 1 Year 2 Pre-instruction to post-testing .72 .56 Pre-instruction to maintenance 1 .69 .71 Pre-instruction to maintenance 2 .53 .35 Pre-instruction to maintenance 3 .39 .42 Preliminary analysis of essay quality and length also indicates improvement. Students in no-treatment control had no significant improvement at any point during the year 1 study (N=226). In the year 2 study (N=129), control students had a small improvement effect at post-testing, however, by the end of the year results indicated negative performance compare to pre-test. Conclusions Linear modeling analysis to determine factors influencing performance gains and to rule out competing/confounding variables will be completed and shared in the presentation. The training context and fidelity of instruction for rural educators is challenging. Given this, the preliminary results for the 2 studies are impressive. Implications for research and practice in rural schools will be noted.
Hisashi Masuda (Hiroshima Shudo University); Takaaki Tabuchi; Chikako Fujita - The role of phonetic radicals in processing Japanese two-kanji compound words
Purpose: Several studies have suggested that Japanese kanji characters are segmented into submorphemic components, or radicals, and are influenced by the phonology of the radicals (Masuda & Saito, 2002). This study addresses whether the phonology of radicals also affects the recognition of two-kanji compound words. Method: Undergraduate students made lexical decisions for two-kanji compound words primed by a single kanji character. The relationship between the prime and the first constituent (left) kanji of the compound target was varied in terms of phonetic radical consistency referring to whether the right-side phonetic radical is consistent with the on-reading of the whole kanji character. In Experiment 1, the prime was a radical-neighbor character with the first constituent. For consistent primes, the phonetic radical was consistent with the character phonology of the prime but not with the phonology of the first constituent, while, for inconsistent primes, the phonetic radical was inconsistent with the primes but consistent with the first constituent. In Experiment 2, the prime was identical to the left constituent of the compound targets. Results: In both experiments, greater priming was observed for inconsistent primes than for consistent primes. Conclusions: The results indicate that the phonology of the right radicals in the primes is activated. The results also suggest that two-kanji compounds can be accessed more quickly following presentation of the left kanji, especially for inconsistent left kanji. These findings indicate that the phonology of the right radical for the first constituent is activated in the recognition of two-kanji compound words.
Urs Maurer (University of Zürich)Enrico Schulz; Sanne van der Mark; Kerstin Bucher; Silvia Brem; Ernst Martin; Daniel Brandeis - Reading for meaning in dyslexic and young children: Distinct neural pathways but common endpoints
Purpose: Developmental dyslexia is characterised by impairments of reading fluency and comprehension. Neuroimaging studies using fMRI and ERPs and comparing sentence reading between dyslexic children (DYS) and chronological age-matched controls (CA) suggest both deviant basic word processing and deviant semantic incongruency processing. We add a younger control group at a similar reading level (RL) as the dyslexic children to clarify which of these effects are specific for dyslexia and which are unspecific just reflecting low reading skills. Method: We recorded fMRI and ERP data from DYS, CA, and RL children (n=19 each) in a sentence reading paradigm. Basic word reading was measured as response to all words; semantic processing as contrast between incongruent and congruent sentence endings. Results: Both the RL and DYS groups showed a similar reversal of incongruency effects in the inferior parietal region (fMRI data) and similarly reduced incongruency effects around 400ms (ERP data) compared to the CA group, suggesting that the semantic impairment in dyslexia resembles a developmental delay. Furthermore, the DYS group showed reduced sentence reading-related activation in the inferior parietal cortex in the fMRI data, and at around 100ms (trend) and 400ms in the ERP data when compared to both the CA and RL groups, suggesting dyslexia-specific deficits in basic word processing during sentence reading. Conclusion: Low reading skills due to young age and due to dyslexia-specific word processing deficits reflect different pathways which impair semantic processing in similar ways.
Genevieve McArthur (Macquarie Centre for Cognitive Science); Max Coltheart; Linda Larsen - Phonics and sight-word training in a group of children with developmental dyslexia
Purpose: The aim of this study was to compare the separate effects of phonics training and sight-word training in children with developmental dyslexia. Method: Ninety children with dyslexia were divided into three training groups. One group (N = 30) did 8 weeks of phonics training followed by 8 weeks of sight-word training. The second group did 8 weeks of sight-word training followed by 8 weeks of phonics training (N = 30). The third group did two 8-week period of a mixture of both phonics and sight-word training (N = 30). Each child was tested for a large number of phonics and sight-word skills before and after an 8-week period of no training (to gauge test-retest effects on the tests) and then after each 8-week training period. Results: Preliminary analyses of the group data show that phonics training had statistically significant and unique (i.e., not present after sight-word training) gains in letter identification, parsing, letter-sound knowledge, and nonword spelling. Sight-word training lead to a significant and unique gain in spoken vocabulary. Both phonics and sight-word training lead to common gains in regular-word reading fluency, irregular-word spelling, and phonological short-term memory. Conclusions: Considered together, these results support the idea that children with different patterns of development dyslexia should respond to different types of reading treatment.
Nicola McClung (UC Berkeley)Diana Arya; Anne Cunningham - A Cross-cultural comparison of home-literacy environments and their long-term association with schooling and fourth-grade literacy achievement
Purpose Existing research has not addressed the question of whether the relationship between children's home environments and their literacy development differs internationally. Moreover, few studies have investigated the generalizability of English-language research. The Progress in International Reading and Literacy Study (PIRLS) 2006 dataset was employed to investigate universal and culture-specific home predictors of reading achievement. Method Hierarchical linear modeling was used to analyze large, representative samples of fourth-grade children in Taiwan (n=3,466), Iran (n=3,480), Norway (n=2,540), and South Africa (n=3,640). Literacy achievement was regressed on home environment variables (e.g., home literacy practices, parental beliefs, and family background characteristics), while accounting for the hierarchical structure of the data-students nested in schools. Results Findings indicate that early parent-child book reading and father's level of education are universal predictors of reading achievement. In contrast, the importance of other home environment variables differs across countries. For example, while parent-child writing practices positively influence reading performance in Iran, Norway and South Africa, these factors are negatively correlated in Taiwan. We hypothesize that this difference may be due to the added challenges of learning in a logographic orthography. Conclusions Children who engage in rich literacy activities in both their home and school environments tend to be the most skilled readers across all languages studied. A multitude of experiences with adult-child book reading helps to support literacy achievement. Parental beliefs and reading practices and the degree of linguistic diversity in a country have a complex and influential role in children's language and literacy development.
Heather McDowell (Birkbeck College, University of London); Marjorie Lorch - Under-specified and multiply-influenced phonemic awareness in teenagers from Mainland China, Hong Kong and England
Purpose: Crosslinguistic differences in phonological awareness development (Ziegler & Goswami, 2005) are indicated by low levels of phonemic awareness in Hong Kong L2 alphabetic literates and strong performance by Mainland Chinese Pinyin-literates (Holm and Dodd, 1996). Meanwhile, English readers display orthographic influences in phonemic awareness tasks (Castles et al. 2003). This research investigated the extent of the phonemic awareness possessed by these three groups within a range of phonological and orthographic contexts. Method: Participants were teenagers from Mainland China, Hong Kong and England (n=90) all undertaking secondary education in the UK. As part of a larger project, an English word and pseudoword phoneme segmentation test was administered. An important innovation was in devising a controlled set of stimuli containing different orthographic and phonological features which allowed a fine-grained assessment of phonemic awareness. Results: Hong Kong participants displayed significantly lower phonemic awareness than both other groups (Mdn: 37.1%), while performance of the Mainland Chinese group (Mdn: 66.1%) surpassed the English native speakers (Mdn: 54.2%). Detailed error analysis reflected orthographic and phonological features of stimuli. Group-specific patterns were consistent with aspects of instructional, orthographic and phonological background and all groups showed difficulty in consistently segmenting at the phonemic level. Conclusions: Results suggest that alphabetically-literate teenagers from Mainland China, Hong Kong and England do not possess fully-elaborated phonemic awareness. Phonemic awareness appears to be derived from multiple sources of information including orthographic, phonological and explicit training input.
Sarah McGeown (Psychology Department, University of Hull); Emma Medford; Naomi Hughes - Investigating the effectiveness of phonological and visual word reading strategies
Purpose: The aim was to examine the effectiveness of phonological and visual reading strategies on children's performance in different literacy tasks (regular and irregular word reading, complex word reading, spelling and reading comprehension) and identify the cognitive skills underpinning these different literacy abilities. Method: One hundred and eleven children (aged 9 -11) completed assessments of reading comprehension, word reading, spelling, regular and irregular word reading, verbal IQ, nonword reading and reading frequency. Analysis of reading errors identified children with either a phonological or visual reading strategy. ANOVAs examined differences in the effectiveness of these strategies and regression analyses investigated the cognitive skills underpinning the different literacy abilities. Results: Children with a phonological reading strategy had superior nonword and regular word reading. Whilst their complex and irregular word reading was better, this was not quite significant (p = .07). No differences were found in reading comprehension or spelling. Decoding skill (nonword reading) accounted for most variance in all word reading and spelling tasks, however verbal IQ was equally important in reading comprehension, and reading frequency only contributed variance in those literacy tasks which included complex words. Conclusions: Despite the non-transparent nature of irregular words, a phonological reading strategy did not impair performance in irregular word reading. It is suggested that better decoding skills and a more phonological reading strategy may facilitate irregular word reading, as even irregular words contain regular elements which provide a cue to pronunciation.
Meredith McKague (The University of Melbourne) - Masked form-priming effects in episodic recognition and lexical decision: an index of lexical acquisition.
Purpose. The research aimed to identify signatures of episodic and lexical memory processes in reading. Method. Sixty-four participants completed both lexical decision and episodic recognition. Two lists of 128 items (64 words, 64 nonwords) were used with every item appearing in both tasks across participants. For episodic recognition, half of the words and nonwords were studied prior to a recognition test. Half of the items in each task were high-N and half were low-N. Targets in both tasks were preceded by one of three different masked form-primes created by substituting a letter, or letters, from the target word; one-letter-different, two/three-letter-different, and an all-letters-different control prime. RTs to targets as a function of task, N and prime served as the dependent variable. Results. Dissociable priming effects were observed between the two tasks, evidenced by a significant Task x Prime x N interaction. In episodic recognition, both studied words and nonwords showed facilitatory form priming regardless of N, even when primes differed by two or three letters, and priming was greatest for high-N items. Significantly less form priming was observed in lexical decision, and was only significant for low-N words when primes differed by a single letter. Conclusions. The dissociable patterns of masked form priming effects can be used to inform studies investigating the acquisition of orthographic representations for newly learned words. They will enable orthographic learning to be traced from recognition based on an imprecise episodic representation to lexical access based on a precisely specified orthographic representation.
Hedva Meiri (University of Haifa); Zvia Breznitz, PhD - Cognitive processing style and reasoning strategies of dyslexic readers during learning tasks
Purpose: Examining the cognitive processing style of dyslexic readers during the execution of learning tasks can make a crucial contribution to understanding the dyslexia phenomenon. Documented differences between dyslexic and skilled readers have been found in experiments employing both implicit and explicit learning. Whether these differences are related specifically to reading or are more general characteristics of the dyslexic learning style is yet another question. Method: Adult dyslexic and skilled readers completed an implicit learning task with visually presented sequential stimuli and two explicit learning tasks, one with linguistic stimuli and the other with nonlinguistic. The implicit learning task was a two-alternative forced choice task which required participants to respond to targets and nontargets. Participants were unaware that the stimuli were presented as a 12-item repeating sequence. On the explicit learning tasks participants were asked to identify the rule by which the presented stimuli were created. Participants were asked to verbalize their thought processes during execution of the tasks. The verbalizations were recorded by the experimenter. When participants felt they understood the governing rule, a test block was completed. Results: Results reveal differences in accuracy, reaction times, and awareness of a governing rule on the implicit learning task. On the explicit learning task marked differences in strategy and willingness to investigate ulterior options characterized the research groups. Conclusions: The findings of this study shed light on the qualitative differences that distinguish the cognitive processing and learning strategies of dyslexic readers from those of skilled readers.
Monica Melby-Lervag (University of Oslo, Dep. of special needs education)Arne Lervåg - Language and literacy skills in L2 learners: A meta-analysis
Purpose: The main aim was to investegate the nature of language and literacy skills in L2 learners, and what moderates these skills. Two main hypotheses were examined; that L2 learnes have a metalinguistic advantage and that language skills transfer from L1 to L2. Method: A comprehensive literature search was undertaken for studies with group comparisons of language and literacy skills in L1 and L2 learners, and correlational studies of L1 and L2 language and literacy skills in L2 learners. Results: For decoding (N L1= 80,563, L2 = 9,100) and phonological awareness (N L1=78,418, L2=7,683) the group difference was small in favour of L1 learners. For oral language (N L1=23,374, L2=8,111) and reading comprehension (N L1=39,279, L2=6,705), the group difference was large in favour of L1 learners. The size of the group differences in language and literacy skills were dependent on moderator variables related to SES and langugage type. The correlation between oral language L1 and L2 (N=2,912) in L2 learners was very small, but large for decoding L1 and L2 (N=2,326), and for phonological awareness L1 and decoding L2 (N=1,602). Conclusions: The results do not support the hypothesis of a metalinguistic advantage as L1 learners have better phonological awareness and decoding skills than L2 learners. Also, L2 learners have a large disadvantage on oral language and reading comprehension. As for cross-linguistic transfer, oral language skills in L2 are relatively independent of oral language skills in L1, while decoding and phonological skills in L2 benefits from these skills in L1.
Maggie Middleton (The Ohio State University)S.A. Petrill - Making information meaningful: comprehension assessment
Title: Making Information Meaningful: Comprehension Assessment Authors: M.E. Middleton & S.A. Petrill Purpose. This study examined the environmental and genetic influences on the relationship between narrative and reading comprehension in young children. Methods. We examined this issue in a sample of 518 identical and same-sex fraternal twin pairs (MZ = 214, DZ = 304) recruited from the Western Reserve Reading Project. The McDonald's Story, The Shipwreck, and The Dragon Story from the Test of Narrative Language, Woodcock Johnson Passage Comprehension, and the Peabody Reading Comprehension subtest were assessed from tester-administered measures. Results. Two latent factors of narrative and reading comprehension were formed using using the language subtests and the comprehension subtests respectively. The variance of the latent factors was divided into estimates of additive genetic, shared environment, and nonshared environmental influences using Cholesky decomposition. The results indicated unique significant genetic, nonshared, and shared environmental influences for narrative comprehension. There was a significant genetic and shared environmental overlap between narrative comprehension and reading comprehension. There were no unique additive genetic or shared environmental influences on reading comprehension not accounted for by narrative language. Finally, there was significant nonshared influences on comprehension suggesting that forty-percent of reading comprehension is not accounted for by language. Conclusions. Performance on the Test of Narrative Language reflected listening skills and language skills. Children's ability to remember information is dependent upon knowledge of words meanings, knowledge of sentence structures, relationships between words and ideas, and the ability to understand and remember the gist of the story.
Fiona Middleweek (University of Oxford, UK); Victoria A. Murphy; Mairéad McKendry; Kathy Sylva - Comprehension skills in EAL children in later primary/early secondary school years
Purpose: While the reading abilities of younger EAL children have been explored in some depth by the research community, few studies have examined how EAL children's reading and language comprehension skills develop in later schooling. Even fewer studies have examined this issue within the UK context. The current study reports on preliminary UK data on the language and reading skills of EAL children from similar linguistic backgrounds (Indo-Iranian L1s) at the end of primary school (age 9-10) and beginning of secondary school (age 11-12). Method: Reported language measures include receptive and expressive naming vocabulary, depth of vocabulary knowledge, morphological awareness and exception word reading administered to the two year groups of Native Speakers (NS) and EAL children. Results: The contribution of semantic factors to the reading and listening comprehension skills of the older EAL children are reported. The semantic profiles of the EAL children who are poor comprehenders are compared to those of the more competent EAL readers as well as to a matched group of poor and skilled English L1 readers. Furthermore, specific comparisons are made across the EAL and NS children from end of primary to early secondary school to identify the extent to which the relationships between these variables change as the children progress. A measure of the home literacy environment for the EAL and English L1 children is included in the analysis. Conclusions: The implications of these preliminary findings for the academic achievement of EAL children in primary and secondary education are discussed with specific reference to how these initial findings compare to research findings from other countries and learning contexts.
Brett Miller (Eunice Kennedy Shriver National Institute of Child Health and Human Development) - Reading, writing, and related learning disabilities research program at the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health
This poster will present current funding opportunities available from the National Institutes of Health (NIH) for graduate students, post-doctoral fellows, and independent researchers interested in additional training, mentorship or research funding. NIH has a long history of providing support for research and training initiatives to increase understanding of both normal and atypical development of reading and written language skills throughout the life course and the training and research opportunities highlighted will focus on scientific topics related to literacy and related learning disabilities. A program staff member will be available to answer general questions about NIH funding or questions about funding for specific research projects or training opportunities for individuals from U.S. or non-U.S. institutions.
Suzanne E. Mol (Faculty of Social and Behavioural Sciences); Adriana G. Bus - A meta-analysis on young children's home literacy environment; An update and replication of Bus et al. (1995)
Purpose: In previous reviews (Bus, Van IJzendoorn, & Pellegrini, 1995; Scarborough & Dobrich, 1994) self-report questionnaires were used as indicators of young children's exposure to print. Such questionnaires, however, are likely to suffer from a social desirability bias. A print exposure checklist is an alternative measure that is thought to be a more objective proxy of reading volume (Stanovich, 2000). In the current meta-analysis, we compared effect sizes for oral language and emergent reading skills as revealed by both measurement methods by matching studies in which a print exposure checklist was completed by parents of 3- to 6-year olds with studies that administered a self-report HLE-questionnaire. Method: For 14 of the 15 included print-exposure studies, we matched a HLE-study on four main characteristics: sample size, mean age, home-language, and SES. Two coders coded 90% of the studies (κ=.96). Results: The association between oral language skills and print exposure checklists was moderate (r=.34***) and did not significantly differ from the association with HLE-composites from the set of matched studies (r=.32***). Emergent reading skills such as alphabet knowledge and phonological processing showed a small association with both the print exposure checklists (r=.29***) and the HLE-composites (r=.18***). Conclusion: Our findings replicate the meta-analysis by Bus et al. (1995) that about 8% of the variance in young children's reading and oral language is related to book sharing. The associations found with HLE-questionnaires resembled the associations with print exposure checklists, indicating that print exposure checklists tap into an environment that familiarizes children with books.
Padraig Monaghan (Lancaster University) - Linking Computational Models with Theories of Multidimensional Impairments in Literacy
Lori Morris (Université du Québec à Montréal) - The effects of parent-child reading on the development of French pre-reading skills in Innu-speaking kindergartners
Purpose This experiment study assessed the impact of participation in a parent-child reading program on the development of the L2 French pre-literacy skills of a group of Innu-speaking kindergartners. Innu children often have low print exposure at home, poor oral vocabularies in both L1 and L2, and low levels of academic achievement. The hope was that teaching Innu parents how to read to their children and supplying them with books would result in significant gains in a variety of pre-literacy measures, particularly in expressive and receptive lexical knowledge. Method The study had 32 child participants. All were pre-tested in October and post-tested in May using 13 independent measures from the Nouvelles épreuves pour l'examen du langage. Twenty families participated in the parent-child reading program and formed the experimental group. The eleven non-participating families (including 12 children) formed the control group. Results While the cohort as a whole made several significant pre-/post- gains, the children whose families participated in the parent-child reading project made significantly greater gains in working memory span, expressive vocabulary, receptive vocabulary, and mastery of a number of basic spatial and numeric concepts. Conclusions The parent-child reading program resulted in good pre-literacy gains in children from a highly at-risk population. Given the current dearth of viable literacy development programs in First Nations communities and the simple, easily replicable design of this particular form of intervention, the implementation of similar programs in Amerindian and similar highly at-risk contexts might be considered by school communities.
Frederick Morrison (University of Michigan) - The role of executive function in growth of early literacy skills
Purpose: The presentation will present results from an ongoing project examining the role of executive function (EF) in early literacy development. Executive function has received increased attention in neuropsychological, cognitive and educational fields as a crucial skill in academic success. We have been systematically examining the extent and nature of individual differences in executive function and the role it plays in growth of literacy skills. Method and Results: Three major findings will be highlighted. First significant and meaningful individual variation in EF skills (attention control, working memory and response inhibition ) emerge early in the preschool years. Further, significant gender differences in response inhibition favoring girls have been revealed (Matthews) by Kindergarten, in both mean level of performance and overall distribution, with boys almost exclusively represented at the bottom end of the distribution. Second, the influence of schooling on growth of EF skills has been studied using a natural experiment (school cutoff), which compares growth of children who just make vs miss the cutoff for school entry. Strong schooling effects have been seen for working memory (Burrage) but less evidence for schooling influences on response inhibition (Burrage, Skibbe). Finally, a recent study of the impact of classroom disruptions (Bell) found that, for children with initially low EF skills, the more disruptions a classroom experienced, the weaker was children's literacy growth. However, for children with average or higher EF skills, amount of classroom disruptions had no effect. Conclusions: Overall findings from this project highlight the existence of major variability in EF skills early in development, the importance of EF for classroom functioning and the differential impact of schooling experiences on different EF skills.
Ian Mundy (University of Warwick); Julia Carroll - Representations of lexical stress in adults with developmental dyslexia
Purpose: The aim of this study was to investigate the ability of adults with developmental dyslexia to accurately represent the prosody (i.e. lexical stress patterns) of words stored in the mental lexicon. Methods: 16 adults with developmental dyslexia (mean age = 24.25, 8 males) and 24 age/IQ matched controls (mean age = 19.46, 4 males) completed a cross-modal priming task in which visually presented target words were preceded by stress-congruous, stress-incongruous and control primes. In the congruous condition the spoken prime was the first two syllables of the target word (e.g. prose/cutor → PROSECUTOR). In the control condition the spoken prime was the first two syllables of a word phonologically unrelated to the target (e.g. accel/erate → PROSECUTOR). In the incongruous condition the spoken prime was the first two syllables of a word derived from the same root as the target, with matching segmental phonology but contrasting stress assignment in the first two syllables (e.g. prose/cution → PROSECUTOR). Results: Both groups of participants showed the same pattern of priming effects. Relative to the control condition, priming for stress-congruous primes (controls = 155.40ms, dyslexics = 314.75ms) was more than twice as strong as priming for stress-incongruous primes (controls = 63.33ms, dyslexics = 116.49ms). Conclusions: The results suggest that adults with developmental dyslexia represent word-level prosody fully and accurately. This contrasts with their imprecise encoding of segmental phonology. Reading interventions may capitalise on these findings by exploiting intact prosodic representations to facilitate the learning of segmental phonology.
Maria Murray (SUNY Oswego - Curriculum and Instruction Dept);Maria S. Murray; Chris Schatschneider; Benita Blachman - The effect of response to instruction on the prediction of kindergarten invented spelling ability
Purpose The purpose of this study was to investigate differences in the amount of variance in kindergarten invented spelling (measured in June) explained by letter-sound knowledge (measured in both January and June). Method Kindergarten children in nine upstate New York inner-city schools (n=306) participated in this study. Data were collected from a larger study that randomly assigned the schools to conditions (some received an intervention and some did not). Influence of the intervention was controlled for in the analyses. Hierarchical linear regression was used to explore the extent to which letter-sound knowledge (and other predictor variables) in January and June of kindergarten explained unique variance in June invented spelling. Results January letter-sound knowledge explained 20% of the variance in June invented spelling, whereas June letter-sound knowledge explained 60% of the variance in June invented spelling, despite the fact that the variance of January letter-sound knowledge (SD = 7.17) was similar to the variance of June letter-sound knowledge (SD = 7.68). An inspection of scatterplots revealed a "response- to-instruction effect," in that the participants who responded to exposure or instruction (i.e., learned letter sounds) were more likely to produce better invented spelling in June. Those who had not learned their letter sounds by June performed poorly on the invented spelling measure. Conclusion Researchers have noticed the predictive validity of measures is sometimes affected by response-to-instruction. Knowing more about the impact of rate of growth of letter-sound knowledge on spelling could provide additional support for emphasizing this skill in the kindergarten classroom.
Bertram Müller-Myhsok (Max Planck Institute of Psychiatry) - Merging kernel-methods and statistical genetic methodology - a promising outlook for the genetic study of dsylexia and sub-phenotpyes
Purpose: To illustrate that the merging of kernel methods (such as support vector machine technology) and statistical genetics methodology may provide very important insight into the genetic basis of dyslexia Methods and Results: Kernel-based methods are successfully used in classification tasks. Many of their properties, such as the ability to deal with vast sets of features (phenotypes) of different origins (such as EEG data, test results, NMR data) make them very attractive for the study of the biological basis of dyslexia. I will show how to combine these properties with the rigorous testing standards implemented in statistical genetics to arrive at a both very flexible and powerful as well as reliable methdology. Conclusions: The merger of the two methodologies works and has some very desirable properties
Kari-Anne B. Naess (Department of special needs education);Solveig-Alma Halaas Lyster; Monica Melby-Lervåg - Language and memory skills in children with Down syndrome: A preliminary meta-analytic review
Language and memory skills in children with Down syndrome: A preliminary meta-analytic review Purpose: The study presented is a meta-analysis of studies comparing language skills in children with Down syndrome and typically developing children of the same nonverbal mental age. Method: The meta-analysis comprises journal articles published before 07.04.2009, identified through ERIC and PsycINFO by using the key words Down* and language. A total of 23 articles met the criteria for inclusion. Each study was carefully reviewed and coded based on both inclusion criteria and coding protocol before the analysis was performed in the program "Comprehensive meta-analysis". Results: Results showed no significant difference between the groups in receptive vocabulary (g = - 0.02), but there were differences between the groups, in favour of the typically developing children, in expressive vocabulary (g = - 0.98), grammar (g = - 0.78) and auditory memory (g = - 0.86). The meta-analysis showed large heterogeneity between the studies in all the outcomes and different moderator analyses have therefore been done. Conclusions: Based on the meta-analysis, it seems reasonable to conclude that children with Down syndrome have a specific deficit related to expressive language skills, grammar and auditory memory. By this, the meta-analysis confirms conclusions that have been drawn from previous research. However, an evaluation of the methods being used in the different studies indicates that matching strategies may have large effects on study outcomes. Matching strategies may therefore be an important moderator variable to consider when planning future studies of children with Down syndrome.
Hannah Nash (UCL Division of Psychology & Language Sciences);Debbie Gooch; Charles Hulme; Maggie Snowling - The overlap between children at family risk for Dyslexia and children with Language impairment in the pre-school years
Purpose: This paper reports preliminary findings from the first phase of the Wellcome at-risk study, a 5 year longitudinal study exploring the overlap between specific language impairment (SLI) and Dyslexia. Method: 202 children aged 3-4 years were recruited to the study and classified as family risk for Dyslexia (FR), language impaired (LI) or typically developing (TD). This resulted in 81 TD children, 79 FR children (20 had an additional LI) and 42 LI children. Results: There were no significant differences between the 3 groups in age but there was a clear step pattern on all the language measures, letter sound knowledge and measures of non-verbal ability TD > FR > LI. The FR children with additional LI differed significantly from the rest of the FR group on all language measures but not in non-verbal ability, letter sound knowledge, SES or age. The FR without LI group still showed weaknesses compared to the TD group in the following, inflectional morphology, nonword repetition, sentence repetition, articulation and LSK. Expressive vocabulary and sentence comprehension were not impaired. Conclusions: 25% of children at family risk for Dyslexia fit criteria for pre-school language impairment. The remaining 75% showed weaknesses in expressive aspects of language. The group as a whole had poor letter sound knowledge, a key pre-literacy skill.
Kate Nation (University of Oxford) - Learning to read changes children's phonological skills: evidence from a latent variable longitudinal study of reading and nonword repetition
Purpose: Although it is clear that individual differences in nonword repetition are associated with language and literacy development, few studies have considered the extent to which learning to read influences phonological skills as indexed by nonword repetition performance. We explored this question using a latent variable longitudinal design. Method: Reading, oral language and nonword repetition were assessed in 215 children at age 6 years and one year later at age 7. Results: Structural equation modeling showed that reading at 6 years predicted growth in nonword repetition between 6 and 7 years, independent of the effects of oral language skills and the autoregressive effect of nonword repetition at 6 years. In contrast, nonword repetition did not predict progress in learning to read. Conclusions: These findings demonstrate that learning to read has a powerful effect on children's language processing systems. We consider how learning orthography might influence speech processing, and discuss the implications of our findings for theoretical accounts of reading disorder.
Sabina Neugebauer (Loyola University Chicago) - Daily differences in reading motivation: Young monolingual and language-minority adolescents' motivation to read inside and outside of school.
Purpose Numerous studies document low reading motivation among young adolescents in school; yet these same students exhibit high reading motivation outside of school (Moje, 2008). The current study of daily in-school and outside-of-school reading motivation will examine whether motivational differences between environments influence the reading comprehension of language-minority (LM) and monolingual students. The focus on LM students reflects their divergent linguistic and socio-cultural contexts outside-of-school and in-school, increasing the potential magnitude of difference in literacy motivation across settings (August, 2005). Method Fifth graders from two urban schools participated in this study (n=127): 65% Latino, 30% white, and 5% African American students, and 50% LM. Measures included a demographic survey, the Motivations for Reading Questionnaire (MRQ) (Wigfield & Guthrie, 1995), and daily administration of a 14-day log for recording reading motivation in outside-of-school and in-school settings, followed by the reading portion of the Massachusetts Comprehensive Assessment System (MCAS) as the achievement outcome. Data were analyzed using the Multi-level Model for change (Singer & Willett, 2003). Results Preliminary findings indicate more variation in daily reading motivation within individual students than between students. The disparity in within student reading motivation across environments negatively predicted performance on the MCAS for LMs and monolinguals: The baseline MRQ was not associated with achievement scores. Conclusions This study's methodology can provide a more comprehensive understanding of the relationship between cross context reading motivation and reading performance; supporting a context specific explanation for reading motivation, and underscoring the limitations of one-time reading motivation assessments.
Tom Nicholson (Centre of Excellence for Research on Children's Literacy, Massey University)Yi Li Huang - A revisionist analysis of miscue analysis - a comparison study of miscues made by 6- and 7-year-old European and Chinese good and poor readers when reading text
The aim of this study was to determine whether there is any strategic difference in the reading process of readers with different abilities and readers who speak English as a first or second language. Twenty European children and twenty Chinese children aged between 6 and 7 years of age participated in the study. They were divided into a good reader group and a poor reader group. All children read two stories from their school's book collection, one approximately at their reading age, and another above their reading age. Their miscues were recorded and analysed based on the Reading Miscue Inventory (Goodman & Burke, 1972). Analyses of variance were carried out to analyse possible differences in strategies between European and Chinese children, and between good and poor readers. The results favoured the print-driven, phonological decoding view of reading, in that good readers made better quality grapho-phonic miscues than did poor readers when reading both easy and hard stories. There was little difference between good readers and poor readers in their use of semantic and syntactic cues when reading at their instructional level, yet miscue analysis theory would predict that good readers would make better use of semantic and syntactic cues than poor readers. The results were also at variance with miscue theory in that good readers made better quality grapho-phonic miscues than did poor readers when miscue theory would predict that the miscues of good readers would only minimally rely on print information.
Mark Noordenbos (no member); Eliane Segers; Willy Serniclaes; Ludo Verhoeven - The allophonic mode of speech perception in children at risk for dyslexia.
Purpose: The aim of this study was to examine the behavioral and neurological outcome of speech perception in preschoolers. We know that dyslexics are better in the perception of speech sounds within the same category (Serniclaes et al., 2001), however is this same effect already present in children without formal reading instruction? Method: 35 children at-risk for dyslexia and 42 control children in the second year of kindergarten participated in this study. Categorical perception was investigated by means of identification and discrimination tasks with speech sounds that differed on the place of articulation. In first grade, the children were tested again behaviorally, and this time, also the Mismatch Negativity (MMN) was used as a neurological measure of the discrimination between the speech sounds. The MMN is an electric brain responses that is automatically elicited by any discriminable change in a repetitive sound. Results: There was no difference between at-risk and control children on the identification of the speech sounds, however there was a difference in discrimination. The at-risk children showed an allophonic mode of speech perception, and the control group did not. Differences between the two groups at a neural level are being investigated at this moment. Conclusions: Children at-risk for dyslexia already show an allophonic mode of speech perception before formal reading education.
Jane Oakhill (University of Sussex); Holly Pellatt - Sex differences in the effects of anticipated interest on good and poor readers' comprehension
In a previous study of the effects of reading preferences on comprehension, we showed that boys and girls had distinctly different preferences as to which topics they would prefer to read about. In addition, the boys' comprehension was better in the case of texts they professed to be interested in reading, whereas the girls' performance was relatively little influenced by interest levels. In the present study, we extend these findings by investigating not the content of the text, but the interest level engendered by the title. Half of the children were provided with a title that was manipulated to make the text appear more interesting and the other half had the original, rather bland, title. The text itself was identical for all children. Those children who received the "interesting" title rated their interest in reading the story higher (thus, the title manipulation was effective) and even maintained that the story had been more interesting after they had read it. The interest manipulation had a particular effect on the boys who were poor readers: those who received the low interest title performed particularly poorly in relation to boys who were better readers, and to girls in general. In addition, the pattern of reading persistence (number of questions attempted) showed a similar pattern - the boys who were poor readers, and who were given the low interest title, tended to be less persistent. The implications of these results for encouraging reading, particularly in boys, are discussed.
Ake Olofsson (Umeå University);Pernilla Juhlander; - Speed and accuracy in the first year of reading development.
Purpose. To provide a detailed description of the development of word recognition during the first year of schooling, with respect to processing speed and decoding accuracy. Method. One year longitudinal study of the first year of reading development in 35 seven year old Swedish children. The study used a computer based measurement procedure with four waves of repeated measurement of reading aloud of single words. The accuracy of word decoding was recorded and analyzed with respect to errors and decoding strategies. Reaction and process times were recorded for reading as well as for rapid naming and letter knowledge. The word material consisted of; (a) very familiar (content and functor) mono syllable simple words; (b) complex words which occur with high frequency in the early stages of children's reading schemes; (c) simple one-syllable pseudowords and (d) two-syllable pseudowords. Also a short spelling task was used. Assessment of letter knowledge and phonological awareness was done prior to school entry. Results. The results show a rapid reading development for practically all children and clear evidence for code oriented and phonologically based decoding strategies. At the first three timepoints monosyllable pseudo words were read more correct than simple words. The development of decoding speed was preceded by an increase in accuracy. Conclusions. When learning to read, in the Swedish orthography, the beginning readers rely on phonological skills and letter knowledge rather than using any holistic strategy. The results give no support to the claim of an initial logographic strategy in reading acquisition.
Richard Olson (University of Colorado)Janice Keenan - Genetic and environmental etiology of vocabulary and reading development from preschool through 4th grade
Purpose: We explored the genetic and environmental etiology of individual differences in vocabulary development from preschool through the fourth grade, and we assessed its genetic and environmental correlations with decoding, word recognition, and reading comprehension in the first, second, and fourth grades. Method: Vocabulary and other reading-related skills were assessed at preschool in nearly 2,000 identical and fraternal twins. Portions of this twin sample were subsequently tested on word reading and comprehension skills at the end of the 1st, 2nd , and 4th grades, and on vocabulary at the end of 2nd and 4th grades. Cholesky models were used to estimate the genetic and environmental influences on latent traits for vocabulary and reading, and on the correlations between the latent traits. Results: At preschool, 80% of the vocabulary variance was associated with shared family environment, and only 15% was due to genes. Genetic and shared environment influences were more equal at second and 4th grade, but the longitudinal genetic and environmental correlations for vocabulary were above .9, as were the genetic and environmental correlations between vocabulary and reading comprehension. Genetic and environmental correlations between vocabulary and word recognition increased across the grades, but not to the high levels seen between vocabulary and reading comprehension. Conclusion: Though the balance of genetic and environmental influence on vocabulary changes from preschool through the early grades, the genetic and environmental factors at preschool are nearly the same as those influencing individual differences in vocabulary at the end of 4th grade.
Sébastien Pacton (Université Paris Descartes / CNRS) - Do French children use information about the morphological structure of new words? Evidence from self-teaching situations
Purpose: We report two experiments investigating whether French grade 3-4 children benefit from the information about the morphological structure of pseudo-words in self-teaching situations. We compare children's learning of the spelling of "opaque" pseudowords with their learning of "morphological" pseudowords. Method: Pseudowords such as 'gilarand' (with the final /ã/ spelled -and as in the words grand, tall, or quand, when) and 'bolirant' (with the final /ã/ spelled -ant as in the words gant, glove, or avant, before) were embedded within a little story, as if they were real words. "Opaque" pseudowords appeared in a context in which nothing justifies the presence of the silent -d or -t (e.g., il se promène avec son gilarand, he is walking with his gilarand). Conversely, "morphological" pseudowords will appeared in a context that justifies the presence of the silent -d and -t (e.g., Un bolirand est quelqu'un qui bolirande des vaches, A /bolirã/ is a man who /bolirãd/ cows). Results and Conclusions: The results show that children learned "morphological" pseudowords better than "opaque" pseudowords, suggesting that they used the morphological information provided by the sentence in which the pseudo-words were embedded.
Prakash Padakannaya (University of Mysore, Mysore, INDIA)Mohanmad Ahmad Panah2, Nandini C Singh3 and R Malatesha Joshi4 2Department of Clinical Psychology, Medical Science University, Hamadan, IRAN. 3National Brain Research Centre, Manesar, 122050, INDIA 4College of Education and Human Development, Texas A & M University, Texas 77843-4232, USA - Effect of opacity and frequency in naming words in Persian abjad
Purpose The study examined the efect of frequency and transparency on word reading in native Persian readers by presenting different types of word lists to read. Persian orthography, though is an alphabetic one, is quite different from other scripts such as Hebrew and English. For example, almost every third word in a sentence could be a homograph representing different meanings. Placing an inappropriate diacritic mark on a consonant may change the structure and meaning of the sentence. Method Thirty tenth-grade Persian students (mean age = 14.5) from an Iranian high school in Tehran were presented with the following lists of words to read: List 1- twenty high frequency transparent words; list 2- twenty low frequency transparent words; list 3- twenty high frequency opaque words; list 4- twenty low frequency opaque words. Words across lists were comparable with regard to length and complexity. The performance was measured in terms of accuracy and the total time taken. Results The main effects for both frequency and opacity and the interaction effect were significant for accuracy scores while for speed the opacity effect alone was significant. Reading was better when the words had vowels. Conclusion The presence of vowels and or context seems to be very important in reading Persian regardless of one's reading level. Presence of vowel and context help to disambiguate homographs.
Jinger Pan (Beijing Normal University); Hua Shu; Catherine McBride-Chang; Yuping Zhang - The role of rapid automatized naming in Chinese reading accuracy and fluency
Purpose:Recent studies in alphabetic languages argue that RAN is not a unique predictor and its effect on reading is limited to fluency. Two studies investigated whether RAN is a unique predictor of Chinese reading, especially for accuracy since to provide evidence for this issue. Method of Study 1: 262 children who were administrated nonverbal IQ test, untimed character recognition test, vocabulary, phonological awareness (PA) and rapid naming. They were tested 3 times with an interval of 2 years between each time. Regression analyses were applied. Results of Study 1: RAN in kindergarten uniquely and consistently contributed to future reading accuracy, with PA, vocabulary, and character recognition at time 1 statistically controlled. Method of Study 2: Untimed character recognition, timed word recognition, PA and RAN were administrated to forty-five dyslexic children and forty-five age controls in grades 4 and 5. Path analyses were applied. Results of Study 2: The dyslexic children were significantly slower than their controls on the RAN task. RAN also predicted unique variance of both reading accuracy and fluency, with phonological awareness statistically controlled. RAN is a significant and consistent predictor of Chinese reading accuracy and fluency, which seems to be different from the situation typically observed in alphabetic languages. We suggested that RAN in Chinese reflects the process of lexical access and importance of RAN in Chinese is attributable to the specific properties of Chinese in which lexical access in extremely salient.
Timothy Papadopoulos (Department of Psychology, University of Cyprus); Maria Constantinidou; Sotiris Douklias - Comorbidity of reading disability and attention-deficit disorder: Findings from a longitudinal study
Purpose: The present study examined longitudinally the possible comorbidity of reading difficulties and attention deficits following up a group of children from Kindergarten to Grade 2. Method: Four groups were formed on the basis of two composite scores of reading fluency and attention criterion measures: a Comorbidity Deficit group, exhibiting both reading difficulties and attention deficits (CD; n=12), a Reading Disability group (RD; n=22), an Attention Deficit group (AD; n=25), and a Control group exhibiting no deficits (CnD; n=116). Children in each of the single-deficit groups performed within normal range on the other criterion measure. The four groups were identified in Grade 1 and they were compared from Kindergarten to Grade 2 on a set of cognitive (from the Cognitive Assessment System, Naglieri & Das, 1997), phonological processing, RAN, sensorimotor, word reading fluency and accuracy, orthographic processing, and passage comprehension measures. The effects of verbal and non-verbal ability, age, gender, and parental education level were controlled among the groups. Results: Results showed that the CD group exhibited greater dysfunction in the vast majority of the measures compared to the single deficit and CnD groups. Also, although the three deficit groups were not easily differentiated in Kindergarten, their differences were maximized in Grade 1 and retained in Grade 2. Particularly, the cognitive, phonological, naming, reading, and spelling deficits found in the CD group persisted across development. Conclusion: These findings suggest that comorbid cognitive deficits may actually hinder further the performance of children with reading disabilities.
Rauno K. Parrila (University of Alberta)Holly Stack-Cutler; Helene Deacon - Dyslexia in the university: What predicts word reading when phonological decoding fails?
Purpose. We examine how university students with a history of reading disabilities (RD) read and spell words. Word reading is highly correlated with phonological awareness (PA) and phonological decoding (PD) in most participants. However, university students with a history of RD can perform word reading tasks at a level significantly higher than predicted from their performance in PA and PD tasks (e.g., Parrila, Georgiou, & Corkett, 2007). We will examine first how common this dissociation is, and second, what processes are associated with unexpectedly high word recognition. Method. Participants include RD (n = 105) and normally reading (control, n = 100) university students. The dependent measures include word reading accuracy and fluency. Independent measures include (a) nonverbal and intelligence and vocabulary; (b) phonological sensitivity; (c) phonological decoding fluency and accuracy; (d) orthographic processing; (e) rapid naming speed; (f) morphological processing; (f) print exposure; and (g) ease of lexical access. Results. Data from RD participants has been collected, and data from control participants is being collected. We will use data from control participants as a criteria to identify strengths and weaknesses in RD participants. Conclusions. Our goal is to identify the mechanisms that allow at least some individuals to develop effective compensatory strategies at the word level. The RD sample in this study is larger than in any existing study so far and we expect to find several different profiles of strengths and weaknesses.
Derek Pasisz (Florida State University); Brooke Soden-Hensler; Chris Schatschneider - Genetic and environmental influences on reading comprehension: a research synthesis
Purpose Determining the etiology of a given behavior is fundamental in order to gain a full understanding of a given phenomena (Waldman, 2007). Relatively little is known about the interplay of genes and environment with academic achievement (Kovas, Haworth, Dale, & Plomin, 2007) and understanding the etiology of academic skills can potentially provide additional information to the effort underway to alleviate academic difficulties. This paper adds to previous research by conducting a meta-analysis to estimate the overall etiology of reading comprehension and identify any potential moderators. Method A literature search returned 220 potentially appropriate studies. After screening for the necessary statistical information, a total of N = 33 articles were used in this study. Results The results indicated that the etiology of reading comprehension is roughly half genetic and half environmental (one quarter shared, one quarter nonshared). Additional analyses revealed variables that moderate these findings including differences between research projects, participants' country, and the measures used. Conclusions This synthesis is consistent with previous findings that genetics have a substantial influence on reading comprehension (Harlaar, Dale, & Plomin, 2007), but indicates somewhat lower estimates. In turn, this brings to light the importance of environmental influences on reading. Studies have begun to uncover environmental aspects central to the development of reading (Petrill et al., 2007) and the results obtained here support these findings, which is one of the significant contributions of this paper. In addition, the finding that moderators can produce differences in etiological estimates is important information for individuals conducting and reviewing this research.
Adrian Pasquarella (Home Residence); Alexandra Gottardo; Amy Grant - Reading comprehension in adolescent first and second language learners: A comparison of simple and multi-component models
Purpose: The underlying components of reading comprehension were examined and compared in adolescents who spoke English as their first (L1) or second (L2) language. Method: One-hundred and nine adolescents (55 L1 and 54 L2) completed measures of reading comprehension, decoding, vocabulary knowledge, working memory, comprehension strategy use, motivation and print exposure in English. Results: Overall English L1 students outperformed English L2 students on measure of reading and language, with English L2 students performing below grade level on measures of comprehension and vocabulary knowledge. Examining models of reading comprehension between groups revealed that vocabulary knowledge is the best predictor of reading comprehension for both groups. In addition, decoding and working memory were significantly related to reading comprehension for English L2 students, while working memory and comprehension strategies were significantly related to reading comprehension for English L1 students. Furthermore, vocabulary knowledge mediated the relationship between motivation and print exposure with reading comprehension for English L1 students. For English L2 students, the relation between motivation and comprehension was mediated by comprehension strategies. For both groups comprehension strategies mediated the influence of decoding on comprehension. Conclusions: The applicability of using L1 models of reading with L2 populations, similarities and differences in the reading comprehension models, and implications for educational instruction, are discussed.
Hanno Pauly (German Institute for International Educational Research);Sven Lindberg; Jan Lonnemann; Janosch Linkersdörfer; Marcus Hasselhorn - Examining the relations between Rapid Automatized Naming (RAN) and reading / mathematical precursors in children at preschool age
Purpose: RAN has been proven to be a valid predictor of later reading and mathematical performance at school age. However, little is known about its relation to other specific precursors of mathematical and reading performance in preschool age. This study examines which specific RAN tests are related to reading / mathematical precursors at kindergarten. Method: 1352 kindergarten children were tested for RAN (quantities, numbers, letters and objects), precursors of mathematical and reading performance using multiple measures. Participants were all one year before school entry and age ranged from 53 to 85 months with a mean age of 70,15 months. Results: Factor analysis revealed RAN, reading precursors and mathematical precursors to constitute three correlated but still distinct factors. In a second step, correlation analysis suggested naming of quantities to be significantly related to mathematical precursors whereas naming of letters was significantly associated with reading precursors. Conclusion: Specific connections between RAN tests and reading / mathematical measures cannot only be displayed in school age but also in kindergarten age on a precursor level.
Riitta Pennala ()Kenneth Eklund, Jarmo Hämäläinen, Paavo H.T. Leppänen, and Heikki Lyytinen - Perception of phonemic length and its relation to reading and spelling skills in children with family risk for dyslexia
Purpose We examined the ability to discriminate phonemic length and the association of this ability with reading and spelling accuracy and reading speed in Finnish children at school age. Method Reading-disabled (RDFR, n = 35) and typically reading children (TRFR, n = 69) with family risk for dyslexia and typically reading control children (TRC, n = 80) were tested through grades 1-3. MANOVA was used for the comparisons of the development of discrimination ability, and hierarchical stepwise regression analyses were used to examine the variance explained by discrimination ability in reading and spelling accuracy after controlling for verbal short-term memory, phonological memory, naming speed, and intelligence. Results The RDFR group made more errors in phonemic length discrimination than the TRC group in grades 2 and 3. Discrimination ability explained unique variance of spelling accuracy in grades 2 and 3 and reading accuracy in grade 3 in the RDFR group. The association was particularly strong in spelling accuracy in the 2nd grade (ΔR2 = 0.21). At the individual level, in grade 2, 31.4 % of the RDFR, 14.7 % of the TRFR group, and 8.8 % of the TRC group performed below -1.25 standard deviations in the discrimination task. Conclusions Phonemic length discrimination was most challenging to dyslexic readers and it was related to reading and spelling accuracy. Problems in discrimination ability are indexing reliably compromised speech perception, and problems in this ability could be one of the accumulating risk factors affecting development leading to dyslexia among Finnish children.
Jill Pentimonti (The Ohio State University); Tricia Zucker; Laura Justice; Shayne Piasta; Joan Kaderavek; Amy Sofka - A standardized tool for assessing the quality of classroom-based shared reading: The SABR(Systematic Assessment of Book Reading)
Purpose: This study sought to describe the general psychometric characteristics of the Systematic Assessment of Book Reading (SABR), an observational tool designed to characterize the quality of classroom-based shared-reading sessions. Method: Participants included 105 early childhood and early childhood special education preschool teachers and their 4-year-old students. SABR was coded from two videotaped, whole-class shared-reading sessions collected from each teacher. Reliability (internal consistency, inter-rater agreement, test-retest) and validity (convergent) were examined. Results: Results from a series of factor analyses indicated that the SABR contained five construct factor scores and internal consistency was established with acceptable alpha levels for each construct (Language development (.73), Abstract thinking (.64), Elaborations (.74), Print/Phonological skills (.5), Session climate (.72)). Inter-rater agreement was generally high with interclass correlation coefficients ranging from .71 to .96 for each of the five constructs. Significant test-retest correlations ranging from .4 to .65 demonstrated that scores on the SABR, administered at two different time points, were consistently and moderately correlated with each other. Validity was established through determining that the SABR's five construct scores and comparable measures (i.e., Classroom Assessment Scoring System- PreK) were significantly and moderately correlated (rs from .24 to .46). Conclusions: Few reliable and valid tools are available to measure the quality of classroom-based shared-reading sessions. The SABR, which this study shows exhibits adequate psychometric characteristics, offers researchers, teachers, and allied professionals an important tool for documenting the quality of teachers' reading practices and, in turn, how these might be related to students' academic development.
Yaacov Petscher (Florida Center for Reading Research)Christopher Schatschneider - Contributions of Reading Comprehension, Text Reading Efficiency, and Spelling in Predicting High Stakes Outcomes in Grades 3-12
Purpose: The purpose of the study is to examine how students' performance on components of a state-wide screening and diagnostic assessment uniquely predict reading comprehension performance on the state's high stakes reading comprehension test (i.e., Florida Comprehensive Assessment Test [FCAT]). Method: Approximately 900,000 students in grades 3-12 from the state of Florida are included in the design. Students are assessed in the Fall, Winter, and Spring of the academic year on a newly released screening and diagnostic system called the Florida Assessments for Instruction in Reading. Students were screened for reading comprehension difficulties in a computer adaptive framework. If performance is less than a pre-specified cut-point, students are exposed to timed silent reading fluency passages (i.e., maze) as well as a computer adaptive spelling task. Both the reading comprehension and spelling tasks produce a developmental ability score that can be used for tracking growth. Similarly, the maze task produces an adjusted maze score, which accounts for both the amount of time it takes for students to complete the task, as well as the difficulty of the passage. Results/Conclusion: The developmental ability and adjusted maze scores were used in individual growth models to estimate the amount of change students made over the school year. The estimated student slopes, as well as the scores from the Fall assessment, were used in a dominance analysis. This analysis examined the amount of unique variance that intercepts and slopes from reading comprehension, spelling, and fluency contributed to the prediction of the FCAT. Keywords: reading comprehension; longitudinal
Linda Phillips (University of Alberta)Denyse Hayward; Stephen Norris - Inaccuracies in and confusions with the successful instruction of phonological awareness in preservice teachers' professional development textbooks
Purpose Successful instruction of phonological awareness (PA) relies on educator knowledge and studies have revealed educators lack knowledge to instruct in PA. Our purpose is to report results across 30 extensively used preservice teachers' professional development (PTPD) textbooks in North America for inaccuracies and confusions about PA. Method Thirty textbooks (2001-2009) were identified from leading publishing sites and cross-referenced with Google book and Amazon. Copies of sections devoted to PA and phonics instruction were procured. Confusions and inaccuracies were classified based on type of error. Results Seven of the 30 textbooks had no confusions or errors, however, four provided insufficient information for educators to instruct PA. Three of the 7 contained activities that required substantial knowledge of PA. The remaining 23 contained confusions and inaccuracies across three areas: grapheme-sound correspondences(inaccuracies occurred for individual letters (X - 12 texts; S - 4 texts); letter combinations (diagraphs - 12 texts; two letters making one sound - 4 texts); blending (consonant blends become one sound - 12 texts and non-consonant blends classified as consonant blends - 8 texts); and segmenting (confusing descriptions, instructions or examples in 23 texts across one of the following areas: (a) number of syllables in word, (b) segmenting syllables versus sounds, (c) determining syllable boundaries, (d) hierarchy of task difficulty, and (e) activities that focused on different skills. Conclusions Many textbooks confuse further and misinform educators. Discussion of these texts is critical in forums such as SSSR in order to prevent additional and unnecessary barriers to children's successful acquisition of phonological awareness skills.
Beth Phillips (Florida Center for Reading Research)Yaacov Petscher - Participation in Florida's universal pre-k program: predictors of completion and kindergarten literacy achievement
Purpose. Recent public policy and community interest supports the potential for state-supported pre-k to enhance the school readiness and early literacy skills of entering kindergarteners, positioning them to take full advantage of literacy instruction. Predictors of completion of and successful outcomes from participating in Florida's Voluntary Prekindergarten Program (VPK) are explored longitudinally from pre-k through the end of kindergarten. A primary research question is whether VPK completion is a significant predictor of kindergarten success once child and site level variables are considered. Method. A total of 481 four-year-old children representing 223 providers participated in VPK and then matriculated to a school participating in the field trial of new kindergarten screening (at entry: letter naming and blending sounds) and progress monitoring (DIBELS Letter Name Fluency (LNF) and Initial Sound Fluency (ISF). Of these, 77% (n = 369) completed the VPK program, whereas 33% (n = 112) did not. Data were collected on VPK program length (300 vs. 540 hours), provider type (public vs. private), provider school readiness ratings, school quality and child demographics (ethnicity, SES, and disability status). Results. Using cross-classified HLM and growth models with children nested in both VPK provider and kindergarten school, findings will indicate the unique and overlapping predictors of success on early literacy skills in kindergarten. Entry, end of year status, and growth analyses will be conducted. Conclusions. Findings will add to understanding of early predictors of kindergarten literacy acquisition and will contribute meaningfully to policy and instructional decisions regarding pre-k implementation in Florida and nationally.
Shayne Piasta (The Ohio State University); Yaacov Petscher; Laura M. Justice - Diagnostic efficiency of prekindergarten letter name knowledge benchmarks: Relations with first-grade literacy achievement
Purpose Various benchmarks for children's prekindergarten early letter name knowledge have been set forth by federal, state, and professional organizations. The present study investigated the adequacy of such benchmarks in predicting later risk for literacy-related difficulties. Method Data from 370 children contributed to this correlational study. Children were assessed on their (a) end-of-prekindergarten uppercase and lowercase letter name knowledge using the Phonological Awareness Literacy for Preschool and (b) end-of-first grade literacy skills using the Letter-Word Identification, Spelling, and Passage Comprehension subtests of the Woodcock-Johnson Tests of Achievement-III. Receiver-operating characteristic (ROC) curves and relevant indices were generated to examine the relations between various letter name knowledge benchmarks and later literacy achievement. Sensitivity, specificity, positive and negative predictive power, and the overall correct classification were used to examine the diagnostic efficiency of various benchmarks for lowercase and uppercase letter knowledge. Results Results indicated trade-offs in diagnostic efficiency for letter name knowledge benchmarks. Sensitivity, specificity, and overall correct classification indices ranged from .37-.90, .32-.88, and .41-.80, respectively, for uppercase letter name knowledge and.51-.98, .09- .81, and .22-.74 for lowercase letter name knowledge. Positive predictive power was generally low for all benchmarks and similar for uppercase and lowercase letter knowledge (range: .15-.60). Negative predictive power was generally high and again similar for uppercase and lowercase letter knowledge (range: .73-.98). Conclusion Findings are important for policymakers and early childhood professionals charged with establishing these standards and indicate that benchmarks ought to be selected based on the probability that attainment in prekindergarten assures later literacy success.
Meredith Pike (University of Guelph); Marcia Barnes; Alain Desrochers - Knowledge-based inference in 3rd to 8th grade children: cognitive underpinnings and relation to reading comprehension
Purpose: The ability to connect information in the text with prior knowledge (i.e., make knowledge-based inferences) is important for reading comprehension. Knowledge-based inferences can maintain causal coherence (coherence inferences), or they can expand on story content without being necessary for comprehension (elaborative inferences). The goals of this study were to determine: (1) if there are age-related changes in the ability to make knowledge-based coherence and elaborative inferences, (2) what cognitive competencies are related to making knowledge-based inferences, and (3) if knowledge-based inference skill is a unique predictor of reading comprehension. Method: 112 typically developing children, from grades 3 to 8, were administered the Knowledge-Based Inferencing Test in which they learned a novel knowledge base prior to hearing a related multiepisode story. They were then asked questions that required them to make both cohesive and elaborative inferences. Measures of vocabulary, word reading skill, attention, working memory, and reading comprehension were also administered. Results: The ability to make elaborative inferences improved with grade whereas coherence inferences were made at a high rate across all grades. Visual-spatial working memory, visual-perceptual abilities, and behavioural inattention were predictive of coherence inferences, whereas aspects of verbal working memory were predictive of elaborative inferences. Finally, knowledge-based inferencing was a unique predictor of reading comprehension after controlling for age, vocabulary, and word reading skill. Conclusions: Results are discussed in relation to cognitive and developmental models of comprehension.
Hannah Pimperton (University of Oxford); Kate Nation - Inhibitory deficits in poor comprehenders
Purpose. Poor comprehenders are children who show significant deficits in their reading comprehension performance despite having age-appropriate reading accuracy. Previous research has suggested that poor comprehenders have weaknesses in cognitive inhibition. This series of studies aimed to explore poor comprehenders' inhibitory skills at two different levels of Nigg's (2000) taxonomy of inhibition; cognitive inhibition and interference control, in order to provide a clearer picture of any inhibitory deficits. These studies also sought to determine the effect of manipulation of the type of stimuli used on inhibitory performance. Method. We compared the performance of poor and skilled comprehenders on a series of tasks that required them to inhibit an irrelevant dimension of a stimulus (interference control), as well as on tasks that required them to inhibit irrelevant information to protect working memory (cognitive inhibition). Both verbal and non-verbal stimuli were used, and the semantic demands of the verbal stimuli were also manipulated. Results. Poor comprehenders showed deficits relative to skilled comprehenders in both the levels of inhibitory control that we studied. Furthermore, manipulating the verbal vs. non-verbal nature of the stimuli, as well as the semantic demands of the verbal stimuli, produced a distinct profile of deficits in the poor comprehenders. Conclusions. Poor comprehenders show inhibitory deficits across two different domains of inhibitory control; interference control and cognitive inhibition. The nature of the stimuli that they are required to inhibit affects the degree of deficit shown by the poor comprehenders relative to the skilled comprehenders.
Kristina Pineau (Florida State University); Kathie Moe; Christopher Lonigan; Laura Lang - The diagnostic accuracy of oral reading fluency progress monitoring for predicting primary-grade students' proficiency on the California Standards Test
Purpose - This study investigated an empirically-based Grade 1 beginning-of-year Words Correct Per Minute (WCPM) benchmark. Analyses examined the diagnostic accuracy of the Reading Lions (RL) oral reading fluency progress monitoring tool for predicting students' proficiency on the California Standards Test (CST). Information gathered from students' performance on RL guides decision-making regarding reading intervention. Method - The sample consisted of 595 Grade 1 students from seven schools in one California district. Sensitivity and specificity were calculated to assess the diagnostic accuracy of RL. Receiver operating curve (ROC) analyses were used to derive the area under the curve (AUC). Results - Sensitivity and specificity were calculated based on CST levels ≥ 3 (Basic). A score of 19 WCPM yielded test sensitivity of .70 and 1-specificity (false-positive rate) of .13. A score of 47 WCPM would allow the district to minimize the false-positive rate with a sensitivity of .21. ROC analysis showed that RL performed significantly above chance at the Basic CST level. The estimated AUC was .86 (SEM= 02, p<.001). Conclusions - Based on these results, to minimize the false-positive rate, the district would need to set a beginning-of-year benchmark of 47 WCPM for Basic performance on the CST. Provided that the district can accommodate a higher than zero false-positive rate, a less stringent beginning-of-year benchmark may be useful. These results suggest that RL is meaningful for guiding district decision-making regarding reading intervention. Analyses are ongoing, and next steps include validation of the Grade 1 mid-year suggestion of 23 WCPM (Hasbrouk & Tindal, 2006).
Hanne Poelmans (ExpORL, Dept. of Neurosciences, K.U.Leuven, Belgium); Heleen Luts; Bart Boets; Maaike Vandermosten; Pol Ghesquière; Jan Wouters - Auditory temporal envelope processing deficit in dyslexia: An auditory steady-state responses study
Purpose: Developmental dyslexia refers to a deficiency in reading and writing skills, caused by a deficit in the development of phonological skills. It is hypothesized that the phonological problems in dyslexia result from a fundamental deficit in low-level auditory temporal processing and is related to a subtle deficit in speech perception. The aim of the study is to trace at what level of the auditory system (brainstem or cortex) possible neurophysiological differences in temporal envelope processing between normally reading and dyslexic subjects occur. Method: In this study, temporal envelope processing was investigated in a group of 31 normally reading and 31 dyslexic adults by means of auditory steady-state responses (ASSRs), in combination with cognitive and psychophysical measures. Multichannel ASSRs were evoked by speech-weighted noise stimuli, 100% amplitude-modulated at 20 and 80 Hz. These modulation frequencies represent peripheral (80 Hz) and cortical (20 Hz) processing. The 20 Hz modulation is important for speech envelope processing; whereas the 80 Hz modulation represents more basal processing. Results: Compared to the matched control group, dyslexic subjects showed different processing at 20 Hz, whereas no differences were found at 80 Hz. Conclusion: This study found evidence for a non-speech-specific, temporal processing deficit in dyslexic adults. This deficit is present at the level of the cortex, whereas peripheral processing is not different from the control group.
Sharolyn Pollard-Durodola (University of Denver, Child, Family, and School Psychology);Jorge Gonzalez; Deborah C. Simmons; Aaron B. Taylor; Matthew J. Davis; Leslie Simmons; Miranda Nava-Walichowski - Analyzing the impact of curriculum implementation, professional development, and instructional cues on bilingual preschool teachers' shared book reading practices
Purpose - This exploratory observation study examined the shared book reading practices (interactive dialogues, general and vocabulary-specific instruction) of seven bilingual preschool teachers and to what degree curriculum implementation, professional development, and instructional cues changed their teaching quality. The research questions are: 1.What is the entry-level quality of shared book reading practices (general and vocabulary-specific instruction, interactive dialogues) prior to curriculum implementation/professional development and instructional cueing? 2. What is the impact of instructional cues and curriculum implementation/professional development on teaching quality (general and vocabulary-specific instruction, interactive dialogues)? 3. Does the entry-level quality of shared book reading change after participating in a 15-week curriculum implementation/professional development and instructional cueing experience? Method - Teachers were observed at two points prior to curriculum implementation/professional development (pretest uncued and cued), and once post curriculum implementation/professional development (delayed posttest cued). To describe patterns of instructional change, observation data were analyzed by (a) a categorical approach (overall and follow-up exact tests) and (b) comparing the means of individual teachers' shared book reading practices by instructional domain (interactive dialogues, general and vocabulary-specific instruction) and observation occasion (pretest uncued, pretest cued, delayed posttest cued). Results - Findings indicated different levels of response to a multi-dimensional (cue + curriculum implementation/professional development) approach with notable differences in the quality of instructional practices characterized by distinctive teacher performance profiles. For some teachers, instructional cues were sufficient to improve book reading quality while others required more intensive support. Conclusions - Implementing a multi-method approach to support/scaffold teachers' knowledge and use of effective shared book reading practices may improve teachers' instructional practices and better prepare teachers to accelerate vocabulary knowledge while reading books out loud.
Tatiana Cury Pollo (Washington University in St Louis); Rebecca Treiman; Brett Kessler; Emily Rosenzweig - Children's initial spelling strategies: 'Bear' is not bigger than 'mosquito'
Purpose: It has been asserted that beginning writers use a referential strategy rather than a phonological one (e.g., Ferreiro & Teberosky, 1982). For example, children may use more letters when writing 'bear' than 'mosquito' because they believe the number of letters should reflect the size of the referent rather than the length of the pronunciation. However, to date, most evidence for this strategy is anecdotal and not rigorously tested. In this study we investigated the strategies that children use in their earliest writing. Method: We studied 91 English-speaking children (mean 4 years 7 months). Monte Carlo techniques identified subgroups who made significant use of phonologically appropriate letters (14 children) or did not (56). Children spelled 24 words of different lengths that represented small objects (e.g., bug, mosquito) or large ones (bear, dinosaur). We asked whether the prephonological writers would use more letters to spell bigger objects, ignoring word length, and whether phonological spellers would evince the opposite pattern. Results: Children who used some phonology in their spellings wrote more letters for longer words (p < .001), whereas the prephonological group did not (p = .808). However, neither group consistently used more letters for bigger objects (p = .237 for the prephonological writers). Conclusions: We found no evidence of referential strategies even among children who had not yet learned to use letters to represent sounds. The referential strategy adduced by previous studies might not describe how the majority of prephonological children spell in controlled experiments.
Heather Poole ()Betty Ann Levy; Beverly Ho - Testing Raney's (2003) model of text processing: effects of text-base and situation model overlap.
PURPOSE. Two experiments tested the relationship between representation of the text base and the situation model, following Raney's (2003) model of text representation. We varied overlap in text propositions between two successively read passages to examine retrieval of the text base of the first passage during reading of the second passage. Orthogonally, we manipulated the theme of the two passages (same or different situation model). The question was whether degree of overlap of text propositions and of situation models would have independent or dependent effects on perceived similarity and reading time of passages. METHOD. Experiment 1 varied situation model overlap and the number of repeated sentences between passage pairs. University students rated the similarity of passage pairs on a Likert-type scale. Experiment 2 measured effects of these manipulations on reading time of repeated text propositions in the second passage. RESULTS. Experiment 1 showed that readers' ratings of the similarity of passages increased with the number of sentences the two passages shared. This pattern was true for passage pairs that had similar or different situation models. In Experiment 2, reading time for the second passage decreased as the number of sentences repeated from the first passage increased. This effect was independent of overlap in situation models. CONCLUSIONS. These findings suggest that the situation model and text base have separate effects on the repetition benefit. Our findings are incompatible with the strong assumption of Raney's (2003) model, which states that surface form and text base are bound to a well-formed situation model.
Mads Poulsen (University of Copenhagen)Holger Juul; Carsten Elbro - Does ability to establish symbol-sound pairings mediate the RAN-reading relationship?
Purpose - Performance on tests to rapidly name letters and digits has been shown to correlate with reading. One possible reason is that these tests probe the ability to learn and automatise symbol-sound associations. However, most studies have not controlled for the amount of experience with the RAN-items, so it is unclear whether it is the experience or the ability to take advantage of the experience that is responsible for the correlation between RAN and reading. Paired associate learning tasks have been shown to differentiate dyslexics from controls, and to correlate with reading in unselected samples. We tested whether the ability to learn and/or established letter knowledge mediated the correlation between RAN and reading in a sample of preschool beginning readers. Method-105 preschoolers were administered traditional reading, letter knowledge, and RAN tests, and taught novel names for animals in a paired associate learning task. These animals were then used in a rapid naming task. Results - Preliminary results show that reading correlated with the amount of training required for learning the animal names (r=-.19, p=.06). RAN speed with the same animals did not correlate with reading, but the number of error produced during this task did (-.27, p<.01). None of the PA-measures mediated the relationship between a traditional RAN and reading (r=-.36), but controlling for letter knowledge did. Conclusions - The results indicate that established letter knowledge to a larger degree than the ability to learn sound-symbol associations is responsible for the RAN-reading relationship in early reading development.
Daisy Powell (Institute of Education, University of Reading); Maureen Dixon; Lynette Chesson - Does SMS text messaging have a negative effect on knowledge of standard spelling?
Purpose: In recent years the sharp rise in mobile phone usage, particularly text messaging, in school-aged children has raised fears of negative consequences of such technology for literacy. The current research used a paradigm developed by Dixon And Kaminska (1997), who showed that exposure to phonetically plausible misspellings (e.g. recieve) negatively affected subsequent spelling performance, though this was true only with adults, not children (Dixon & Kaminska, 2007). The current research extends this work to directly investigate the effects of exposure to both SMS text abbreviations (textisms) and misspellings in adults and 12-14 year-old children. Method: Spelling of a set of key words was assessed both before and after an exposure phase where participants read the same key words, presented either as SMS abbreviations (textisms; e.g. 2nite), correctly spelled (tonight) or misspelled (tonite) words. Results: Adult sample: Spelling scores decreased from pre- to post-test following exposure to misspellings, whereas performance improved following exposure to correctly spelled words and, interestingly, to textisms. Child sample: Similar results were obtained in the child sample. Unlike Dixon and Kaminska (2007) we also found a trend towards a negative effect of exposure to misspellings, but there was no effect of textisms. Conclusions: Data suggest that exposure to textisms, unlike misspellings, does not impair knowledge of standard spellings, and indeed has a positive effect on adults' spelling. These findings are interpreted in light of other recent research (e.g. Plester et al., 2009) suggesting a positive relationship between texting and some literacy measures in school-aged children.
Liron Primor (); Margaret E. Pierce; Tami Katzir - Predicting Reading Comprehension of Narrative and Expository Texts among Different Profiles of Readers
Purpose Reading comprehension is the ultimate goal of reading development and as such, has been extensively studied. However, the interaction between text genre, reading proficiency, and reading comprehension has yet to be sufficiently addressed. Therefore, the aim of the current research was to investigate which cognitive and linguistic skills make independent contribution to reading comprehension of narrative and expository texts among 4th grade Hebrew readers. Another aim was to compare the relative contribution of different components to reading comprehension among different reading profiles. Method The study was based on a national database, from which 190 readers with RD and 190 readers with no RD were selected. IQ, decoding, reading comprehension, and various linguistic and cognitive skills were assessed and SEM were constructed. Results Results exhibited differences in reading comprehension predictors between the two genres and among the different reading profiles. Both lower level processes (such as decoding and orthography) and higher level processes (such as verbal IQ and Raven matrices) predicted narrative and expository reading comprehension among all readers to different extents. In addition, a significant interaction was found between reading group and genre. Conclusions This study adds to the existing literature on reading comprehension, supporting the shared and unique components that are related to each genre in different populations of readers. In addition it suggests a language specific model of reading comprehension in which different linguistic and cognitive measures contribute to reading comprehension based on the orthography that is being read.
Athanassios Protopapas (University of Athens); Katerina Fakou; Stella Drakopoulou; Christos Skaloumbakas; Angeliki Mouzaki - Spelling errors of Greek dyslexic and nondyslexic children in Grades 3-4 and 7-8
Purpose: To propose a classification system for spelling errors and determine the most common spelling difficulties of Greek children without and with dyslexia. Method: Spelling skills of 566 children from the general population and 61 children with dyslexia, in Grades 3-4 and 7-8, were assessed with a dictated common word list and an age-appropriate passage. Spelling errors were classified into broad categories, including phonological (graphophonemic mappings), morphological (inflectional suffixes), etymological (word stems), stress assignment (diacritic), and punctuation. Errors were further classified into specific subcategories. Chronological-age and reading-age control groups were formed by selected nondyslexic children. Results: Relative proportions for a total of 15066 errors were derived by calculating the opportunities for each error type. Nondyslexic children of both age groups made primarily morphological and stress errors, followed by etymological errors. Phonological and punctuation errors were negligible. Most frequent specific errors were in derivational affixes, stress diacritics, inflectional suffixes, and vowel historical spellings. Older children made fewer errors, showing greater improvement in morphological spelling. Dyslexic children differed from nondyslexic in making more errors of the same types, in comparable relative proportions. Older dyslexic children differed more from nondyslexic children in error counts but not in error types, and made more errors than reading-matched younger children. Conclusions: Most spelling errors of both dyslexic and nondyslexic children indicate persistent difficulty with internalizing regularities of the Greek orthographic lexicon, including derivational, inflectional, and word (stem) families. This difficulty is greater for children with dyslexia. No qualitative differences emerged in dyslexic spelling patterns.
Cyntia Puranik (University of Pittsburg);Stephanie Al Otaiba; Jessica Folsom; Luana Greulich - The effect of writing instruction on writing productivity in kindergarten children
Purpose: This study examined the relation between classroom writing instruction and students' writing performance. Method: Spontaneous end-of-year writing samples were collected from 242 students attending kindergarten in 12 different schools. Students' writing was further examined using letter fluency and spelling tasks. Classroom instruction was observed, videotaped, and coded for writing instruction. Based on their average writing productivity, classrooms were divided into three groups (low, mid, and high) to compare observations of instructional practices. Instructional practice variables coded included: amount of time students watched teacher write, amount of time spent on teacher-modeled writing, amount of time students spent writing stories, amount of time spent in handwriting instruction, amount of time spent by teacher providing feedback on children's written work. Results: Large differences were noted among classrooms for productivity; much less differences among classrooms for spelling. Overall, there was very little writing instruction observed in the classroom. Students who showed the most writing productivity came from classrooms where the student spent more time on writing activities, receiving handwriting instruction and watching the teacher write. Preliminary analysis indicates that code-focused instruction directly impacts spelling but productivity in writing requires more than code-focused instruction. Conclusions: Results with these younger students are consistent with conclusions drawn by Moats, Foorman et al. (2006) for fourth graders that quality of writing instruction impacts the length of students' writing composition. The classroom environment and models provided by teachers appear to be associated with students' writing skills starting as early as kindergarten. Implications will be discussed.
Pauline Quemart (Centre de Recherche sur la Cognition et l'Apprentissage)Severine Casalis - Effects of orthographic and phonological opacity in the processing of morphologically complex words : A masked priming study in children from grade 3 to 7
Purpose- Masked priming studies conducted in adult expert readers (Rastle and Davis, 2008) and recently in developing readers (Quemart and Casalis, in preparation) indicate that morphological processing is triggered by the orthographic properties of morphemes at the earliest steps of word reading. Our study was designed to examine to what degree children's morpho-orthographic representations are specified: at a strict letter level, or at a more abstract (lexical) level. For this purpose, we examined whether morphological decomposition occurs when stems are orthographically (abundant - abound) and/or phonologically (natural - nature) altered. Method- We asked French-speaking children from grade 3 to 7 to perform a lexical decision task associated to a masked priming paradigm (60 ms prime duration). Three prime-target relationships were included in the experiment: - Morphological without alteration: Fillette - FILLE (little girl - girl). - Morphological with phonological alteration: Jardinage - JARDIN (gardening - garden). - Morphological with phonological and orthographic alterations: aviateur - AVION (pilot - plane). An orthographic control condition was added: Abricot - ABRI (apricot - shelter) where "cot" is not a suffix in French. Results- Morphological priming effects were shown only in the "morphological without modification" condition in children from grade 3 to 5. However, children in grade 7 benefited from the prime independently of the phonological and/or orthographic modifications of the stem. Conclusion- Orthographic and/or phonological alterations of stems prevent morpho-orthographic decomposition to occur in youngest children. However, morpho-orthographic representations of 7th graders are more flexible, and established well enough to capture relationships that might have been obscured by differences in orthography and phonology.
Ralph Radach (Florida State University);Christian Vorstius; Ronan Reilly - The science of speed reading: Exploring the impact of speed on visuomotor control and comprehension.
Purpose: Basic research suggests that there may be some scope for faster reading, providing a potential base for at least a moderate amount of 'speed reading'. This study focuses on individual changes in visuomotor control, word processing and comprehension associated with increased speed of reading. Methods: Readers were trained over six sessions including pre and post testing. One group received non-specific instructions to increase speed in steps of 20 percent, while the other group used specific speed reading techniques. This included avoiding regressions back to earlier words (implemented via eye movement contingent acoustic feedback) and visually highlighting word groups as meaning units. Comprehension questions and speed feedback followed each passage. Results: Participants almost doubled their speed over the four training sessions. Eye movements changed quite dramatically, including a sharp reduction in refixations and frequent skipping of short words. In the specific training group, regressive saccades back to earlier positions in the text dropped by 50 percent, indicating that this aspect of the training was very effective. Overall, the doubling of reading speed was achieved without much loss in comprehension. Surprisingly, the group receiving specific speed training interventions had no advantage over the group with unspecified individual fluency training. Conclusions: Skilled readers can almost double their speed without losing text comprehension after a modest amount of training. Critically, the increase can be achieved via deliberate practice in the absence of any specific speed reading techniques. Implications of this remarkably large intra-individual variability for theories and models of reading will be outlined.
Gloria Ramirez (Thompson Rivers University);Xi Chen-Bumgardner; Yang C. Luo - How simple is the view of reading for ELLs?
Purpose: This study tested the Simple View of Reading (SVR) in three language groups: Spanish-speaking ELLs, Chinese- speaking ELLs, and English monolinguals. It tested whether for the three groups reading comprehension could be mainly explained by word recognition and language skills or if additional factors played an important role for ELLs. Method: The reading comprehension of 260 fourth graders and seventh graders (78 English monolinguals , 90 Spanish-speaking ELLs, and 76 Chinese-speaking ELLs) was examined. Three regression models were run to test the independent contribution of the SVR as assessed by word reading, receptive vocabulary knowledge, and morphosyntactic skills over and above non-verbal ability, memory and phonological awareness. Results: The pattern of results was similar across the three groups with word recognition and language skills explaining the larger proportion of variance on reading comprehension. However, differences were identified in the proportion of variance contributed by the SVR from group to group, the involvement of other factors such as phonological awareness, and the individual contribution of the separate components of the SVR (word recognition, vocabulary, and morpho-syntactic sensitivity). Conclusions: Results suggest that the SVR accurately describes reading for ELLs from different L1 backgrounds and therefore can be used as a reference model for instruction (e.g., importance of word decoding and language skills) and assessment.
Anne Rau (University of Tubingen, Germany); Karin Landerl - Reading acquisition in English and German - development of the word length and the word frequency effect
Purpose: Our study sets out to compare reading development in English and German in a sentence-reading eye tracking paradigm. Given the low orthographic consistency of English, systematic grapheme-phoneme conversion is less effective for English-speaking beginning readers than for children learning to read the regular German orthography. We therefore expect reading strategies to differ between English and German-speaking beginning readers. Specifically, this study investigates the orthography-specific development of the word length and the word frequency effect. German primary school children are expected to show a stronger word length effect whereas English-speaking children are expected to show a stronger word frequency effect. No differences are expected between adult readers of the two languages. Method: Participants include German and English-speaking readers of different age groups (grades 2, 3 and 4 and young adults). The reading task consists of 72 sentences nearly identical in the two languages (e.g. Lilly's hamster Birsul is hungry / Lillys Hamster Birsul ist hungrig). Each sentence comprises a target word manipulated with regards to word length (3 to 10 letters) and word frequency (high and low-frequency words and nonwords). Results: Data for German readers have already been collected. Analyses show standard effects of word frequency and word length (especially for nonwords) for central eye-tracking parameters (first fixation duration, gaze duration, number of fixations). We consider the item set appropriate for the intended cross-linguistic comparison. English data are currently being collected. Conclusions: Findings will be discussed in the context of reading models that consider cross-linguistic differences (orthographic depth hypothesis, grain size theory)
Dominique Rauch (DIPF, German Institute for International Enducational Research); Johannes Hartig - A differential analysis of abilities relevant for answering open-ended test items versus multiple-choice test items in reading comprehension assessment
Reading comprehension is typically assessed with a combination of multiple-choice items and open-ended items. This leads to multidimensionality of test performance, since multiple-choice items focus mainly on basic reading proficiency aspects (e.g. the extraction of local information), whereas open-ended items assess higher reading proficiency aspects (e.g. comprehension of main ideas and interpretations of texts). We analyze data from a German reading comprehension test of a nationwide assessment of language competencies in Germany. A two-dimensional IRT model with a basic reading proficiency factor common to all items and an additional factor specific to open-ended items (representing higher reading proficiency aspects) is more appropriate than a unidimensional model. Including gender, language spoken at home and general cognitive abilities as predictors in latent regression models reveals the diagnostic benefit of the two-dimensional model: The better performance of girls compared to boys is not due to basic reading proficiency aspects but to higher reading proficiency aspects. Non-native speakers of German perform worse than natives on both dimensions, yet the effect of language spoken at home is twice as high for higher reading proficiency aspects. Conversely, general cognitive abilities are strongly related to basic reading proficiency aspects but only weakly to higher reading proficiency aspects.
Dorit Ravid (Tel Aviv University)Rachel Schiff; Ronit Levie - Spelling in Hebrew-speaking dyslexic and non-dyslexic university students
Purpose: Spelling acquisition and processing in Hebrew is dependent on a complex of interrelated lexical, phonological, morphological and orthographic domains. Reduced sensitivity to cues in these domains may be detrimental to reading and spelling. Method: The current study investigated spelling skills in 36 Hebrew-speaking dyslexic university students compared with 36 normally-reading peers. Two spelling tasks were administered to the participants. One, a dictation task, focused on homophonous function letters denoting grammatical and lexical functions such as past tense or abstract noun. Participants heard each target word in the context of sentence to avoid ambiguity. The target homophonous letters were analyzed in terms of their morphological function and position in the word (beginning or end). A second task consisted of pairs of homophonous words differing in one root letter. Participants heard the target word in the context of a sentence and had to decide which member of the pair appeared in writing. Results showed that the dyslexic students scored significantly less well on both tasks than their normally-reading peers and also took more time on lexical decision. Moreover, they did particularly worse on those function and root letters which are less frequent and occupy less salient positions in the Hebrew word, such as K כ denoting second-person singular marking or vowel letters such as A א. This was apparent in both spelling tasks. Conclusion: These findings indicate that dyslexic undergraduates are less able to analyze written Hebrew words into morphemes and make use of this analysis for spelling.
Keith Rayner (University of California, San Diego); Jinmian Yang; Monica S. Castelhano; Simon P. Liversedge - Eye Movements, the Perceptual Span, Preview Benefit, and Disappearing Text Effects in Older and Younger Readers
Purpose: Recent experiments examining the nature of eye movements of older readers will be presented. Method: A series of experiments will be described in which the following paradigms were used to compare older and younger readers: (1) the moving window paradigm, (2) the boundary paradigm, and (3) the disappearing text paradigm. Results: The results indicate that (1) older readers have a slightly smaller and less asymmetric perceptual span compared to college aged readers, (2) preview benefit is slightly attenuated in older readers as on some fixations they obtain less preview benefit than younger readers, and (3) it takes older readers longer to encode the fixated word than younger readers, although, for both groups, how long the eyes remained in place was strongly influenced by the frequency of the fixated word (even though it was rapidly replaced by a mask). Conclusions: This is very strong evidence that for both older and younger readers cognitive/lexical processing is the primary determinant of when to move the eyes in reading. Older readers also apparently adopt a different reading strategy than younger readers
Susan J. Rickard Liow (Home); Shi Min Chua; Sara Da Silva Ramos; Stephanie Yeong - Spelling: Identifying bilingual children at-Risk using a single measure
Purpose: Tests of phonological awareness, rapid naming and letter knowledge are often used to predict future reading performance in English-speaking kindergarten children. These tasks are individually-administered, and may be less reliable for some bilingual populations (e.g., Jongejan et al, 2007). This paper describes how scores on a group-administered spelling test, can identify Year 1 Mandarin-English bilingual children who are at-risk of reading difficulties in Year 2. Method: 110 children (n= 52 English-L1; n=58 Mandarin-L1, aged 5-6 years) completed WRAT-4 Spelling Test, phonological awareness tasks (phoneme and syllable deletion), rapid naming (digits) and receptive vocabulary tests in English, as predictor variables at Time-1, and WRAT-4 Reading Test at Time-2. Results: Children were classified as at-risk using WRAT-4 reading if their scores fell within the lowest quartile for their language background. Logistic regression analysis, with all four measures in the model, showed that spelling (p = .034, Exp(B) = 1.487) and phonological awareness (p = .048, Exp(B) = 5.788) significantly predict at-risk status of bilinguals over and above the either rapid naming or letter knowledge. More importantly, the ROC analysis revealed that screening with WRAT-4 Spelling alone has a Relative Risk (RR) of .669, marginally lower than for all four predictors (RR =.710), but better than either phonological awareness (RR= .605) or rapid naming (RR=.596) alone. Conclusions: Group screening through spelling is a reliable method for teachers to identify children at-risk of reading difficulties in large bilingual kindergarten classes provided the word-list captures the salient phonological and orthographic knowledge.
Jessie Ricketts (University of Reading, UK)Fiona Duff; Jackie Masterson; Morag Stuart - Semantics and word-level reading: Does the relationship change with age?
Purpose: Two cross-sectional studies address the proposal that semantic knowledge contributes to word recognition, especially the reading of words that contain irregular spelling-sound mappings (e.g., Ouellette & Beers, in press). So far, few studies have taken a developmental approach. Further, many studies have used an oral vocabulary measure as a proxy for semantics. The present study takes a novel developmental approach and uses two measures of semantic knowledge. Method: In Study 1, a group of 478 children aged 4-12 years completed a standardised measure of oral vocabulary and read 30 nonwords, 30 regular words and 30 irregular words. In Study 2, an additional sample of 75 children aged 6-11 years read 20 regular and 20 irregular words and completed two semantic tasks; they provided oral descriptions of words, and the relationships between words. Results: Regression analyses showed that performance on both semantic tasks predicted word recognition after controlling for age and decoding skill. The proportion of variance explained by semantic knowledge was relatively consistent across both word type (irregular vs. regular) and age group. Nonetheless, there was evidence that the relationship between semantic knowledge and reading was weaker in younger children. Conclusions: These studies build on previous research by demonstrating a relationship between semantic knowledge and word recognition across a wide age range. Further, they indicate that the relationship is weaker for children at the beginning stages of learning to read. The findings will be interpreted within the context of models of reading acquisition (Plaut et al., 1996; Share, 1995).
Mabel Rivera (University of Houston)David Francis - Teaching Literacy to English Language Learners: What Have We Learned?
Purpose Over the last decade the study and validation of effective ways to teach literacy to English Language Learners has increased. The purpose of this study was to collect and disseminate knowledge of effective reading and writing interventions, for ELLs in grades K through 12. Method We examined evidence of enhanced instruction in elements of literacy for ELLs performing below grade level. This information was culled from a variety of sources, including major literature reviews and single studies and summarized in a literature review. Results The findings from this review indicate that assessment plays a crucial role in the educational experience of all ELLs. Assessments inform the process of educational placement, which in turn influences instruction and academic outcomes. ELLs need comprehensive, early, explicit, intensive interventions that reflect a close match between their source(s) of difficulty and the intervention. In addition, high-quality teacher preparation and professional development programs on teaching ELLs should address theoretical knowledge and pedagogical methods on second language acquisition. Conclusions Meeting the needs of ELLs requires a multi-dimensional strategic approach, with significant attention to three areas: 1) assessment, 2) intervention programs or models, and 3) professional development. The findings of this study demonstrate the significant positive impact that effective reading and writing interventions may produce, particularly for ELLs with low performance outcomes. Finally, teachers with strong content knowledge and specialized training have a positive impact on student performance.
Andrew Roach (Georgia State University); Tamika LaSalle; Dawn McGrath - The relationship of IEP quality to reading/language arts access and outcomes
Purpose: This paper will present results from an investigation of the quality of IEP goals developed using a web-based IEP tool. Method: Approximately 150 teachers completed two rating scales regarding the curriculum and instructional experience and current level of performance of a one of their students in special education. Teachers also completed a survey evaluating use of a web-based Individualized Education Plan (IEP) program. Quality of IEPs developed with the web-based program was then evaluated. Results: Findings will address the relationship between IEP quality, access to standards-based grade level content and subsequent performance on large-scale reading/language arts assessments. Data also will be presented regarding the relationship between teacher experience and training and the acceptability of the web-based IEP tool. Conclusions: Attendees will develop an increased awareness about how IEP quality may support curricular access and student performance in reading/language arts. We also will discuss the implications of using IEP quality as an index of curricular access for students with disabilities who participate in alternate assessment based on modified achievement standards.
Fiona Roberts (University of Oxford Department of Education)Kathy Sylva - Training parents to support their children's reading at home: a randomised control trial
Purpose: This paper reports a randomised control trial of a parent-training programme aimed at improving reading and behaviour in 5-6 year old children. The focus here is the literacy component of the programme, which aims to help parents support their children's reading at home. Methods: The children whose families are taking part in the study are from two disadvantaged areas in England and are 'at risk' of learning and behaviour problems. The SPOKES (Supporting Parents on Kids Education in Schools) intervention combines two treatments: a behaviour programme, alongside a programme designed to train parents in ways they can support their children's reading development. A previous trial of this two-pronged intervention compared the effects of the combined programme (reading and behaviour) on parents receiving the intervention versus a control group. This current study disaggregates the two treatments and compares them individually and jointly with a no-treatment control. Pre- and post- literacy assessments are used to investigate the effects of the intervention on children's word reading, phonological skills and reading comprehension. Results: Results demonstrated significant improvements in both behaviour (Parent Account of Child Symptoms) and literacy scores (single word reading and phonological skills). Conclusions: The robust RCT design of the original trial has provided evidence that a combined programme was effective in supporting parents with their children's reading development. The new trial disaggregates the effects of the literacy versus the behavioural programmes, and also demonstrates that experimental programmes in a single site can be replicated in everyday local government services.
Theresa Roberts (California State University, Sacramento) - Levels and patterns of kindergarten literacy performance under different instructional conditions
* Purpose - This study compared the levels and patterns of kindergarten literacy when children participated in structured and explicit instruction with instruction where skills were embedded in a meaning-focused program. It was hypothesized that instruction would shape patterns of literacy development: explicit instruction would lead to more clearly differentiated patterns of development that reflected both theoretical relationships amongst variables and instructional emphasis. * Method - Participants were 125 kindergarten children from low SES families whose primary language was English, Spanish or Hmong. Children were drawn from all 5 kindergarten classrooms in one school in each of two subsequent years. Letter names and sounds, phonological awareness, word recognition, and oral language achievement were measured. 1/2 the children participated in explicit instruction and the other half participated in embedded instruction. * Results - There were no significant differences between children on age, English oral proficiency, gender or language in the two instructional groups. Children of all three language who received explicit instruction performed significantly better on all measures than comparison children: F (6, 114) = 28.12, p < .001. Effect sizes ranged from 1 to almost 2. Explicit instruction was associated with the emergence of clear alphabetic and phonological awareness factors while comparison instruction revealed only a clear association between English proficiency and rhyming. * Conclusions - Explicit instruction led to significantly greater alphabet, phonological awareness, and beginning word recognition achievement for children from three ethnolinguistic groups. Type of instruction was associated with differences in patterns of literacy development. The theoretical importance of instructional context on achievement and developmental trajectories is highlighted.
Sarah Robins (Philosophy-Neuroscience-Psychology Program, Washington University in St. Louis); Rebecca Treiman - Learning about writing begins informally: How parents and children talk about letters
Purpose: Acquiring letter knowledge is important for the development of literacy. Our research explores how children and parents talk about letters, and how it changes over the preschool years. Method: We searched the CHILDES database, which contains child language transcripts from many researchers, for use of 'letter' and individual letters. We focused on interactions between U.S. English-speaking parents and children from one to five years of age. Results: Parents' and children's talk about letters often involved physical attributes of letters such as size, shape and color. The focus on shape increased at older ages. Talk about letters' sounds was infrequent throughout. From age 1 to 2, parents and children were more likely to talk about letters at the beginning of the alphabet than at the end, and a high proportion of the letter sequences uttered were alphabetic. At this early age, the frequency of a letter in the language did not determine how much parents and children talked about it. By age 4 to 5, letter frequency was a significant contributor in addition to alphabet position. Both parents and children mentioned letters in association with the child's name, linking letters more often with the name than with any other word. The relative importance of the child's name was smaller for older children. Conclusion: Looking at talk about letters is a valuable way to study emergent literacy. It offers information about what young children know and about how they learn it, information that would be difficult to glean in other ways.
Louise Roenberg (Department of Education, University of Aarhus); Dorthe Klint Petersen - The relationship between component skills, reading experience, and reading comprehension in Danish 3rd graders
Purpose The main aim of this study was to investigate the contribution of aspects of vocabulary, word reading abilities, and reading experience to reading comprehension, and to analyse sub-samples of students with comprehension difficulties. Method The study employed a cross-sectional design. Full data sets were obtained from 224 Danish Grade 3 pupils including minority pupils. Participants were given a standard reading comprehension test requiring multiple-choice answers to six different texts of various length and type. Orthographic and phonological coding, as well as non-verbal problem solving were assessed by means of standard tests. Assessments of reading experience and aspects of vocabulary (size and semantic lexical structuring) entailed new assessment tools developed for group administration by the authors of this paper. Lexical structuring was assessed with odd one out tasks (e.g. which one does not belong here: path, street, road, river). Results Data analyses showed that in the entire sample, skills of semantic lexical structuring and reading experience made strong contributions to reading comprehension. Analyses of the pupils below the 25%-percentile in reading comprehension revealed that for the vast majority simultaneous phonological and orthographic coding difficulties were found as well. Conclusions The results suggest that access to semantic structuring of lexical items play an important role in reading comprehension abilities for the entire group of 3rd graders. However, for those with poor comprehension basic word reading abilities still made the most constraints on their comprehension. The results are discussed in the light of Ricketts et al. (2008) and Perfetti & Hart (2002).
Kathleen Roskos (John Carroll University)Karen Burstein ; Jeremy Brueck - Title: The eBook in Preschool Vocabulary Instruction: Promise and Pitfalls
Purpose: This study describes the use of eBooks in supplemental, small group Tier 2 instruction for preschoolers at risk in vocabulary development. Research evidence suggests that e-shared book experiences boost word learning and cultivate basic skills for future e-reading over conventional shared book reading, although generally limited to one-on-one settings. Method: Using a formative research design, high quality eBooks were incorporated into a standard treatment protocol--Word Play Time--as Read Aloud books for vocabulary teaching. eBooks were selected using evidence-based instructional design principles and analytic techniques to establish a mixed genre corpus of 25 eBooks that included 250 target words aligned with the preschool curriculum, consisting of basic, root and subject matter vocabulary. The sample included three preschool teachers in Head Start classrooms and nine kinder-bound children (3-member small groups) who scored in the lowest quartile of the PPVT-IV. Over an 8-week cycle, three structured iterations of eBook shared reading in the context of the standard treatment protocol were tracked using PERT, Iterative Sequential Checklists, and Criteria of Merits lists to measure progress toward end goals of functionality and quality. CBM was used to progress monitor children's word learning in expressive, receptive and word meaning skills. Results: Results yielded dimensions and indicators in three domains of instructional interaction: (i) small group organization; (ii) instructional supports; (iii) play supports, and showed evidence of strong word learning gains on target words. Conclusions: Embedded in a standard treatment protocol, the high quality eBook shows promise as a robust instructional resource for improving at risk preschoolers' vocabulary development.
Susie Russak (Beit Berl Academic College, Israel); Elinor Saiegh-Haddad - Predictors of word reading in Hebrew (L1) and English (FL) among normal (NR) and reading disabled (RD) adults
Purpose: The study examined predictors of reading among Hebrew (L1) and English (FL) normal and reading disabled adults (N=60). Method: Three cross-linguistically equivalent experimental tasks which tapped into phonological, orthographic and morphological awareness were developed along with word reading tasks in both languages. Results: Results showed different predictors of reading for group and language. Within language correlations showed that for English all linguistic skills correlated with reading for both groups, whereas for Hebrew linguistic skills correlated with reading for the NR group only. Multiple regression analyses in each language with group, as well as phonological, morphological and orthographic awareness as predictor factors showed that English reading was significantly predicted most strongly by morphological awareness, followed by orthographic and phonological awareness. In Hebrew, none of the linguistic skills significantly predicted reading. Yet, group was found to be a significant predictor. The multi-linear regression analysis also revealed a significant interaction between group and orthographic awareness in English with higher orthographic awareness among the NR, but not the RD, predicting higher levels of reading. No such interaction was found for Hebrew. The results also showed a significant interaction between phonological and morphological awareness in English. Morphological awareness predicted word reading among individuals with higher, but not lower levels of phonological awareness. A similar interaction was found in Hebrew between phonological and orthographic awareness. Orthographic awareness predicted reading among individuals with high, but not low phonological awareness. Conclusions: Results will be discussed within the framework of universal versus language specific predictors of reading.
John Sabatini (Educational Testing Service (ETS))Kelly Bruce - Investigating the Range and Variability in Reading Subskills of Middle School Students
* Purpose - The purpose of this session is to investigate the range and variability in reading subskills of middle level students as they relate to overall comprehension as measured by state literacy tests. Results will be interpreted with respect to the use of screening/diagnostic assessments to inform instructional decision-making and resource allocation at the classroom and school level. * Method - About 3372 students were tested at nine middle and elementary schools in a small urban district, evenly distributed across grades 6-8. Roughly half were females; 12% classified Limited English Proficient; 15% classified as Special Education students. Students completed an experimenter based, computerized battery of individual reading component tests: Word Recognition & Decoding, Vocabulary, Morphology, Sentence Processing, Efficiency of Reading for Basic Understanding, Reading Comprehension. Each subtest was scaled to state test basic and proficient benchmark levels. * Results -showed good subtest reliability (r=.78-.91); moderate correlations to the previous Spring's state test scores (r=.62-.73); one factor underlying all subtests; and each subtest resulted in a significant change in variance in predicting state test scores in multiple regression models. We will also discuss analyses designed to understand the unique contributions of each subtest to understanding subtypes of student profiles. * Conclusions - To better understand middle level reader needs, we designed an assessment battery that targets a wide range of reading subskills (from lexical to sentence to discourse comprehension) that can be administered in about 1-2 class periods. The results build the empirical base for understanding the range of reading difficulties facing middle level readers. * Keywords: reading comprehension; longitudinal
Elinor Saiegh-Haddad (Bar-Ilan University)Haitham Taha - Morphological transparency and spelling accuracy in Arabic
Purpose. Arabic is a Semitic language with a rich morphology, which is reflected in its orthography. Specifically, many Arabic words consist of a root and a pattern (e.g., root K.T.B. + pattern maCCu:C = maktu:b 'written' ). Thus, spelling success in Arabic is assisted by morphological knowledge about the structure of the word. Method. The current study tested the use of morphological processes in word and pseudo word spelling accuracy in 2nd, 4th, and 6th grade normal and reading disabled Arabic native children. Two types of words and pseudo words were compared: a) morphologically transparent- constructed according to productive word patterns, and b) morphologically opaque- constructed according to non-productive word patterns. A total of 289 participants were tested. Results. The study showed a significant effect of morphological transparency on spelling accuracy in both groups and across all grade levels tested. The positive effect of morphological transparency on word spelling emerged particularly strongly in the case of pseudo words, and among young and reading disabled children. Conclusions. These findings indicate the central role of morphological structure in learning to spell in Arabic.
Paula Salmi (Niilo Mäki Institute); Tuire Koponen; Kenneth Eklund; Tuija Aro - RAN and counting skill: predictors of reading and calculation fluency after controlling for verbal short-term memory and phonological awareness
Purpose. Although arithmetic and reading skills have been studied for decades, the underpinning cognitive factors underlying these two skills are not fully understood. An interesting question is to which extent these two academic skills share common cognitive predictors. This study examined to what extent children's performance in rapid serial naming in kindergarten and counting at the beginning of first grade predict fluency of reading and arithmetical calculation skills in early school years. The effects of verbal short-term memory and phonological awareness were controlled to evaluate unique predictive associations of RAN and counting with arithmetic and reading. Method. Participants were 200 children followed from birth in the Jyväskylä Longitudinal Study of Dyslexia (JLD). Half of the participating children had incidents of dyslexia in their immediate family (n=108) and half did not (n=92). To examine the inter-correlations between the measures in a comprehensive model, the Sructural Equation modelling (SEM) approach was applied using the MPlus program (version 5.21). Results. The results show that both counting and rapid serial naming predict later reading and calculation fluency above and beyond what is explained by verbal short-term memory and phonological awareness. Conclusions. Results will be discussed.
Lana Santoro (Pacific Institutes for Research & Instructional Research Group); Scott K. Baker; Hank Fien; Yonghan Park; David Chard - Enhancing expository read alouds with tier 2 vocabulary and language instruction
Purpose: Our study investigated the impact of Tier 2 instruction on the comprehension and vocabulary of first grade students identified with low language and vocabulary skills. Specifically, we examined whether students participating in interactive read alouds during regular classroom instruction (Read Aloud Curriculum - Tier 1) benefited from additional small group instruction in vocabulary and comprehension during a later time in their reading/language arts time. Method: A randomized block design was employed. We blocked by classroom and randomly assigned students within 18 participating classrooms to condition. All students participated in a whole class Read Aloud Curriculum (Tier 1). Students in the intervention group received additional small group instruction for 20-minutes, two times per week (Tier 2). Small-group instruction included additional expository read alouds and opportunities to pre-teach, review, and enhance vocabulary and comprehension. The comparison group did not receive additional small group instruction. Pretest and posttest vocabulary and retell measures were administered. Because subjects were nested within classrooms, hierarchical linear modeling provided a framework to explore the intervention effect. Results: Outcome data favor students in the intervention group on vocabulary and expository retellings with effect sizes of .66 and .57 observed. Conclusions: Results are particularly noteworthy given that significant effects of .66 and .57 were observed with modest levels of instruction. A critical aspect of this Tier 2 intervention is the instructional focus on vocabulary and use of dialogic interactions to increase student opportunities to practice and expressively use language. Tier 2 instruction structured discourse about expository topics by making the interactions accessible to students with language delays without compromising the complexity of content. Overall, results support adding a small group component to whole class read alouds.
Kate Saunders (University of Kansas)Megan Stein - The development of computerized instruction for generalized sound-print relations: long- and short-vowel word pairs
Purpose: Our ultimate goal is to develop procedures that promote the acquisition of the alphabetic principle in individuals with intellectual disabilities (ID). We do this using a computerized "constructed-spelling" task. Previous studies have demonstrated generalization to new combinations of the onsets and rimes contained in taught consonant-vowel-consonant (CVC) words. The purpose of this study was to extend this work to CVC and CVCe words. Methods: Participants were adults with ID who had first-grade-level sight-word reading skills and who selected the onsets of spoken words correctly, but did not correctly select rimes differing by the presence or absence of the final e. We used a multiple-baseline-across-rime-sets design. There were 7 rime sets, 5 containing an equal number of long and short 'a' words (at/ate, an/ane, ap/ape, ad/ade, af/afe) and one rime set each with 2 other vowels (et/ete and ot/ote). Each rime set contained at least 16 words (8 onsets with each of the 2 rimes). After pretesting, participants were taught to construct pairs of words by selecting the onset and rime from a choice pool. Results: After mastering the construction of some words in a rime set, participants began to construct correctly other words with the same rimes. In addition, probes throughout the study showed generative whole-word reading following mastery of the word-construction task. Conclusion: Although it is an empirical question, we would predict similar outcomes for individuals with similar preexisting skills. These findings suggest that word-construction training is a valuable addition to comprehensive word-attack instruction.
Robert Savage (Department of Educational and Counselling Psychology); Louise Deault; Julia Daki; & Julie Aouad - Orthographic analogies and early reading: Evidence from a multiple clue word paradigm.
Two experiments using a variation of the 'clue word analogy task' (Goswami, 1986) explored whether children can make orthographic analogies when given multiple clue words, beyond the known effects of phonological priming. In experiment 1, 42 children (mean age 6 years and 8 months) were first taught three 'clue' words (e.g. 'fail', 'mail', 'jail'), and then shown transfer words sharing orthographic rimes (e.g. 'hail'), phonological rimes (e.g. 'veil'), orthographic vowel digraphs (e.g. 'wait'), phonological vowel digraphs (e.g. 'vein'), or unrelated controls (e.g. 'bard'). All word types were advantaged at post-test over unrelated controls. A small additional advantage for orthographic rimes over phonological rimes was evident in by-participant analysis. Finally, regression analysis showed a specific relationship between onset-rime phonological skills and orthographic rime clue word task transfer. Experiment 2 replicated experiment 1 with 30 children (mean age 7 years, 0 months) and added a distinct group of children taught multiple clue words sharing vowel digraphs (e.g. 'gait', 'maim', 'maid'). Results showed advantages for all words over unrelated controls and a small additional advantage for orthographic vowel digraphs over phonological vowel digraphs in the by-participant analysis. Results suggest that young children do have the capacity to make orthographic analogies when given multiple exemplars but that most transfer reflects phonological priming effects.
Chris Schatschneider (Florida State University)Jeannette Taylor - Genetic influence on reading constructs in kindergarten and first grade: evidence from a diverse twin sample
Purpose: Historically, twin research on reading has been conducted on older children and the generalizability of results across racial/ethnic/socioeconomic groups is unclear. To address these gaps, early reading achievement was examined among 960 twin pairs in Kindergarten and 1061 twin pairs in First Grade (ages 5-7) from a population based sample of twins. Method: These analyses were conducted on data from the Florida Twin Project in Reading (FTPR) which currently has 4,363 individual twins enrolled from grades K-5. Reading and reading related assessments are collected on these twins every year as a part of a much larger statewide effort to assess children in public schools across the state of Florida. The ethnic/racial composition of the twins in our study reflects the diversity of the State of Florida, with about 25% Hispanic, 20% African American, and 47% White and 8% reporting more than one race. Results: Univariate models were fit to data from the total sample and to subsamples of racial/ethnic and neighborhood income groups within each grade. In Kindergarten, variance in letter naming and phonemic segmentation fluency was associated with shared and non-shared environmental factors in the total sample and across most subsamples. In First Grade, variance in oral reading fluency was associated with large additive genetic effects for White and middle/upper neighborhood income twins, but shared environmental influence was substantial for low neighborhood income twins and for Black and Hispanic twins. Conclusions: Results suggest that the etiological architecture of some early reading skills may differ across cultural and economic contexts, while others remain fairly constant.
Rachel Schiff (Bar Ilan University)Dorit Ravid - Morphological processing in dyslexic and normally reading students: Plural adjective agreement in Hebrew
Purpose: Adjective agreement in number and gender with nouns constitutes a stumbling block to Hebrew poor readers, especially in the context of irregular morphology and misleading syntactic cues. The current study investigates the impact of interference by linguistic factors on adjective agreement with a noun in a compound structure .Method: 36 dyslexic and 36 normally reading Hebrew-speaking university students judged the correctness of sentences containing subject compounds and predicative adjectives. Task sentences were subdivided by head noun gender (masculine or feminine) and semantics (human or non-human), head noun morphology (regular or irregular suffix, changing or non-changing stems), and compound structure (with two same-gender nouns or with contrasting gender nouns). For example, batey ha-soxarot hayu gvohot 'the-houses,Masc (of) the-merchants,Fm were tall, Pl,Fm' is a non-grammatical sentence with regular noun morphology, contrasting noun gender within the compound, and an incorrectly marked plural adjective. Measures were accuracy and latency. Results: Dyslexic students achieved significantly lower scores and longer RTs than normal readers. Masculine gender, human (agent) semantics, and regular morphology of the head noun in the compound resulted in higher scores and faster RTs in judging sentence correctness (though less in dyslexics), while contrasting-gender interfered with correct judgment, especially in the dyslexics. Conclusions: Adult dyslexic students find it more difficult than normally-reading peers to use morpho-syntactic cues in judging plural adjective agreement with nominal compounds, especially when faced with morphological irregularity, absence of semantic salience, and noun gender interference. This suggests the involvement of grammatical deficits in dyslexia.
Gerd Schulte-Körne (Department of Child and Adolescent Psychiatry, Psychosomatics, and Psychotherapy, University of Munich)Darina Roeske; Nina Neuhoff; Jessica Becker; Jürgen Bartling; Jennifer Bruder; Felix F. Brockschmidt; Andreas Warnke; Helmut Remschmidt; Per Hoffmann; Bertram Müller-Myhsok; Markus M. Nöthen - First genome-wide association scan on neurophysiological endophenotypes points to transregulation-effects on SLC2A3 in dyslexic children
Purpose: It has been shown that event related potential measurements reveal differences between dyslexic children and age-matched controls. This holds particularly true for mismatch negativity (MMN), which reflects automatic speech deviance processing and is altered in dyslexic children. Method: We performed a whole-genome association analysis in 200 dyslexic children, focusing on MMN measurements. Results: We identified rs4234898, a marker located on chromosome 4q32.1, to be significantly associated with the late MMN component. This association could be replicated in an independent second sample of 186 dyslexic children, reaching genome-wide significance in the combined sample (p=5.14e-08). We also found an association between the late MMN component and a two-marker haplotype of rs4234898 and rs11100040, one of its neighbouring SNPs. In the combined sample, this marker combination withstands correction for multiple testing (p=6.71e-08). Both SNPs lie in a region devoid of any protein-coding genes, however, they both show significant association with mRNA-expression levels of SLC2A3 on chromosome 12, the predominant facilitative glucose transporter in neurons. Conclusions: Our results suggest a possible trans-regulation effect on SLC2A3, which might lead to glucose deficits in dyslexic children and could explain their attenuated mismatch negativity in passive listening tasks.
Mila Schwartz (University of Haifa, Oranim Colledge)Michal Lerner; Zvia Breznitz - Cross-linguistic perspectives on reading difficulties: Comparison of literacy profile between bilingual and monolingual adult dyslexic students
Purpose: The aim of the present research was to investigate the profile of Russian-Hebrew speaking adult dyslexic students and to compare it to Hebrew-speaking adult dyslexic students. We hypothesized that the bilingual dyslexic students will show difficulty: (1) in spelling and reading rate in Russian (L1), and less problems with decoding and (2) in reading accuracy and fluency in Hebrew (L2) deep unpointed script. Method: The overall sample included 48 adult dyslexic students: 24 bilinguals and 24 monolinguals. The selection of the adult dyslexic students was based on a diagnostic database collected in the Support Unit for Students with Learning Disabilities. The bilingual dyslexic students were asked to furnish oral reports on history of reading and writing acquisition in L2. A wide range of linguistic, meta-linguistic, cognitive and literacy tasks in Hebrew (L2) and in Russian (L1) were administered to bilingual and monolingual adult dyslexic students. Group comparisons were conducted using a series of t-tests and Cohen's d was used as a measure of effect size. Results: The bilingual dyslexic adults showed the same salient deficiencies in their L2 as in L1, namely low reading rate and insufficient orthographic knowledge. Conclusions: The results stress the necessity to address the unique characteristics of the orthographies involved in interpretation of bilingual dyslexic students' performance on reading and spelling in either L1 or L2 with accordance to Orthographic Depth Hypothesis.
Lucja Segal-Seiden (OISE/UT)Esther Geva - Development of writing conventions in L1 and L2 children between grades 4 and 6
* Purpose: Investigating development of written language between grades 4 and 6 in children who are native speakers of English (L1) and children for whom English is a second language (L2). * Method: This was a longitudinal study, matched pair design, involving 44 L1 and 44 L2 (total n=88) normally developing children. Children were matched on language status and nonverbal cognitive ability (Raven Test score). Study included analysis of the stories written by the same children as part of TOWL-3 test in grades 4 and 6 in response to the picture stimulus. This poster focuses on results from TOWL-3 test, Subtest 6, "Contextual Conventions ", which tests the mastery of the rules of writing in English. Data analysis was carried through Repeated Measures ANOVA with language group (L1/L2) and cognitive ability (Low/High) as the independent variables. * Results: The interaction of grade and language group was significant. Specific writing convention, eg.,use of paragraphs, use of comma to offset direct quotation and correct spelling were mastered by the L2 significantly earlier than by L1, but by grade 6 L1 closed the gap with L2. High cognitive ability children significantly outperformed low cognitive ability children in each group. * Conclusions: Overall, in the area of acquiring contextual conventions in English story writing it was not the language status but the cognitive ability that determined performance, with high cognitive ability children outperforming low cognitive ability children. Interestingly, L2 children were not disadvantaged and sometimes performed better than the L1 children.
Eliane Segers Segers (Radboud University) - The Use of Mnemonics in Learning Grapheme-phoneme Connections in Children with SLI
Purpose - Letter knowledge is an important predictor of literacy development. The question of the best way to teach children letter sounds has not, as yet, been answered satisfactorily, especially not in children with learning problems. De Graaff, Verhoeven, Bosman and Hasselman (2007) found normal language learning children to benefit from using mnemonics in a computer-supported environment, especially in a condition in which the mnemonic picture is used in a fading condition. The computer was used without teacher support. Children with Specific Language Impairment are at risk of developing reading problems and are known to have problems in learning letter sounds. Visual support could enhance their learning, and it was investigated whether mnemonics would boost their letter learning. Method - Participants were 20 children with SLI who had just started formal reading education. Each kindergartner learned letter sounds under three conditions: (a) a fading condition in which letters are taught using a picture-supported first-sound mnemonics procedure in combination with a fading procedure in which the picture slowly disappears; (b) an embedded condition in which letter sounds are taught using the picture-supported first-sound-mnemonics procedure only and (c) a without-picture condition in which letters are taught using a first-sound procedure without-picture support. Results - Repeated measures analyses showed children to have higher learning gains in the two mnemonic conditions, but - contrary to normal language learning children - there were no differences between the fading and embedded condition. Conclusions - These findings indicate that an integrated-picture mnemonics procedure is beneficial for children with SLI.
Mark S. Seidenberg (University of Wisconsin-Madison) - Computational tools for understanding reading
Gabriela Seidlova Malkova (Charles University Prague);Markéta Caravolas (University of Wales, Bangor UK, voting member SSSR); - Phoneme training and its impact on letter learning
Purpose. In our earlier work (Caravolas & Landerl, in press - SSR; Hulme et al., 2005), we argued that, contrary to views that phoneme awareness emerges as a consequence of letter learning, phoneme awareness is separate from but reciprocally related to letter knowledge. However, in those studies, children with phoneme awareness typically knew some alphabet letters, thus precluding strong claims about the separate ontogeny of these two skills. In the present 3-phase study, we tested whether training awareness of phonemes would help children learn letter sounds. Method. Participants were 44 Czech children (48-63 months), who at screening could not manipulate any phonemes and knew no alphabet letters. Phase 1: children were randomly assigned to one of two pre-training conditions (n = 22): Phoneme awareness (5 days training of 5 phonemes /s, m, k, l, p/) or Non-alphabetic exercises (5 days graphomotor training). Phase 2: all children underwent a 5-day letter training programme of 10 letter-sound associations (including phonemically pre-trained S, M,K,L,P). Phase 3: Phoneme awareness and letter knowledge were post-tested after letter training (wk. 2) and one week later (wk.3) to assess training effects. Results: (1) phoneme awareness, but not graphomotor training promoted general and durable growth in phoneme awareness; (2) letter training promoted item-specific letter learning regardless of pre-training condition; (4) importantly, there was no item-specific or general facilitation of phonemic training on letter learning. Conclusion. Our results support the "dual-foundation view" that both skills emerge independently; they become reciprocally associated through explicit linkage, primarily through literacy instruction.
Willy Serniclaes (applying for membership); Miguel Lopez; Nora Termeche; Olivier Dufor; Juan-Luis Luque; Jean-François Démonet; Liliane-Sprenger Charolles - A review of the evidence in support of allophonic perception in developmental dyslexia.
Purpose: Previous studies suggest that children with developmental dyslexia (DD) display "allophonic perception": they do poorly when discriminating between different phonemes and do better when discriminating acoustic variants of the same phoneme (e.g. Serniclaes et al., JECP, 2004). Here we present a critical review of the evidence in support of allophonic perception in DD and we discuss the link between allophony, auditory processing and speech production. Method: an internet search with the keywords "allophonic perception", "allophonic"& "dyslexia", "speech production & dyslexia" was conducted. Results: different studies support allophonic perception (e.g. Bogliotti et al., 2008, JECP) while other studies present counter-evidence (e.g. Ramus & Szenkovits, 2008, QJEP). Discrepancies between studies seem to be related to differences in the articulatory status of the allophonic contrasts used in these studies. Differences between DD and controls are larger for contrasts which are difficult to produce. The link between allophonic perception and speech production is supported by the results of behavioural and neuroimaging investigations suggesting a deficit in the relationship between the auditory and articulatory representations of speech sounds (Boets et al., 2007, Neuropsychologia), located in left hemisphere in DD (Dufor et al., 2007, NeuroImage; Hommet et al., 2009, Neuropsychologia). Further, adult DDs are more "allophonic" when they activate left pre-motor areas whereas the controls are less "allophonic" when they activate the same areas (Dufor et al., 2009, NeuroImage). Conclusions: allophonic perception in DD seems to be due to a distortion in the articulatory mapping of the various auditory features contributing to the same phoneme.
Donald Shankweiler (University of Connecticut - Haskins Labs)David Braze; Robert K. Fulbright; W. Einar Mencl; Kenneth Pugh; Whitney Tabor - Tracking the supramodal language brain in skilled adult readers
Purpose - At SSSR 2008 we reported a result of an fMRI study comparing brain activity elicited by matched spoken and printed sentences in young adults who varied widely in reading skill. It was found that degree of speech-print convergence during sentence processing was positively correlated with skill level at some cortical sites. Subsequently, a follow-up study based on the more skilled readers in the sample (N = 18) enabled us to investigate the scope of convergence within supramodally-responsive cortical zones. Method - Brain activity was monitored while the participants (18 skilled readers) performed a semantic categorization task with matched printed or spoken sentences that were either well-formed or contained anomalies of syntactic form or pragmatic content. Voxel-wise similarity of patterning across modalities of the cortical response to the three sentence types (two anomalous and corresponding nonanomalous) was assessed quantitatively by applying a cosine similarity metric to the fMRI data from whole brain scans (Braze et al, Cortex, in press). Results -The results of applying the cosine similarity metric identified all clusters of voxels that showed a high degree of pattern coherence for the three sentence types. These varied considerably in volume and were found in both hemispheres, the largest in left temporal and lateral frontal regions within cortical zones that receive polysensory inputs. Conclusion - The findings suggest that an essential feature of skilled reading is reaching a state in which speech and print versions of structurally varying sentence material evoke the same pattern of response within the supramodal language brain.
Linda Siegel (University of British Columbia, Canada); Orly Lipka - The Development of Reading Comprehension Skills in ELL Children
Purpose and Method: In order to address the development of reading comprehension skills in children learning English as a second language (ELL), a longitudinal study was conducted of 942 children from grades 2-7. Results: Although the ELL children initially had poorer reading comprehension skills, these differences disappeared by grade 5. The development of age appropriate reading skills was due, at least in part, to the word level skills that were taught during the kindergarten and grade 1 years and the district wide program to foster reading comprehension skills. We examined the differences and similarities among the children as a function of first language (Farsi, Chinese, South Asian and Romance) and we found no differences among these groups. We developed a reading comprehension task that did not require background knowledge and found that the ELL children performed relatively better than the children with English as a first language (L1). Conclusions: These results show that with appropriate instruction, ELL children can catch up to L1 children in the area of reading comprehension.
Macarena Silva (Universidad de Chile); Kate Cain - Narrative comprehension and production: the role of oral language and home literacy
Purpose: Narrative skills are predictive of reading comprehension and are a useful index of discourse comprehension in non-independent readers. The current study investigated the role of oral language and home literacy practices on narrative production and comprehension, both concurrently and longitudinally. Method: Eighty-two children aged 4 - 6 years completed assessments of nonverbal IQ, short-term memory, receptive vocabulary and receptive grammar. Narrative comprehension and production were assessed through the narration of a wordless picture book and answers to questions about the book's content. Parents completed a questionnaire about the home literacy environment. One year later, 69 children of the original sample completed the same measures. Results: After controlling for memory and nonverbal IQ, the influence of oral language on narrative skills was limited. Vocabulary and grammar knowledge explained variance in narrative comprehension at Time 1, but not in production. Home literacy practices were more strongly related to narrative skills, explaining unique variance in narrative comprehension and production at both time points. Of note, different types of home literacy practice were uniquely related to comprehension and production. Longitudinally, the home literacy environment at Time 1 predicted narrative comprehension skills at Time 2, over and above nonverbal IQ, oral language skills, and the autoregressive effect of early narrative comprehension. Conclusions: Word and sentence oral comprehension have only a moderate influence on narrative skills in young children. In contrast, specific home literacy practices exert a stronger influence on young children's ability to comprehend and produce coherent stories, and the development of this ability.
Gabriela Silva-Maceda ()Josephine Chen-Wilson; Kenneth Manktelow - Are better communicators better readers? An exploration of the associations between expressive language and reading comprehension.
Purpose. There is an increasing need to examine the association between wider oral language skills and reading comprehension (RC). However, most of the research focuses on receptive skills with expressive skills largely overlooked. This study aims to explore the relationship between RC and language skills using both receptive and expressive measures in 7- and 8-year-old children, at the time when they start transitioning from decoding to RC. Method. Seventy-five children were recruited; preliminary results are available for 21. Receptive language skills, non-verbal intelligence, RC and a measure of story content were examined using standardised tests. Expressive measures were based on narratives elicited from picture-books. Using CLAN software (MacWhinney, 2000) expressive vocabulary was measured using parameter D (i.e. VOCD), while Mean Length of Utterance in T-Units was used for expressive syntax. Children's language skills at the discourse level were also assessed. Results. In the available subsample, correlations with RC were only significant for receptive grammar (r=0.58, p<0.01) and narrative story content (r=0.46, p<0.05). Expressive grammar showed a non-significant trend. The magnitude of these correlations is likely to increase with an increased sample size. Conclusions. These findings show that receptive grammar and the ability to relate story content were the strongest associations for 7-8-year-olds' RC. Preliminary analyses seem to suggest that socio-economic background might also play a role on how language variables are related to RC. Data from a larger and more socially diverse sample are currently being collated. A more comprehensive review of factors associated with RC will be further discussed.
Darcey Sims (Florida State University/ Florida Center for Reading Research)Christopher Lonigan - A multi-method examination of preschool children's inattentive/hyperactive behaviors and emergent literacy skills
Purpose: This study was designed to examine how distinct symptoms of ADHD and different methods of assessing these symptoms relate to preschoolers' emergent literacy skills. Given prior research on older children, it was hypothesized that measures of inattention, but not hyperactivity/impulsivity, would correlate with emergent literacy skills, and that behavioral ratings of inattention (i.e., teacher ratings) would be more strongly related to emergent literacy skills than would cognitive inattention (represented by CPT omission-errors). Method: Measures of emergent literacy skills and a computerized Continuous Performance Test (CPT) were completed by 166 3- to 5-year-olds. Teachers also rated these children's ADHD-characteristic behaviors. Results: Teachers' ratings of inattention and CPT omission-errors were significantly correlated with an overall index of emergent literacy skills, whereas ratings of hyperactivity/impulsivity and CPT commission-errors (i.e., hyperactivity/impulsivity) were not. Multiple regression analyses indicated that both measures of inattention contributed unique variance to emergent literacy skills, even controlling for general cognitive ability. An examination of the individual components of emergent literacy revealed an identical pattern of results for phonological awareness skills and a similar pattern for print knowledge, with CPT commission-errors rather than omission-errors uniquely predicting print knowledge. Results were non-significant for definitional vocabulary. Conclusions: Results confirm the strong association between behavior problems and early reading skills. Although teacher-ratings and the CPT are commonly used to measure inattention, this study indicates that they are measuring distinct constructs. Clarification of the constructs represented by these measures would facilitate a better understanding of the critical underlying psychological deficits that are associated with, and potentially exacerbate, reading problems.
Aparna Sivapurapu-Pisupati (Kennedy Krieger Institute);Raj A.Stewart; Sheryl L.Rimrodt; Sarah H.Eason; Katherine M.Young; Lindsay F.Goldberg; Nicole Davis; Ken R. Pugh; Laurie E.Cutting - FMRI correlates of a dynamic measure of word learning in children.
Purpose: Previously, it has been demonstrated that after completing a dynamic training paradigm attaching meaning to pronounceable pseudowords ("Trained Pseudowords", TP), young adults show fMRI activation to TP that is more similar to reading real words than reading untrained pseudowords. The purpose of the present investigation was to determine, using the same paradigm, if these same patterns of fMRI activation were found in typically developing adolescent readers. Method: fMRI data of 23 typically developing adolescent readers in the age range of 9-15 years old were collected. In the scanner, participants were exposed to 40 TP, 40 novel pseudowords, and 80 real words. Statistical Parametric Maps were generated to see the differences in the activation pattern for all stimulus types. Results: Similar to previous findings in adults, few differences in activation were observed between TP and real words, suggesting that after training, BOLD response to TP is similar to that of real words. Also consistent with previous findings, novel pseudowords showed greater activation than both words and TP in left inferior occipital and right inferior frontal regions, also providing support that words and TP are more alike than pseudowords. Conclusions: Neurobiologically, and similar to adults, TP showed more similarity in activation to real words than untrained pseudowords in regions associated with reading. Findings suggest that with exposure, activation to TP is similar to real words, implying that plasticity in adolescent readers is similar that of adult readers; it remains to be determined if similar activation patterns are observed in adolescents with reading disabilities.
Thomas Skiba (Yale University, Child Study Center, New Haven, Connecticut CT 06519, USA)Jodi Reich1, Adam Naples1, Mei Tan1, Kelly Nedwick1, Philip E. Thuma2, & Elena Grigorenko1 1Yale University, Child Study Center, New Haven, Connecticut 230 South Frontage Road, New Haven, CT 06519, United States 2Macha Mission Hospital P.O. Box 630166, Choma, Zambia - Reading assessment in Chitonga: A comprehensive study of Zambian children
Purpose: The purpose of our study is to understand the prevalence and etiology of specific reading disabilities (SRD) in rural Zambia. Here, we report on the assessments most predictive of reading competence in rural Zambian children. Method: An assessment protocol including measures of ability and achievement was developed for Chitonga to examine reading competence in rural Zambian children. The protocol included selected nonverbal cognitive ability subtests adapted from the Universal Nonverbal Intelligence Test (UNIT) and the Kaufman Assessment Battery for Children, Second Edition (KABC-II). Additional reading-related componential measures were developed or adapted including: letter/digit span for working memory, phonological memory, and others. The Zambian Achievement Test (ZAT: Chitonga Version) was administered to measure reading. Data were analyzed to establish patterns of correlation between componential measures of reading. Results: Preliminary results suggest that the three ZAT reading measures (Reading Comprehension, Pseudo-word Decoding, and Reading Recognition) are all highly correlated (with adjusted R2 values between 0.82-0.85) and explain variance in each at a much higher level compared to the effects of other predictors. Comparing the individual ZAT subtests to the componential measures, working memory predicted 10% of the variance for reading comprehension and pseudo-word decoding. Working memory and the KABC-triangles were predictive of reading recognition. The UNIT subtests were not very predictive of the ZAT subtests. Conclusion: We will discuss the performance of our assessment protocol, including further measures. We will also question the role of IQ indicators in predicting reading skills in societies where standardized IQ tests are not commonly used.
Lori Skibbe (); Carol Connor; Frederick Morrison - Schooling effects on preschoolers' self-regulation, early literacy, and language growth
Purpose: The present study examined the extent to which self-regulation, early literacy development (i.e., letter knowledge and decoding) and language growth (i.e., vocabulary) were related to children's first and second years in preschool. Method: Using the school-cutoff technique, children (n = 60) whose birthdates fell two months on either side of the arbitrary cut-off date for preschool entry were assessed using measures of early literacy from the Woodcock-Johnson III Tests of Cognitive Abilities and self-regulation (i.e., Head-to-Toes task). By comparing the two groups of children almost identical in age, but in different academic environments, we directly examined the extent to which growth was due to age-related or schooling-related processes. Results: Using hierarchical linear modeling, our results suggested that the first and second years of preschool are both systematically associated with decoding and letter knowledge gains, and the effects are cumulative (two years predicted greater gains overall than did one year of preschool). Although children's self-regulation scores increased from fall to spring, results indicated that this growth was attributable to general development rather than time spent in preschool. Vocabulary gains were also attributed to chronological age rather than school. Follow-up analyses will include discussion of the instructional emphases within these classrooms. Conclusion: Results suggest that more preschool is better than less, as children demonstrated significant growth in early literacy skills during both years of school. However, not all skills were affected by preschool (i.e., vocabulary and self-regulation). Implications for preschool curricula will be discussed.
Daisy Smeets (Leiden University)Marianne van Dijken;Adriana Bus - Multimedia Storybooks: Suitable for Children with SLI?
Purpose. Word learning is facilitated when words are concretized by images. Multimedia additions to digital storybooks (i.e., live action video, music and sounds) attract children's attention to details of pictures that match to the oral rendition of text. As children with specific language impairment (SLI) heavily rely on visual support, multimedia features are expected to be beneficial for vocabulary acquisition just as they are in normally developing groups. Method. Participating children (N=18) were diagnosed with SLI by standard criteria as used in special education schools in the Netherlands. A counterbalanced, within-subjects design was used to compare children's word learning in three different conditions: a) digital stories with static pictures; b) multimedia stories; or c) control stories (only pre- and posttested). Experimental stories were presented four times during a one-month intervention period. Children's book-based expressive vocabulary was pre- and post-tested. Results. Vocabulary gains for pooled experimental conditions were superior to the control condition (p <.001) with a large effect size (Cohens'd = 2.17). Contrary to what we expected, static stories gained more than multimedia stories (p <.01, d = 1.79). Conclusions. SLI-children benefit from digital storybook exposure. However, multimedia stories encompassing film-like pictures, music and sounds are not as effective as static stories in which video-effects, sounds and music are absent. Since children with SLI have been reported to experience auditory processing difficulties, we hypothesize sounds and music in the multimedia versions to interfere with understanding and processing the storybook language. A follow-up experiment testing this hypothesis is currently carried out.
Allan Smith (University of Maine, Orono); Susan Lambrecht Smith; Jenny Roberts; John L. Locke; Katharine Farrell - Speech timing characteristics of young children's longest utterances and later reading disability
Purpose - Information from spontaneous language samples may allow early detection of reading disability (RD) in children younger than three years. One behavior that may predict RD is slow speaking rate, related to increased pausing. Many child utterances are not well suited for studying pausing, because the formulation demands are presumably low (e.g., simple agreements "yeah", negations "no", joint references "what's that?"). Assuming that pausing would be common when language formulation and processing demands are high relative to an individual child's capabilities, the longest utterances may provide useful information. The purpose of this study is to conduct a thorough analysis of timing in young children's longest utterances in order to find behaviors that are predictive of later RD. Method - 28 children (mean age 29 months) were recorded interacting with an adult. Most children were at high familial risk for RD. Later assessment showed 9 participants had RD. For each child, longest utterances were examined for their grammatical and phonological complexity, timing characteristics, including speaking rate, errors, pauses, and pause locations. Results - Preliminary results suggest that longest utterances offer a useful and interesting aspect of language to study. For example, children with RD may produce shorter stretches of speech between pauses, as compared with children who later read normally. Conclusions - It is hoped that the results will contribute to the understanding of language production and early predictors of RD.
Patrick Snellings (University of Amsterdam, Department of Psychology); Petra de Waard - Enhancing reading fluency beyond the word level: Text level repeated reading and sentence level reading acceleration
Purpose Repeated reading at text level and reading acceleration training at sentence level were used to experimentally enhance poor readers' fluency. We tested Dutch poor readers at word, sentence and text level to assess whether learning was item specific or generalized to untrained words for each level. The asynchrony hypothesis predicts generalized effects for reading acceleration training. We predicted stronger specific effects for repeated reading, especially at the text level. Method Two groups of children (n = 52) matched on reading fluency and IQ were randomly assigned to the accelerated and repeated condition and trained in 10 sessions. Pre- and Post-training we used standardized assessments and specific trained words versions of single word reading speed, sentence word reading speed and text reading speed. Results Preliminary results suggest that both groups improved on all fluency measures. Both types of training yielded generalized effects on standardized tests, yet specific effects on trained words were larger. The reading acceleration training was more effective in increasing sentence reading speed. Conclusions Results show that both repeated reading and accelerated reading improve reading fluency of poor readers at word, sentence and text level. Acceleration training was not more effective than repeated reading for untrained words, providing no evidence for generalized effects. Still, acceleration training at sentence level was equally effective as repeated reading for specific words at all levels, even at text level. Results suggest that both approaches effectively increase fluency at all levels but that the learning mechanisms underlying training are specific rather than general.
Catherine Snow (Harvard Graduate School of Education) - Discussant contribution
Catherine Snow will discuss the scientific and educational implications of the results of the four paper contributions.
Brooke Soden (Ohio State University, Dept of Psychology); Chris Schatschneider; Jeanette Taylor - Do reading subskills share the same etiology?: Investigating genetic overlap of reading subskills in a diverse twin sample.
Purpose The current study investigated the genetic overlap of reading subskills in the early school grades in a socioeconomically and ethnically diverse twin sample. It has been proposed that such skills are largely influenced by the same genetic factors (Plomin & Kovas, 2005), and the current study tests this hypothesis on a more diverse sample as well as extending the research to additional measures and age groups. Method Both longitudinal progress monitoring and achievement data were collected from 4,278 twins in a large southeastern state. Students were assessed using DIBELS measures of letter naming fluency, initial sound fluency, phoneme segmentation fluency, nonword fluency, and oral reading fluency in kindergarten through third grade. Genetic correlations were computed to assess genetic overlap of skills. Results The majority of the reading subskills showed moderate to substantial genetic overlap, however, some reading skills showed very little overlap of genetic factors. Conclusions The current study adds to the growing body of research that supports that many, but not all, reading subskills share common genetic factors. Importantly, the magnitudes of these relationships change over time. Understanding the roles of both genetic etiology and environmental factors of these skills can inform the role of environmental factors such as intervention. Timing of influences will also be discussed.
Jonathan Solity (KRM-PER Ltd)Janet I. Vousden, janet.vousden@coventry.ac.uk, Department of Psychology, University of Coventry, Membership Status - Active Member - 'Real books' or 'reading schemes': Evidence from instructional psychology on the texts which best facilitate children's phonic and sight vocabulary skills.
Purpose: The research involved an empirical investigation into the widely held assumption that 'reading schemes' provide children with more opportunities to apply their phonic and sight word reading skills to text than 'real books'. Method: Five databases were analysed that contained adult literature (the MRC Psycholinguistic Database of fiction and non-fiction); children's real books (a random sample of texts suitable for beginning readers) and three reading schemes (texts that reflected analytic and synthetic phonic approaches to teaching reading). An electronic database was constructed for each set of materials enabling each word's orthographic and phonemic representation to be noted along with its frequency count. Results: Within each database the 100 most frequently occurring word types accounted for over 50% of all word tokens. However, the next 50 frequently occurring word types accounted for less than 4% of the remaining word tokens. After excluding the 100 high frequency word types, 64 grapheme-phoneme correspondences (out of a possible 461) accounted for over 65% of all monosyllabic word tokens in the adult and real books databases and over 75% of all word tokens in the reading scheme databases. Conclusions: A small number of high frequency word types and grapheme-phoneme correspondences accounted for approximately 90% of the monosyllabic word tokens in all five databases. Surprisingly each database provided comparable opportunities for the beginning reader to apply their phonic and sight vocabulary skills. The databases reflect Pareto's Law which proposes that a minority of causes, inputs or effort usually leads to the majority of the results, outputs or rewards.
Matt Solomon (Florida State University);Albrecht Inhoff; Ralph Radach - Vowel length in silent and oral reading: Implications for 'inner speech' and the eye voice span.
Purpose: There is growing interest in the relationship between oral reading and a potentially similar stream of 'inner speech' when reading silently. The present work study examined readers' eye movements while reading silently and out loud, with the specific goal of comparing effects of word frequency and vowel length on word viewing times under these two conditions. Method: Participants were asked to read a series of sentences silently vs. out loud. Each sentence contained a target word that was a member of a similar sounding word frequency pair (high/low) or word pair matched on word frequency but differing in vowel length. Eye movements and audio recordings were collected. Results: Differences in word frequency had a localized effect on target viewing times when reading both silently and out loud. In contrast, differences in spoken vowel duration produced asymmetrical patterns of results. During silent reading there was an increase in word viewing time measures for longer vowels both on the pre-target and target word. When reading out loud, however, this target word effect was delayed until post-target word viewing. Conclusions: Our results suggest that featural vowel information is utilized in the generation of a phonologically informed inner voice during silent reading. Moreover, we add to the growing body of evidence on the eye-voice span by showing an increased temporal delay for speech information when reading out loud. Interestingly, this difference corresponds to the initial viewing of a word and its subsequent articulation, generating a likely span size of approximately 1 -2 words.
George Spanoudis (University of Cyprus);Timothy C. Papadopoulos; and Spyroula Spyrou - Evidence for expressive and receptive language difficulties in children with specific language impairment, but not in children with reading difficulties.
Purpose: Some researchers (Vellutino, Scanlon, & Tanzman, 1994;) have proposed that expressive and receptive language may influence reading development. Indeed, some studies (Wise, Sevcik, & Morris, 2007) that have focused on the relationships between expressive/receptive language and dyslexia in English-speaking populations, have confirmed that expressive and receptive language abilities facilitate word identification skills in children with reading difficulties. Method: Nineteen children with reading difficulties (RD) and twelve children, with specific language impairment (SLI) were assessed with a test battery consisted of two standardized expressive and two standardized receptive language measures, and a nonword repetition task in Grades 2 and 4 in Greek. Results: The analysis showed that the children with reading difficulties outperformed the children with specific language impairment on all measures. The scores of the RD group fell within the normal range on all tasks, whereas the scores of the SLI group were evidently below the normal range. In addition, group membership, as yielded in subsequent discriminant function analysis, was highly predicted on the basis of these group differences. Conclusions: The study highlights the need for future cross-linguistic research to assess both reading and expressive/receptive skills of RD and SLI school-age groups in different languages.
Erin Sparks (); S. Hélène Deacon - Morphological awareness and vocabulary development: Understanding the direction of their relationship
Purpose: Both morphological awareness and vocabulary have clear links to reading ability, and although we know that there is a strong association between these two abilities, we know very little about the direction of their relationship. In this longitudinal study, we investigate the possibility of a bidirectional relationship between morphological awareness and vocabulary in English-speaking children. We do so with a focus on the morpheme types that are most central to English: inflectional (including measures of past tense inflections as well as a broader range of inflectional forms) and derivational morphology. Method: In a cross-lagged longitudinal study across grades 1 to 3, 100 children completed measures of vocabulary and of past tense, inflected, and derived morphological awareness, along with measures of phonological awareness, word reading, pseudoword reading, and nonverbal reasoning. Results: We conducted hierarchical regression analyses using stringent controls, including the autoregressive effect of the Time 1 outcome variable. The direction of the relationship between vocabulary and morphological awareness depended on morpheme type: vocabulary predicted subsequent growth in past tense morphological awareness, whereas, in the opposite direction, broader measures of inflected and derived morphological awareness predicted subsequent vocabulary growth. Of the two, derived morphological awareness emerged as a stronger predictor of vocabulary growth. Conclusion: The change in the direction of the relationship between vocabulary and different types of morphological awareness suggests a dynamic relationship that is sensitive to the morphological distinctions within a language. These results further our understanding of the complex relationship between these reading-related abilities.
Rachael Sperring (The University of Reading); Patricia Riddell; John Stein - Assessing magnocellular function at different levels of the visual system in Specific Reading Deficits
Purpose - Lovegrove, Heddle & Slaghuis (1980) have suggested that abnormalities of the magnocellular system can result in visual confusion in people with reading difficulties. However failures to replicate these finding have lead some to disregard this theory (Skottun, 2000). Much of the controversy can be attributed to the levels of processing that psychophysical tests assess. Therefore this study aimed to investigate the relationship between magnocellular processing and reading ability at two contrasting levels of the dorsal pathway; the retina, and V5/MT. Method - Single word reading was compared with two tests for magnocellular function in 66 children aged 7-12 years. Random Dot Kinematogram integration time (RDK-IT) was used to assess magnocellular function at MT/V5. Frequency doubling illusion thresholds (FDT) assessed magnocellular function at the retinal level. Results - Regression analyses showed a significant relationship between reading ability and RDK-IT (r² = 0.11, p = 0.006), reading ability and FDT (r2 = 0.096, p = 0.01), and RDK-IT and FDT (r2 = 0.22, p< 0.0001). This suggests that early magnocellular processing difficulties are transmitted to higher processing levels. Conclusions - A proportion of the variation in reading ability can be accounted for by variability in magnocellular processing measured at either the retina or MT/V5. This suggests that magnocellular processing can be measured reliably at different levels of the visual system.
Nicole Stadie (University of Potsdam) - Treatment of the non-lexical (sight-word) route in a boy with hyperlexia
Purpose: This single case study evaluates the effectiveness of a phonics-based (non-lexical) treatment programme in a boy whose word recognition was far better than his word comprehension (hyperlexia). Method: JM was a 6-year-old boy with hyperlexia and high functioning autism. The treatment programme focussed on explicit training of grapheme-phoneme associations and blending. It was conducted in two stages: letter sounding and blending of trained items (96 words and 96 non-words). Pre- and post treatment assessments included trained and untrained items, as well as spelling and phonological processing. Results: There was a statistically significant treatment effect for trained items. This effect generalised to untreated words and non-words, as well as to spelling, and some aspects of phonological processing. All effects were still presented 8 weeks after the treatment had finished. Conclusions: These results support the idea that reading accuracy and reading comprehension skills dissociate in developmental disorders. In addition, the results show that intact lexical (sight-word) reading ability does not necessarily ensure adequate mechanisms for reading comprehension. The present study supports the use of cognitive neuropsychological models to facilitate treatment of developmental dyslexia.
Rhona Stainthorp (Institute of Education, University of Reading); Daisy Powell;Morag Stuart ; Daisy Powell; Morag Stuart - Does RAN performance contribute to word spelling?
Purpose: There is now ample literature which testifies to the contribution that rapid automatized naming (RAN) makes word reading performance. Thus the double deficit theory of developmental dyslexia proposes that RAN deficits may be found independently from phonological awareness deficits in children identified as having developmental dyslexia. However, there has been much less research investigating the relationship between RAN and spelling. Given that spelling performance potentially involves both phonological processes and orthographic processes (dual route model of spelling), this study investigated whether individual differences in RAN are related to word spelling performance. Method: The study involved a comparison between a group of 72 children with poor RAN performance, and a control group (N = 77) who were matched on age (7-9 years), phonological awareness, phonological memory and other key cognitive variables but with no RAN deficit. Between-group differences in performance on word spelling were analysed using ANCOVA. Regression analyses were used to investigate the contribution that RAN performance made to spelling performance. Qualitative analyses of spelling errors of phonological and orthographic error patterns were also conducted. Results: The Low RAN group showed significantly poorer spelling than the control group. Regression analysis showed that RAN performance made a significant unique contribution to word spelling. Conclusions: These data suggest that RAN performance is also implicated in individual differences in word spelling over and above the contribution of phonemic awareness. Qualitative data illustrate these differences. These findings can contribute to our understanding of the potential processing difficulties which underlie dyslexia.
Laura Steacy (Vanderbilt University); John R. Kirby; Lesly Wade-Woolley - Early identification and the double deficit hypothesis: stability of groups from Kindergarten to grade 2
Purpose The Double-Deficit Hypothesis of dyslexia posits that students can be grouped into four distinct groups: (a) average readers, (b) students with phonological deficits, (c) students with naming speed deficits, and (d) students with double deficits (those having both (b) and (c)). We examined the stability of the Double-Deficit groups across four time points from Kindergarten to Grade 2. Method 214 children were assessed in Kindergarten and subsequently in early Grade 1, late Grade 1, and Grade 2. Tests administered at each time included multiple measures of naming speed and phonological awareness, and a variety of reading measures. Naming speed and phonological awareness construct scores were formed by averaging z-scores at each time point. Groups were formed at each time based on these construct scores, with a buffer zone between groups. Results Discriminant analyses indicated that approximately 70% of Grade 2 children were successfully classified by Kindergarten measures. Contingency analyses indicated moderate stability from Kindergarten to Grade 2 with most movement between groups occurring between Kindergarten and Grade 1. Analyses of variance indicated that the groups differed in reading achievement at each testing time, with the Double Deficit group obtaining the lowest scores. Conclusions The results indicate that Double-Deficit group membership remains relatively stable between Kindergarten and Grade 2 and that group membership is related to concurrent and future reading achievement. The results suggest that identification in Kindergarten may be helpful, especially for the Double Deficit group. These findings could have important implications for early assessment and intervention practices.
Esther Steenbeek-Planting (Radboud University Nijmegen, Behavioural Science Institute); Wim van Bon; Robert Schreuder - Aptitude-Treatment Interactions in improvement of reading speed at the word level
Purpose A randomized controlled trial with 79 Dutch poor readers assessed the effects of a flashcards training. The training used phonologically transparent CVC words. Two trainings were contrasted, either focusing on the repeated reading of successes (S-group) or failures (F-group). Method Children received 10 training sessions. Each session, children read 100 words randomly selected from a training set consisting of 845 words. Words were flashed one at a time. In the S-group, incorrectly read words were eliminated from the training set and in the following session, another 100 words were randomly selected from the reduced training set. Conversely, in the F-group, correctly read words were eliminated from the training set. Results Significant aptitude-treatment interactions were found in reading skills related to trained words and words with more complex orthographical structures (CCVCC, multi-syllable words and pseudowords). Children with a low reading performance (based on the number of words read correctly per minute) improved their reading speed more in the S-group than in the F-group. In contrast, children with a high reading performance improved their reading speed more in the F-group than in the S-group. Additionally, children with a low reading performance increased reading speed at the sentence level more than children with a high reading performance. No speed-accuracy trade off effects were observed. Conclusions To improve reading speed at the word level, low performing poor readers should focus on successes, and high performing poor readers should focus on failures. Improved reading speed transfers to untrained, more complex words.
Claudia Steinbrink (Department of Psychology II, University of Kaiserslautern, Germany); Katarina Groth; Axel Riecker; Irene Muthmann; Thomas Lachmann - Developmental dyslexics show deficits in the processing of temporal auditory information in German vowel length discrimination
Purpose: The present study investigated auditory temporal processing in developmental dyslexia using a vowel length discrimination task. Method: Both temporal and phonological processing were studied in a single experiment. Seven German vowel pairs differing in vowel height were used. The vowels of each pair differed only with respect to vowel length (e.g., /a/ vs. /a:/). In German, vowel length is characterized by temporal and spectral information. Three types of differences between long versus short vowels were varied: In the phonological condition, pairs of natural vowels were used, differing in both their temporal and spectral content. In two temporal conditions, in contrast, a natural vowel was always combined with a manipulated one to keep spectral content of long and short vowels identical. Thus, the only distinguishing feature between the two vowels was temporal in nature. Vowels were embedded into pseudo-words and presented successively in a speeded same-different task. Twenty dyslexics and twenty age-matched controls participated in the experiment. Results: In both groups, discrimination accuracy decreased with increasing vowel height in the two temporal conditions. This result is consistent with former findings on the relevance of temporal information for vowel length identification in German and extends this topic to cover discrimination demands. In the phonological condition, groups did not differ in discrimination accuracy. In both temporal conditions, however, dyslexics performed worse than controls. Conclusions: These results suggest that developmental dyslexia is associated with impairments in processing basic acoustic parameters of the speech signal, in particular, with a deficit in temporal processing.
Raj Stewart (Kennedy Krieger Institute); Aparna Pisupati; Matthew Ryan; Lindsay Goldberg; Katherine Young; Sarah Eason; Sheryl Rimrodt; Laurie Cutting - Neural correlates of sentence type in adolescent readers
Purpose: Previous studies have provided insights into the functional neurobiology of word recognition in adolescents; however, fewer studies have focused on the neural correlates of sentence comprehension in adolescents, especially neural correlates of sentence length and complexity above and beyond word recognition. This study focused on determining activation patterns underlying sentence comprehension versus word recognition in adolescent readers, while also investigating the role of sentence complexity. Method: Twenty-five typically developing adolescent readers, 9-15 years old, viewed sentence and individual word stimuli in a 3.0T scanner. Sentence stimuli included meaningful and non-meaningful (containing semantic or syntactic errors) sentences ranging from 3 to 7 words in length; word stimuli were real words and pseudowords. Participants were asked to make a comprehension decision after each stimulus. Data collected were used to generate Statistical Parametric Maps to examine stimuli-specific activation patterns. Results: Similar to previous findings, areas including the left supramarginal gyrus and inferior temporal gyrus showed activation for meaningful sentence stimuli. Greater activation for nonmeaningful sentences than for meaningful sentences was observed in the left superior frontal gyrus. Also, analysis of word stimuli showed increased activation compared to all types of sentence stimuli areas including the right dorsolateral prefrontal cortex and supramarginal gyrus. Conclusions: Overall, findings were consistent with previous studies, and showed varying response as related to sentence type. Subsequent analyses of this data, as well as data from a concurrent group of reading-disabled adolescents, may demonstrate further specificity of activation patterns.
Sue Stothard (); Maggie Snowling; Paula Clarke; Emma Truelove and Charles Hulme - What are the main points? Assessing summarisation skills and reading comprehension ability in 11-16 year olds
Purpose: This study explored individual differences in the development of summarisation skills in a large representative sample of adolescent readers. Method: 41 UK schools representing a range of socio-economic and geographic backgrounds participated in the project. A nationally representative sample of 1277 students aged 11-16 years were asked to read a series of fiction and non-fiction passages. At the end of each passage they were asked a range of comprehension questions which tapped literal, inferential (cohesive, knowledge-based, elaborative, evaluative and predictive inferences) and vocabulary-based comprehension skills. Each student was then asked to give a short summary of the passage, making clear what the main points were. Students were also given a test of single word reading allowing the classification into three groups: dyslexia, poor comprehender and typically developing. We assessed the relationship between the content of the summaries and different types of inferencing skills in the three groups. Results: Analyses explore individual differences in the development of summarisation skills from age 11 to 16 years and the relationship between summarisation skills, inferencing and comprehension skills. We also present qualitative analyses of the summarisation transcripts of good and poor comprehenders, and on the reliability of the summarisation measure. Conclusions: The ability to summarise what has been read is a key component of reading comprehension. We will examine the theoretical and practical implications of the current findings and suggest summarisation can be used as an alternative measure of reading comprehension skill.
Ron Stringer (McGill University); Brenda Linn; Olivia Leung; Anna Tirovolas - Relative contributions of fixation durations and overall RAN performance times reading skill.
Purpose: Performance on Rapid Automatised Naming Tasks (RAN) is known to be related to reading fluency. Until recently, however, we have had no way of determining what aspect of continuous RAN causes difficulty for poorer readers. Eye-tracking technology now allows us to observe directly the eye-movements of readers engaged in RAN tasks. Preliminary findings and clinical observation suggested that differences in RAN times might reflect less efficient scanning patterns in poorer readers. Method: The eye-movements of 34 university students (12 male, 22 female, average age 24.1 years) were tracked as they performed a continuous form of the RAN letter-naming task. Measures of reading achievement were also administered. Results: Completion time and average progressive fixation duration correlated with each other and with all reading variables; however, progressive fixation durations proved the better predictor of single word reading, and accounted for more unique, extra-phonological variance. Fixation duration interacted with position in the visual array in a complex way, suggesting multiple interacting visual effects on stimulus processing. Discussion: The fact that most of the predictive power of the RAN score stemmed from the length of time readers spent examining the individual stimuli suggests that scanning patterns are not the principle source of variance tapped by RAN. The actual distance the eyes travelled in regressive fixations was strongly correlated with IQ, and with reading comprehension, but not with single word reading. The implication of these findings is that continuous RAN may tap more than one extra-phonological process related to reading skill.
Nadia Taibah (King Abdulaziz University)John Everatt, Gad Elbeheri, Abdessatar Mahfoudi, Mosaad Abu Al-Diyar - Word decoding and reading comprehension in Arabic speaking children
The current study investigated the contribution of phonological processing to nonword decoding accuracy (NWRA), fluency (NWRF), and comprehension fluency (CF) in two Arabic countries, Saudi Arabia (SA) and Kuwait (KU). Two studies were conducted, first one was done in both countries with 450 children (divided into 3 cohorts, 1, 2-3, 4-5 graders) investigating NWRA and NWRF; follow up study was conducted in SA with 200 children (divided into 2 cohorts 2-3, 4-5 graders) investigating NWRF. Cohorts represent the nature of reading Arabic texts where young children would move from reading transparent text that includes short-vowel markers to more opaque text where these diacritics disappear. Independent measures included phonological awareness (PA), rapid naming (RAN), and phonological memory (PM). Analyses of variance showed significant differences between SA and KU at different grades in all dependent measures. Regression analyses showed that PA with both KU and SA children had more predictive power than RAN in NWRA and NWRF where R² of PA increased with age. However, role of RAN with KU 2-3 graders in NWRF measure is almost equal to role of PA. Because CF requires several reading skills, the predictive power of PA and RAN were almost equivalent regardless of age. These findings support the Psycholinguistic grain size theory whereby children relay on using grapheme-phoneme recoding strategies in processing literacy in transparent languages. The results were discussed considering the variability between the two countries in dialectics and methods of teaching reading.
Jo Taylor (Medical Research Council Cognition and Brain Sciences Unit);Matt Davis - Does semantic knowledge support irregular word reading? A meta-analysis of neuroimaging data.
Purpose: In the triangle model of reading aloud (Plaut et al., 1996), semantics supports irregular/inconsistent word reading. This position is supported by data from typical adults (Strain et al., 1995) and neuropsychological patients (Woollams et al., 2007). This meta-analysis of neuroimaging experiments tested this idea. Method: Twenty PET/fMRI experiments fit our inclusion criteria and compared activity during word and pseudoword reading. Four contrasted irregular and regular words. The activation likelihood estimate (ALE) technique identified reliable areas of activation which we compared with areas implicated in semantic processing in Binder et al.'s (2009) meta-analysis. Results: A left hemisphere cluster, including the inferior frontal gyrus (IFG) and insula, was more active when reading irregular than regular words. This cluster did not overlap with areas consistently more active for words than pseudowords, many of which were implicated in semantic processing by Binder et al.. Instead, the ALE analysis revealed greater IFG activation for pseudowords than words. Conclusions: Results were consistent with the idea that word reading activates semantics. However, there was no evidence for enhanced semantic processing during irregular word reading. Instead, both irregular and pseudo words were associated with increased responses in the IFG. We suggest that this reflects phonological output demands; compatible with both the dual route (Coltheart et al., 2001) and triangle models. Pseudowords require increased phonological computation and irregular words evoke conflict between phonological outputs from sublexical/orthography-phonology and lexical/semantic pathways. Simple contrasts in neuroimaging studies fail to isolate reading pathways but rather highlight phonological computations common to both routes.
William Teale (University of Illinois at Chicago) - Discussant
Jurgen Tijms (IWAL Institute); Anika Bexkens; Wery van den Wildenberg - Rapid automatized naming in children with dyslexia: are inhibitory control mechanisms involved?
Purpose: To examine whether inhibitory control mechanisms are involved in the association between Rapid Automatized Naming (RAN) performance and the reading skills of children with dyslexia, as well as their responsiveness to treatment. Method: 81 children with dyslexia and 30 normal readers were examined on measures of RAN, response inhibition (interference control and motor control), general processing speed (choice reaction time), phoneme awareness (PA), and reading and spelling. A subsample (n=45) of the dyslexic group received a phonological-based intervention for dyslexia, and was post-tested on reading and spelling. Results: Dyslexic readers differed from normal readers on PA and RAN, but not on inhibitory control or processing speed. Interference control (but not motor control), general processing speed, and PA all contributed significantly to RAN performance. Hierarchical regression analyses showed that RAN did not make a significant contribution to reading any more, after reading variance related to PA, processing speed, and interference control was partialled out. Both interference control and processing speed were significantly associated with treatment responsiveness. RAN did not explain additional variance in treatment effectiveness in reading. Conclusions: The relation between RAN performance and reading in dyslexia can be explained by shared variance accounted for by PA, inhibitory control, and general processing speed. Dyslexic children differ from normal readers on PA, but not on inhibitory control or processing speed. Both inhibitory control and general processing speed have negative impact on treatment effectiveness, presumably by different mechanisms. The implications of these results on the double deficit hypothesis will be discussed.
Anna Kristina Tirovolas () - Extending the RAN task to musical pitches and durations
(willing to present an interactive spoken paper) Purpose The rapid automatized naming (RAN) task is a standardized instrument often used in the assessment of text reading fluency. In this study, we sought to create an equivalent task aimed at assessing music reading fluency. Here we report on the creation of two new musical rapid automatized naming tasks, one for pitches, and one for durations. Inter-correlations between standard and musical RAN tasks were examined. In addition, RAN performance was used to predict fluency in a sight-reading performance of piano music. Method We created two musical versions of the RAN task, employing the common 5 x 10 matrix format, and pitch and duration symbols as stimuli. RAN performance was measured using a digital voice recorder, and operationalized by using the inter-onset intervals of vocal naming responses. Musical sight-reading performances were recorded using a Disklavier piano. Our participants were 30 pianists, aged 18 to 36. Results Analyses revealed correlations amongst all RAN tasks, including music and standard RAN. Preliminary findings, however, show that the music RAN is a better predictor of fluency in a musical sight-reading performance than standard RAN tasks. Significant positive associations were also observed between musical training and performance on the music RAN, as well as the colour RAN. Conclusions These results point out the importance of domain-specific processing in different versions of the RAN task. Further research is needed to address the potential application of this new RAN task in the assessment of music reading dis/ability.
Xiuli Tong (Dalhousie University); Catherine McBride-Chang - Uncovering young children's spelling strategies: An analysis of spelling error patterns in Chinese
Purpose: This study examined young Chinese children's strategies for spelling two-character Chinese words by analyzing the error patterns of 171 Hong Kong Chinese children's spelling mistakes in one word spelling test administered at ages 7 and 8. We also looked at these errors in relation to subsequent rates of spontaneous spelling errors in Chinese essay compositions at age 9. Method: In this longitudinal study, children were first asked to spell 20 two-morpheme Chinese words with different phonological, orthographic and semantic cues, among other tasks at time 1 and 2. They were then asked to write a short Chinese composition at time 3. Results: Young children's spelling errors could be categorized into phonologically, orthographically, and morphologically-based errors. Of these, 97% and 95% of all errors were either morphologically- or orthographically-based at times 1 and 2, respectively, whereas 3% and 5% of all errors were phonologically-based errors at times 1 and 2. In the essay-writing follow-up, only morphologically and orthographically based errors were detected. At the same time, however, a number of these errors had to do with homophone substitutions. This underscores the fact that phonological confusions occur in Chinese, but that they tend to have more to do with homophones at a holistic level, rather than grapheme errors at the level of a phonetic radical or sound-based segment, as in English. Conclusions: These findings suggest that young Chinese children use multiple cues in spelling Chinese words, but that the morphological cue is the most prevalent strategy.
Minna Torppa (Jyvaskyla University, Department of Teacher Education)Rauno Parrila; Pekka Niemi; Anna-Maija Poikkeus; Marja-Kristiina Lerkkanen ; Jari-Erik Nurmi - Double Deficit Hypothesis Tested in a Large Finnish Sample
Purpose. We examined the independence of phonological awareness (PA) and rapid naming (RAN) by evaluating the evidence for double-deficit hypothesis (DDH). Based on DDH (Wolf & Bowers, 1999), the presence of four groups is expected: one with no deficits, two with single deficits on either PA or RAN and one with double-deficits on both PA and RAN. To date, the double-deficit hypothesis has been examined predominately in English, with smaller sample sizes, and only few studies have been longitudinal (e.g., Kirby et al., 2003; Manis et al., 2000; Wolf & Bowers, 1999). Method. We followed 1800 Finnish children from Kindergarten to Grade 2 (four assessments). Double deficit subgroups were formed on the basis of PA and RAN performance in Kindergarten and subgroups' reading and spelling development in Grades 1 and 2 were compared. Results. The data are currently analyzed. Preliminary analyses suggest that subgroups imposed by the DDH can be identified even after controlling for letter knowledge. With a 15th percentile cut-off the following deficit subgroups were identified: a Double-Deficit group (n=62), a Phonological Deficit group (n=107), a Naming Deficit group (n=157), and a matched Control group (n=200). As expected, all deficit groups had difficulties in reading and spelling tasks, with the Double-Deficit group experiencing the most severe and persistent difficulties. The Naming Deficit group had more fluency than accuracy problems, whereas the Phonological Deficit group showed the opposite pattern. Conclusions. The findings expand DDH literature to orthographically consistent languages and provide information on early identification of children with risk for reading difficulties.
Mirjam Trapman (University of Amsterdam)Jan Hulstijn; Amos van Gelderen; Roel van Steensel - Determinants of literacy in monolingual and bilingual at-risk adolescents
Purpose: For adolescents' school and future professional careers adequate reading and writing skills are important. However, many adolescents, especially adolescents in multilingual contexts, fall into severe arrears in their literacy development. This study addresses the role of individual attributes in the literacy skills of monolingual and bilingual at-risk students. Method: A number of literacy-related variables and their relationships to text comprehension and written composition are examined. We measure among others word decoding skills, linguistic skills and metacognitive knowledge. We relate performance on these tests to scores on reading comprehension and writing tasks in order to uncover predictors of literacy of 63 Dutch monolingual and bilingual at-risk students in grades 7 to 9. Results: This contribution presents the associations between the variables and discuss the findings, offering explanations for associations observed. In contrast with the bilingual students, for the monolingual students decoding skills are significantly related to reading comprehension. However, for the bilingual students, there is a stronger effect of vocabulary and grammar on reading comprehension. Further explorations into the role of L1 (home) language use and skill of the bilingual students suggest that being literate in L1 and using that L1 frequently at home is no advantage for their L2 skills in reading and writing. Conclusions: The findings are relevant for theories concerning the role of lower and higher order literacy skills in reading and writing development, in particular of at-risk adolescents. In addition, the data provide new insights in the role of L1 literacy for L2 literacy development.
Susannah Trotter (); Kate Cain - The role of discourse skills and working memory in the reading comprehension of dyslexic children
Purpose - In contrast to our knowledge about the phonological skills of children with dyslexia, we know little about their working memory and discourse-level comprehension. Children with dyslexia typically show poor short-term and working memory when phonological coding is required for good performance. When non-phonological coding and rehearsal can be used to encode and maintain verbal stimuli, difficulties are not reliably found (Pickering, 2006). In addition, some poor decoders have age-appropriate reading comprehension (Catts et al., 2006). The current research investigated two questions: Do children with dyslexia have difficulties with discourse comprehension when word reading is minimised? Do short-term and/or working memory relate to any discourse comprehension difficulties? Method - Dyslexic children (aged 8-10 years) were selected from mainstream primary schools and individually matched to chronological-age (nonverbal IQ) and reading-age controls (word reading accuracy and nonverbal IQ). They completed assessments of reading and listening comprehension, inference and integration, comprehension monitoring, knowledge of story structure, short-term verbal memory, and two versions of a verbal working memory task, one designed to minimise non-phonological coding. Results - Initial analyses indicate that the dyslexic group differed from chronological-age controls on all comprehension measures. Significant differences were not evident on any of the memory measures. Conclusions - Poor listening comprehension suggests that the dyslexics' reading comprehension difficulties are not wholly attributable to poor word reading. Their weaknesses on a range of higher-order discourse measures together with age-appropriate memory suggest that i) their difficulties with discourse comprehension are broad, and ii) these may not be related to underlying memory problems.
William Tunmer (Massey University); Wes Hoover; James Chapman - Vocabulary knowledge and the development of word recognition skills: The missing link?
Purpose: This study aimed to extend recent research on the role of oral vocabulary knowledge in the development of word recognition skills. Vocabulary was hypothesized to influence word learning indirectly through set for variability, the ability to determine the correct pronunciation of mispronounced spoken English words derived from regularized pronunciations of irregularly spelled words, the incorrect pronunciation of words containing polyphonic spelling patterns, and approximations to correct pronunciations based on the application of context-free spelling rules. Method: An original cohort of 152 beginning readers participated in a longitudinal study of factors associated with early literacy achievement. Reading and reading-related measures were taken at the end of Year 1 (M age = 5.75 years), middle and end of Year 2, and middle of Year 3. Tests included measures of phonological and syntactic awareness, oral vocabulary, set for variability, decoding, word recognition, exception word reading, and reading comprehension. Results: Multiple regression and path analyses indicated that vocabulary knowledge influenced future growth in exception word reading indirectly through set for variability; that vocabulary and phonological awareness made independent contributions to variance in set for variability; that Year 1 vocabulary directly influenced Year 3 reading comprehension and indirectly influenced Year 3 decoding and word recognition through set for variability; and that Year 1 set for variability influenced Year 3 reading comprehension indirectly through both decoding and word recognition, controlling for autoregressive effects. Conclusions: The findings suggest that vocabulary has a direct effect on reading comprehension but indirectly influences the development of word recognition skills through set for variability.
Shih-Jay Tzeng (National Taitung University, Taiwan); Shu-Li Chen; Shu-Jan Lin; Yu-Zen Chang - The effectiveness of two Chinese remedial reading programs on the intervention outcomes of Taiwanese first- and second-grade poor readers
Purpose Many anxious parents want to see the outcome of reading intervention in a short time. Teachers and children are often under pressure. We compare the effectiveness of 2 remedial Chinese reading programs in order to help the parents with program selection. Method Participants are 124 first- and second-grade poor readers. Students were assigned to 2 different literacy intervention programs, namely, the AUX (24 Ss) and the ADJ group (30 Ss). The AUX group received a systematic and structural reading instructional kit in which the materials taught were independent of the school reading textbook. The ADJ group used simplified reading materials which was adapted from the textbooks used by students' original class. A control group of 70 students received an after-class homework-assistance program. The experiment provided 1 on 2 tutoring 4 sessions a week for 15 weeks. Pretests, posttests of a Chinese reading competence battery, midterm and final examination scores of reading, and IQ were collected before and after the intervention. Results 1. Both experimental groups showed better reading improvement than the control group. 2. When comparing midterm and final scores, the ADJ outperformed the AUX. However, when comparing standardized reading competence tests scores, the AUX obtained significantly better progress than ADJ. Conclusions In ADJ program, intervention outcome will reflect in midterm scores in as short as 40 days. This makes ADJ a good selection for anxious parents. If parents are patient enough, we recommend the AUX program because it builds up Chinese reading competence in a more systematic way.
Sonja Ugen (Université du Luxembourg); Martin Brunner; Sylvie Bodé; Jacqueline Leybaert - Development and cross-linguistic transfer of specific and general reading related processes
Purpose Acquiring literacy skills involves specific processes such as matching letters to sound as well as more general processes on a semantic level. The aim of this study was to look at the development of these broader reading-related concepts longitudinally in two languages (i.e. German and French) within biliterate children at the primary school level. Method In Luxembourg biliteracy is acquired through education, as children acquire reading skills in German in grade 1 and start written French classes in grade 3. Sixty-four children attending the national educational curriculum since kindergarten participated in the study from grade 2 (mean age = 7:7) until grade 4 included. Measures of specific reading processes included reading and spelling tasks in German and in French over three school years. Indicators of more general processes involved reading comprehension, receptive grammar and receptive vocabulary in German and in French measured after two years of instruction in each language. Results General and specific processes were each included in two separate structural equation models.In both models, the hypothetical constructs of specific and general reading-related processes were affirmed. For general processes no influence from the first acquired language on the second acquired language was found. For specific processes, results indicate influences from one academic year on the following within each language, and moderate influences from German to French. Conclusions There seems to be no cross-linguistic transfer from the first to the second acquired language for general reading-related processes, but there is evidence for cross-linguistic transfer for specific reading-related processes.
Joanna Uhry (Fordham University)Joanna Uhry - Can finger-point reading be taught in kindergarten? And if so, does it benefit emergent literacy?
Purpose. Finger-point reading (FPR) involves matching oral language in memorized poems and songs to print. According to Morris (1993; Morris & Perney, 1987; Morris et al., 2003), children must understand concept of word to tag words correctly. There is converging research (e.g., Ehri & Sweet, 1991; Morris et al., 2003; Uhry, 1999; 2002) indicating a strong association between FPR ability and other early literacy skills (e.g., invented spelling, phonemic awareness). While FPR is widely modeled in kindergarten, there is virtually no research on its effects. It is the purpose of this training study to see if FPR can be taught, and if so, whether this benefits other emergent literacy skills. Method. Stratified random assignment of 69 pre-readers was used to form three groups of urban kindergarten children. Invented spelling, CTOPP, word reading, and a videotaped FPR measure were used for grouping and post-testing. Individual intervention for Group 1 (FPR) involved finger-point reading poems and songs with the focus on one-to-one correspondence and letter-sound matching. Children in Group 2 (PA) were taught to listen and repeat poems with focus on segmenting onsets and matching words by alliteration without looking at print. Children in Group 3 (classroom controls) worked with their teachers on writing projects. Results. While there were no group differences at pre-test, both intervention groups outperformed controls at post-test. The FPR group outperformed both other groups on post-test FPR and invented spelling. Conclusions Findings justify the use of FPR as a kindergarten instructional method. Close observations of the videotapes indicate patterns of difficulty focusing on internal letters in multisyllabic words, consistent with Ehri's partial letter cue stage.
Akira Uno (University of Tsukuba);Noriko Haruhara;Masato Kaneno;Noriko Awaya;Takashi Gotoh;Junko Kozuka - Essential recognitive disabilities for Kanji writing and/or reading in Japanese children with developmental dyslexia
[Purpose] The aim of this study was to investigate the essential recognitive abilities for Kanji reading and/or writing in Japanese children with developmental dyslexia and agraphia. [Methods] 612 Japanese-speaking primary school children with normal reading ability (N), 23 children with developmental dyslexia, who showed both reading and writing disorders(RW), and 11 children with developmental dysgraphia, who showed only writing disorders(W), participated in this study. Tests for visual recognition, and visual short and long-term memory, and tests for phonological awareness, rapid automatized naming task, oral reversal task of words, and auditory long-term memory were administered to the children with RW and W group. The RW and W group revealed impaired performance on the phonological and visual information processing, while they showed normal FIQ using WISC-Ⅲ. [Results] The possible predictors for Kanji reading were vocabulary size and the ability of phonological awareness, while those for Kanji writing were visual recognitive ability and memory, and vocabulary size by multiple regression analysis in N group. The score of vocabulary test and phonological awareness task in W group were significant higher than those in RW group. The score of visual memory task in both W and RW group were significantly lower than those in N group. [Conclusion] These results suggested that there is very important contribution of vocabulary for Kanji reading in normal and dyslexic group, while visual memory ability is related to the attainment of writing especially in children with writing disorders with/without reading disorders.
Anniek Vaessen (Maastricht University); Leo Blomert; Daisy Bertrand; Dénes Tóth; Valéria Csépe; Luís Faísca; Alexandra Reis - Cognitive development of fluent reading skills and the modulating influence of orthographic depth
Purpose: Most reading development models assume a shift from sub-word decoding to automatic word recognition. We hypothesized that this shift is reflected in a concomitant shift in the relative importance of the cognitive skills underlying reading. Study 1 tested this hypothesis in a large Dutch sample ranging from beginner to experienced readers. Study 2 investigated whether orthographic depth of a language influences this developmental cognitive shift. Method: study 1: 1430 Dutch children (grade 1 to 6). Dominance analyses were used to investigate the relative importance of rapid naming (RAN) and phonological awareness (PA) to reading fluency of high-frequency, low-frequency and pseudowords at different reading expertise levels. Study 2: 674 Hungarian, 954 Dutch and 616 Portuguese children (grade 1-4). LMER analyses were used to investigate the modulating influence of orthographic depth on the cognitive development of reading fluency. Results: PA and RAN both significantly contributed to reading fluency in all grades and in all orthographies, but PA contributions decreased as a function of grade, while RAN contributions systematically increased. Moreover, this cognitive shift was most pronounced for high-frequency words. Orthographic consistency systematically modulated the strength of the cognitive contributions of PA, but not the overall developmental pattern. Conclusions: the relative involvement of the cognitive components underlying reading seems to depend both on word familiarity and reading experience. Furthermore, the cognitive development of reading skills is fairly similar across alphabetic languages and differences in orthographic depth will not recruit different cognitive processes, but will mainly be expressed in rate of reading development.
Jyotsna Vaid (Psychology Dept., Texas A&M University); Hsin-Chin Chen(National Chung Cheng University, Taiwan) - Cross-language differences in sound segmentation show a robust effect of the first learned orthography: evidence from Hindi-English users
Purpose Recent research has shown that judgments of the "initial sound" are highly phoneme-based in readers of English but are predominantly syllable-based in readers of Indic scripts (Hindi, Kannada, and Malayalam). The present research sought to examine the relative influence of the first learned language (Hindi vs. English) and length of use of each language (prolonged use of English vs. of Hindi) on readers' judgments of "the initial sound" of Hindi and English homophones. Method Participants consisted of two groups of fluent Hindi-English bilingual adults (one with prolonged exposure to English and the other with more exposure to Hindi) and two groups of college-age English learners of Hindi. All were auditorily presented with a list of homophonic words in the two languages and were to take away the first sound and report aloud what remained. Percent choice of phoneme as the unit deleted was analyzed as a function of group, language set and stimulus type (mono- vs. bisyllabic). Results Phoneme deletion was higher for native English- than for native-Hindi readers, for mono-syllabic than bisyllabic words, and for words presented in the English set than in the Hindi set. Interestingly, cross-language differences in segmentation strategy were most pronounced in the Hindi readers with greater (rather than less) exposure to English. Conclusions The results suggest that there are robust differences in how words are segmented in bilingual users of languages that differ in their orthographic design principles. Furthermore, these differences persist (and may even heighten) after several years of exposure to a monolingual/monoliterate (English) language environment.
Elsje van Bergen (University of Oxford, Experimental Psychology); Peter F. de Jong; Anna Plakas; Judith Bekebrede; Aryan van der Leij - Examining the association between children's reading outcome and parents' literacy levels within families with a background of dyslexia
Purpose - Previous longitudinal studies following children at family risk for dyslexia have compared groups of at-risk dyslexic, at-risk non-dyslexic, and not at-risk control children. Little attention has been devoted to the reading skills of the parents of these groups of children. In a previous study (van Bergen et al., in preparation), we found a difference in reading fluency between the dyslexic parents of the at-risk dyslexic and at-risk non-dyslexic children, which might indicate a difference genetic liability between these groups. In this study we explore parental differences between the two at-risk groups in a new sample. Method - Dutch children with (n = 50) and without (n = 15) family risk have been followed from birth. The children will be classified as dyslexic or not at the end of 2nd grade. The parents of the three groups of children will be compared on measures of reading fluency, spelling, phonological skills, and verbal IQ. Results - The reading skills at the end of 2nd grade of all children will be known by June 2010. Preliminary analyses show that the parents of the control group outperform both at-risk groups on measures of reading and phonology. In contrast, the differences between the two at-risk groups are small. No group differences on verbal IQ are found. Conclusion - Parental differences between the two at-risk groups seem to be less pronounced than in our earlier study. In the latter, children's reading status was assessed later (i.e., in 5th grade), which might (partly) explain the different findings.
Victor H.P. van Daal (Edge Hill University); Vibeke Rønneberg; Malin Wass - Learning to read and spell in a second language: Effects of the orthographic depth of L1
Reading and spelling is harder to learn in some languages than in others, because of differences in orthographic depth. This paper examines whether differences in orthographic depth of L1 has repercussions for the learning of L2. In particular, we looked at whether Danish children were slower at learning to read and spell in English than their Norwegian and Swedish counterparts. We hypothesised that, due to the less transparent orthography, Danish children would be slower in becoming an independent reader, and would rely on visual memory skills. The Norwegian and Swedish children on the other hand were expected to rely more on phonological skills and use these to build up an orthographic lexicon. Participants were assessed on reading and spelling in L1 and L2 with frequency-balanced cognate word lists and frequency-balanced word lists, respectively. L1 vocabulary was used a covariate to ensure that the samples were comparable with respect to home and school background variables. Orthographic skills were assessed with a novel-word learning task. Visual memory skills were measured with a matrix test. The results indicate that Danish readers in all age bands were slower and less accurate in naming L1 and L2 words, and made more spelling errors, again in both L1 and L2, than the Norwegian and Swedish children. Furthermore we found that Danish children relied more on their visual memory skills when reading high-frequency words, whereas Norwegian and Swedish children used their orthographic skills, even when reading low-frequency words.
Wim Van den Broeck (Vrije Universiteit Brussel)Astrid Geudens - Poor nonword-reading is not characteristic for disabled reading: An empirical study using an alternative matching methodology.
This contribution presents empirical evidence challenging the received wisdom that a nonword-reading deficit is a characteristic trait of disabled readers. As we have demonstrated in our Developmental Psychology article (in press), the finding of a nonword-reading deficit in the context of the classic Reading-Level-Match (RLM) design can be explained as the consequence of normal developmental changes in word-specific knowledge. In this paper we propose an alternative methodology to study the nonword-reading deficit in which the problem of confounding variables is solved for. In order to match the subjects simultaneously on age and reading level, words and nonwords of varying difficulty were presented to normal and disabled reading subjects of the same age. Method Using state-trace analysis normal readers and disabled readers were effectively equated on reading level and on age. In order to test the effect of lexicality, each word was rigorously matched with a nonword on a number of structural word characteristics; i.e. word length and positional redundancy, and on three features that were also used to manipulate word difficulty: number of syllables, number of consonant clusters and body frequency. By crossing these three features and lexicality, a 3x3x2x2 within subjects design was created. Results and Conclusions The state traces of both groups proved to be identical, indicating no evidence for a nonword-reading deficit. The proposed method can also be used to test other hypotheses concerning typical effects of disabled reading.
Paul van den Broek (Leiden University, The Netherlands);Andrew Elfenbein;Sarah Carlson;Mark Lewis - Expository text reading comprehension: Evidence from eye-tracking
Purpose In the current study we investigate the cognitive processes that occur during comprehension of expository texts. In particular, we contrast different types of expository text structures (Compare/contrast, problem/resolution, chronology, and description) using eye movements (supplemented by think-aloud data) to identify unique processing patterns. Methods 64 (26 male; 38 female) college-aged students read texts of each subtype while their eye movements were recorded. Participants were asked to recall each text. In addition, students read texts of each subtype while performing think-aloud protocols. Results With respect to specific types of eye movements, we considered forward fixations which reflect the initial reading, reinspections within the same sentence which reflect intra-sentential integration processes, and look backs from subsequent sentences which reflect text-level integration processes. The compare/contrast texts elicited longer fixations/character for first pass F(1,61) = 25.952, p < .001, reinspections F(1,61) = 5.421 p =.023), and look backs F(1,61) = 11.935, p = .001. Chronological texts elicited shorter fixations/character for first pass F(1,61) = 59.421, p < .001, and reinspections (F(1,61) = 19.885, p < .001). Detailed follow-up analyses of the specific locations within each text type that elicited the above patterns -informed by the think-aloud data- revealed the unique patterns of processing associated with each text type. Conclusion The results show that within expository texts distinct subtypes exist. Each of these subtypes is characterized by unique patterns of cognitive activities -as revealed by eye movements. These patterns reflect the unique integration processes that comprehension of each subtype requires.
Julie Van Dyke (Haskins Laboratories); Clinton L. Johns; Anuenue Kukona - A retrieval interference account of poor sentence comprehension in young adult readers
Purpose: To evaluate divergent hypotheses explaining memory-related individual differences in reading comprehension. The working memory capacity approach posits a single, limited pool of resources that supports both processing and maintaining text-relevant information. Individuals with smaller capacity may exhibit impaired reading comprehension because they are unable to efficiently balance both processes. However, this explanation cannot account for a large body of research investigating the speed with which items are retrieved from memory. This research indicates that the amount of information that can be actively maintained during sentence processing is substantially smaller than memory span tasks predict. Here, the source of individual differences must relate to the memory retrieval mechanism. This account is supported by recent evidence that readers are vulnerable to retrieval interference during sentence comprehension. Method: A community-based sample of 65 young adults (age 16-24) completed a battery of individual difference measures and a self-paced reading experiment. In the experiment, participants read complex sentences that varied according to the factors memory load and interference and answered comprehension questions. Results: After factoring out IQ, readers with low capacity were unimpaired relative to those with high capacity. Readers with poorer receptive vocabulary skills showed greater interference effects and impaired sentence comprehension. Conclusions: Our findings indicate that capacity per se is not predictive of sentence comprehension skill. In contrast, receptive vocabulary skill was the most important predictor of variation in reading comprehension in our dataset. Our results suggest that the quality of readers' lexical representations determines retrieval success and comprehension accuracy.
Claudia van Kruistum (PhD Student, Universiteit Utrecht); Ilona de Milliano; Paul Leseman; Peter Sleegers; Amos van Gelderen; Roel van Steensel - The importance of attitudes and out-of-school reading in the promotion of reading skills of at-risk adolescents
Purpose: Generally, reading attitudes, out-of-school reading and reading skills are known to be correlated. The question is whether this is also true for specific subgroups of readers. This study examines the importance of attitudes and out-of-school reading for reading skills of at-risk adolescents. Method: Attitudes were assessed with a questionnaire including statements on reading in terms of pleasure, relevance, effort and confidence. Out-of-school reading was assessed by a questionnaire in which students estimated the frequency of their literacy engagement in a broad sense; in addition, in-depth interviews were conducted to explore students' motives. Reading skills were tested with a reading comprehension test. All measures were administered to 63 7th grade students with reading arrears. Results: Correlational analyses revealed positive relations between reading attitudes, out-of-school reading and reading skills. However, the correlations are quite small and conceal that although many students have quite positive attitudes only a small part of them is frequently involved in out-of-school reading. Interview data provide possible explanations, as it was found that motives for engaging in written media ("Reading is important") are often outweighed by negative motives ("Reading is boring"). Furthermore, written media face strong competition from new media which are very popular and particularly used for social/entertainment purposes. Written media are especially used for informational, educational purposes. Conclusions: The results imply that promoting at-risk adolescents' reading skills via interventions might be more fruitful if new media activities are incorporated; students need to be motivated and guided in exploring the potential of written and new media.
Maaike Vandermosten (Centre for Parenting, Child Welfare and Disabilities, K.U.Leuven, Belgium); Bart Boets; Heleen Luts; Hanne Poelmans; Narly Golestani; Jan Wouters; Pol Ghesquière - The role of phonetic and temporal aspects of auditory processing in adults with dyslexia
Purpose: There is ample evidence that individuals with dyslexia exhibit problems in the representation of speech sounds, and that these may be rooted in a more fundamental auditory processing deficit. Despite substantial research effort, one central issue of debate concerns the nature of these auditory problems in dyslexics, in particular whether the problems are speech specific and/or temporal specific. Method: In the present study we tested 31 dyslexic and 31 matched normal reading adults on four types of stimuli: (1) a speech continuum exploiting temporal cues (/bA/-/dA/), (2) a speech continuum defined by non-temporal, steady-state cues (/u/-/y/), (3) a non-speech temporal continuum (rotated /bA/-/dA/), and (4) a non-speech non-temporal continuum (rotated /u/-/y/). Results: Based on the slope value of the identification curves, our study demonstrated that categorization is impaired in dyslexics subjects for the /bA/-/dA/ and the rotated /bA/-/dA/-continuum, but not for its non-temporal counterparts. Conclusions: By applying this balanced design with stimuli that are adequately controlled for acoustic complexity, we find evidence for a 'low-level' temporal processing deficit in dyslexics which is not speech-specific. Future work: The same subjects are involved in a diffusion tensor imaging (DTI) study which enables a non-invasive examination of the white matter bundles. As intact myelination is important for rapid conduction of action potentials, we hypothesize that a disturbance of myelination would be extremely detrimental for temporal processing. Besides investigating the group differences in white matter, we will relate these anatomical data to the behavioural data. Preliminary results of the DTI-study will be presented.
Sharon Vaughn (University of Texas at Austin); Colleen K. Reutebuch; Coleen D. Carlson; Amory Cable; David J. Francis - Enhancing social studies vocabulary and comprehension for 7th grade English language learners: findings from three experimental studies
Purpose: Adolescent ELLs who lack academic English language knowledge and demonstrate low literacy levels are at risk for academic failure in content area classes. Our goal was to examine the effects of enhanced social studies instruction designed specifically for students who are ELLs that would benefit all students. Method: Three experimental studies in three successive school years with non-overlapping samples were conducted. Three different samples of classes of 7th grade students (N=380; N=507; N=680) were randomly assigned at the classroom level to a social studies intervention or to a business as usual comparison group. The treatment intervention was comprised of: a) vocabulary instruction, b) the use of brief videos and purposeful discussion to build concepts, c) the use of graphic organizers and other writing activities, and d) structured paired grouping. All students were assessed with a researcher-developed content-based measure (CBM) at pretest and posttest. Results: In Studies 1 and 2, analysis of posttest scores revealed statistically significant differences between students in treatment and control sections for both comprehension and vocabulary. Treatment effects did not interact with student status as ELL or non-ELL indicating that ELL and non-ELL students benefitted equally from participation in treatment sections. Results for Study 3 are preliminary and will be completed by the time of the conference presentation. Conclusions: Although this intervention was developed to address instructional and language needs of ELLs, students who were not limited English proficient in the intervention classes also benefited. This finding is particularly relevant for teachers who have both ELLs and non-LEP students in their classrooms.
Anneli Veispak (PhD student at Katholieke Universiteit Leuven); Pol Ghesquière - Relating braille reading difficulties to developmental dyslexia: first empirical evidence
Purpose A proportion of individuals with visual impairment (VI) have specific reading difficulties which cannot be easily explained. The existence of braille dyslexia has been proposed. The pan-modal temporal processing deficit has been assumed to be causal to both the phonological and literacy problems observed in dyslexia. Struggling braille readers have been reported to have difficulties with phonological awareness and tactual perception, domains also affected in dyslexics. The aim of our study is to investigate the relation between auditory temporal processing and reading measures in individuals with VI in Belgium and in Estonia. Methods The participants are individuals with VI (n=approximately 30, minimal age 9 years) reading braille (normal intelligence, no gross deficiencies in audiology). Participants speak either Dutch or Estonian as their first language, they are attending or have attended local special schools. The experiments used in the study: (1) Reading-ability assessment (real-word, non-word, story-reading tasks); (2) Speech perception (speech-in-noise perception tasks: words and sentences); (3) Phonological tests (phoneme deletion, spoonerism, categorical perception task); (4) Verbal-short-term memory (non-word repetition, digit-span forward); (5) Tests for auditory processing (Gap, 2Hz frequency modulation and tone-in-noise detection tasks). Keeping in mind the limited amount of participants (although a big part of braille users in both regions) the study will have a correlational design. Conclusions Auditory temporal processing seems to be related to reading and its cognitive correlates also in braille readers. With our findings we hope to help practitioners, teachers as well as parents to find effective teaching methods for those in trouble.
Ludo Verhoeven (Active Member, Regular, Journal Editorial Board) - Speech perception, lexical retrieval and phonological awareness as precursors of early literacy.
Purpose: The goal of this study was to examine the role of speech perception, lexical retrieval and phonological awareness as precursors of early literacy. Following a longitudinal design, an attempt was made to relate the development of speech perception, lexical retrieval and phonological awareness in kindergarten to children's early literacy attainment, i.e., before formal reading instruction has started. Method: A cohort of 169 Dutch children in the Netherlands was followed and tested during the second year of kindergarten. At the beginning and at the end of the school year, a series of tests was administered in order to measure children' speech perception (5 tests) , lexical retrieval (2 tests), and phonological awareness (6 tests). By the end of kindergarten early literacy skills were assessed. The dependent variables of early literacy comprised tests on letter knowledge, cipher knowledge, word recognition, and word spelling. Results: Structural Equation Modelling showed speech perception to predict children's development of phonological awareness. Phonological awareness, on its turn, predicted the development of receptive and productive letter knowledge and word decoding. Lexical retrieval ability was found to be an additional predictor measure of letter knowledge. Conclusion: It can be concluded that the emergence of literacy skills can be predicted from children's phonological awareness and lexical retrieval abilities. In addition, speech perception abilities can be seen as a precursor of both lexical retrieval abilities and phonological awareness.
Megan Vinco (Florida Center for Reading Research- PIRT); Christopher J. Lonigan; Beth M. Phillips - Home literacy environments and children's problem behaviors mediate the influence of socioeconomic factors on children's vocabulary skills
Purpose: Socioeconomic factors may be an integral component of early literacy. Not surprisingly, socioeconomic factors often impact the literacy environment provided to young children; socioeconomic status is also sometimes associated with behavioral ratings. This investigation sought to examine how one aspect of early literacy-vocabulary-relates to preschool home literacy, child behavior, and socioeconomic factors. Method: Over 200 preschoolers across a range of socioeconomic strata were assessed for vocabulary ability on the Test of Preschool Early Literacy (TOPEL). Parents reported child behavior using the Conner's Parent Rating Scale (CPRS) and completed a questionnaire addressing home literacy and socioeconomic factors. Results: Preliminary analyses indicated that socioeconomic factors (e.g. maternal education and income), in conjunction with parental literacy practices, had the strongest associations with preschool vocabulary scores. High levels of parental endorsement for some home literacy practices were inversely related to vocabulary performance. Socioeconomic factors showed a significant negative correlation with child behavioral problems. The impact of socioeconomic status was mediated by both literacy practices and behavioral problems. Conclusions: Reinforcing the relevance of socioeconomic factors to educational outcomes, the results of this study identify two possible pathways through which socioeconomic status has an impact. Parents with greater socioeconomic status may engage in more literacy-related practices, particularly those associated with higher vocabulary scores. Further, behavioral problems, which are linked to lower vocabulary scores, are reported less frequently in children from higher socioeconomic backgrounds. Extended more broadly, vocabulary researchers may benefit from considering the indirect relation socioeconomic status has on the factors relevant to early vocabulary performance.
Janet Vousden (Coventry University); Michelle Ellefson; Nick Chater; Jonathan Solity - Comparing the content of UK reading programs using the Simplicity Principle
Purpose: A popular approach to teaching reading in the UK is based on learning and applying the connections between letter patterns and sounds - termed phonics. To date most research has focussed on comparing the methods employed rather than content of what is learned (e.g. phonemes, rimes, etc.). In this paper we investigate the potential impact that the content, rather than the method, of various phonics programs could have on reading outcome. Method: Phonics programs used in UK schools that taught predetermined letter-sound correspondences were compared with an optimal set of correspondences theoretically motivated by applying the Simplicity Principle. The measures for comparison were: the number of correspondences taught, the proportion of words that could be correctly pronounced using each program, and an overall measure of program simplicity (using the Simplicity Principle, according to which, simpler programs are preferred). Results: The number of correspondences taught varied considerably between programs, from 43 to 106. Programs that taught more correspondences generally allowed a higher proportion of words to be pronounced correctly and were simpler according to the Simplicity Principle. However, the simplest program, also a top scorer on correct pronunciations, was not the program with most correspondences. The difference between the optimal set and the programs was mostly large for correct pronunciations and simplicity. Conclusions: Potential reading outcome varies considerable, depending on which letter-sound correspondences are taught. The potential outcome of all the programs considered here could be improved by using the Simplicity Principle to select the correspondences.
Lesly Wade-Woolley (Queen's University); Michelle Kotanko - Prosodic sensitivity, reading with expression, and punctuation skill in eight- and twelve-year-old children
Educators know that "reading with expression", including attention to punctuation and prosodic features, is a key index of reading fluency and comprehension. What is unclear is whether a reader's sensitivity to the prosodic structure of her language is normally recruited as an on-line aid in reading with expression, or whether the ability to render a good prosodic reading is a simply an artifact of good word reading and comprehension skills. Therefore, the key research questions are whether prosodic sensitivity predict expressive reading and punctuation skill after controlling for word reading, reading comprehension and baseline punctuation knowledge, and how these relationships change as a function of reading skill. This study will examine the role of prosodic sensitivity in expressive reading in eight- and 12-year-old children, as well as the relationship between prosodic sensitivity and children's ability to correctly punctuate a written passage. Participants are 40 8-year-old and 40 12-year-old children who are assessed on prosodic sensitivity, reading ability and intelligence. Key experimental tasks are short, punctuated and unpunctuated paragraphs matched for readability level, in which the syntactic structures marked by punctuation should elicit the marked pause duration and pitch increment noted by Miller and Schwanenflugel (M&S, 2006). Participants are asked to read the passages aloud with "good expression"; pause length and pitch increment are measured to attain a measure of prosodic reading as was seen in M&S. Following a baseline assessment of punctuation familiarity, participants are asked to insert the correct punctuation in unpunctuated passages. Regression analyses are used to determine the relationships among variables, particularly whether prosodic sensitivity predicts expressive reading and punctuation skill over general reading ability.
Richard Wagner (Florida State University) - Relations between Components of Reading and Reading comprehension
Purpose The purpose of this study was to examine potentially causal relations between reading comprehension and the components of vocabulary, decoding, and working memory in a longitudinal study. Method Two groups of participants were included in the study: 182 students followed from grades two through four; 187 students followed from grades four through six. Multiple indicators were available for all constructs examined. We examined relations between reading comprehension and the three components using three analytic methods: structural equation modeling of longitudinal data with autoregressors; growth curve modeling; and latent change-score modeling. SEM is useful for modeling causal relations among developing constructs represented in longitudinal data, but can be problematical when modeling stable constructs such as reading comprehension. Growth curve modeling is not affected by construct stability, but does not address causal relations: Latent-change score modeling represents a new technique that adopts desirable features of both structural equation modeling and growth curve modeling. Results The SEM results confirmed the problem of high stability for modeling causal relations with the autoregressor swamping out other predictors. Growth curve modeling showed relations between the component predictors and growth curve parameters. The latent change score modeling results illuminated aspects of relations between components and reading comprehension that were not found using the other two techniques. Conclusions Latent change score models provide a useful tool for assessing relations between reading comprehension and key component skills. Keywords Reading comprehension; longitudinal
Richard K. Wagner (Florida State University, Department of Psychology);Young-Suk Kim; Danielle Lopez; Liz Foster; Yusra Ahmed; Patricia Thatcher Kantor - Developmental relations among oral reading fluency, silent reading fluency, and reading comprehension: A two-year latent variable longitudinal study
Purpose. The purpose of this study was to examine changing relations among oral reading fluency, silent reading fluency, and reading comprehension. Method. Participants were 290 students who were assessed in first- and second-grade. The sample consisted of approximately equal numbers of boys and girls. Multiple indicators were obtained for the following constructs: Oral Reading Fluency: The indicators were three oral reading fluency passages from the DIBELS. List Reading Fluency: The indicators were two forms of the Sight Word Efficiency subtest of the TOWRE. Silent Reading Fluency: The indicators were two forms of the Test of Silent Reading Efficiency and Comprehension (TOSREC). Listening Comprehension: The indicators were the WJ-III Oral Comprehension subtest and two experimental listening comprehension passages. Reading Comprehension: The indicators were passage comprehension from the WJ-III and WRMT-Revised, and two experimental passages. Results. The results of confirmatory factor analyses were that oral reading fluency and silent reading fluency represented correlated, yet distinct abilities. Differences were found both between good and poor readers and between first- and second-grade students in the relative contributions of list reading fluency and listening comprehension to oral and silent reading fluency, and in the contributions of oral and silent reading fluency to reading comprehension. Conclusions. Oral and silent reading fluency represent correlated yet distinct reading abilities. The relative contributions of components of oral and silent reading fluency vary as a function of reading skill and developmental level, as do the relative contributions of oral and silent reading fluency as predictors of reading comprehension.
Hua-Chen Wang (Department of Cognitive Science, Macquarie University /Centre of Excellence in Cognition and its Disorder (CCD)); Lyndsey Nickels; Kate Nation; Anne Castles - Story or spelling list? Context effects on orthographic learning of regular and irregular words
Purpose: The self-teaching hypothesis (Share, 1995) proposes that orthographic learning takes place via phonological decoding in meaningful texts, that is, in context. Context is proposed to be important in the early stages of reading especially when decoding is only partial. However, little research has directly explored this hypothesis. This study looked at the effect of context on orthographic learning and examined whether there were different effects for novel words given regular and irregular pronunciations. Method: Forty-one second grade children participated in the study. In order to simulate natural orthographic learning, the children were exposed to the sound and meaning of eight new words prior to orthographic exposure. After four days of pre-orthographic-exposure training, the participants were then given the eight novel words in written form, either in stories (Context condition) or a list of words (No Context condition). Subsequently, spelling and lexical decision tasks were used to assess the success of orthographic learning. Results: The results revealed a significant main effect of context for irregular stimuli on the lexical decision task, where items presented in contexts were identified correctly more often than were words presented in lists. However, no significant effect of condition was found for the regular items. Conclusions: The findings support the view that contextual information is important to orthographic learning but only when the words to be learned are irregular. However, context does not seem to have an effect on novel word learning when there is a one to one phonological mapping.
Min Wang (Dept. of Human Development)In Yeong Ko - Morphological, semantic and orthographic sensitivity in visual word recognition among Korean children and adults
Purpose: We examined the developmental changes in sensitivity to the morphological, semantic and orthographic information in Korean Hangul visual word recognition among native Korean-speaking children (grades 4, and 6) and adults as a comparison group. Method: A group of 96 native Korean-speaking children and adults (32 in each of the 3 groups) made visual lexical decisions on stem targets preceded by 4 different types of masked primes. The 4 conditions of prime-target pairs were: (1) orthographically overlapped only (-M-S+O: 도시락-도시, scandal-scan), (2) orthographically overlapped, morphologically decomposable but semantically unrelated (+M-S+O: 구두쇠-구두, archer-arch), (3) morphologically decomposable, semantically related, and orthographically overlapped (+M+S+O: 음악가-음악, bravely-brave), and (4) semantically related only (-M+S-O: 반대-이의, accuse-blame). 3 (grade: 4th vs. 6th vs. adults) x 4 (priming conditions: -M-S+O vs. +M-S+O vs. +M+S+O vs. -M+S-O) design was employed. The dependant variable was the priming effect. Results: There was a significant interaction effect between grade level and priming condition on the priming effect. Grade 4 children did not show any priming effect across the 4 conditions. Grade 6 children showed a facilitative priming effect under the +M-S+O condition but an inhibitory effect under the -M-S+O condition. Adults showed a facilitative priming effect under both +M+S+O and +M-S+O conditions and continued to show an inhibitory effect under the -M-S+O condition. Conclusion: There is a developmental shift in terms of sensitivity to morphological information in relation to semantic and orthographic information. Morphological information plays an independent role beyond orthographic and semantic information.
Gloria Waters (Boston University);David Caplan; Jennifer Michaud; Julia Bertram - Assessment of auditory and written language comprehension in middle and high school students
Purpose: Research has shown that older struggling readers have deficiencies in aspects of language processing at all levels of the language code-sub-lexical, lexical, morphological, sentential and discourse. The goal of this study was to develop a battery that assessed students' skills at all levels of the language code in the auditory and written modalities and determine the relationship of these skills to effective reading. Methods: 1,356 students in grades 6 to 10 were tested on 12 tasks that assessed language at the levels outlined above in the auditory and written modalities. Reaction times and accuracy were recorded. Within each level, materials were chosen to assess specific features of language that psycholinguistic studies have shown to be important in determining performance. Results: Significant effects of grade and of the linguistic parameters manipulated were found on each of the tasks. Factor analyses revealed very similar patterns of results in the two modalities and in the RT and accuracy data and resulted in three factors- one involved in recognizing the form of items (e.g., lexical decision), one involved in accessing meaning from complex forms (e.g., affixed words) and one in accessing meaning from simple words. Stepwise regressions showed that of all of the auditory and written factor scores the best predictor of written passage comprehension was comprehension of morphology and syntax in the auditory modality. Conclusions: The results suggest that comprehension of higher-level language structures in spoken language is the best predictor of discourse comprehension for middle and high school students.
Brendan Weekes (University of Hong Kong) - Written Age-of-Acquisition Effects Reflect Family Resemblance in the Lexical Network.
Purpose The arbitrary mapping hypothesis assumes that age-of-acquisition (AoA) effects on oral reading depend on whether phonology can be predicted from orthography. One prediction that follows is that AoA effects will be larger for written words with inconsistent rime spellings than for written words with consistent rime spellings. An alternative hypothesis is that AoA effects on oral reading depend on the family resemblance between lexical items that share orthography-to-phonology mappings. Method: 100 underdraduates read aloud over 500 English monosyllables varying in written age of acquisition, spelling rime consistency, number of rime neighbours, written frequency, imageability, length and other lexical variables. Results Multiple regression analyses show that the effects of written AoA on oral reading interact with the number of words containing the rime. Specifically, written AoA has a larger effect on the oral reading of consistent words with few family members than on that of inconsistent words with many family members. Conclusion A binary distinction between predictable and unpredictable mappings is not sufficient to explain effects of written AoA on oral reading in English.
Brendan S. Weekes (University of Hong Kong, China) - Testing the Arbitrary Mapping hypothesis across scripts: A comparison of Chinese and English
Two experiments examined the effects of age of acquisition (AoA) on word recognition in Chinese and English adult readers. In Experiment One, Chinese speakers in Beijing were asked to read aloud early acquired characters taught in elementary school texts and late acquired words that did not appear in those texts (written AoA). Items were matched for frequency using token word and cumulative frequency values but varied in the consistency of mappings between orthography and phonology. Results showed a significant effect of written AoA and evidence of an interaction between character consistency and written AoA. The findings revealed an effect of family resemblance whereby AoA effects were greater for items that had relatively few orthographically similar family members. Written AoA effects on oral reading were reduced if a printed character shared print to sound mappings with several other written characters in the lexicon. This finding formed the hypothesis for the second experiment i.e., there may be an effect of AoA on oral reading of English words if a written word has a relatively low family resemblance (FR). The FR hypothesis generates the rather strong prediction that written AoA will have an effect on oral reading of consistently spelled words in English if the family resemblance of those words is limited. In Experiment 2, over 600 English written words with consistently pronounced orthographic rimes were classified according to whether they had many shared family members (e.g. stay = 22) or few (e.g. carve = 0) and whether they appeared in texts used by UK children (early AoA) or not (late AoA). Results showed larger effects of AoA on the oral reading of consistently spelled words with little family resemblance to other words in the lexicon. A regression analysis confirmed these effects. The findings lend support to the claim that the locus of the AoA effect on oral reading depends on the nature of the mappings between orthography and phonology.
Mary Jane White (University of Minnesota, USA);Kristen McMaster; Catherine Bohn; David Rapp - Good and poor readers: Evidence from eye-tracking
Purpose This study was conducted to identify patterns of eye-movement for good and poor developing readers at three different grade levels. These patterns would help determine if poor readers at varying ages exhibit similar difficulties when making interconnections and generating inferences during moment-by-moment reading. Method Students from 4th (n=241), 7th (n=193), and 9th (n=162) grades each read four different texts (2 narrative and 2 expository) that were controlled for levels of cohesion, goal or idea structure, word count, and grade level. At each grade level ANOVAs were conducted with reading group (poor and good) as the independent variable and the number and duration of fixations during first pass reading (forward fixations and reinspections) and second pass reading (look backs and look froms) as the dependent variables. Results First pass reading At each grade level, good and poor readers were similar in the number and duration of forward fixations. However, in all grades the poor readers engaged in longer reinspections than good readers. Additionally, 7th and 9th grade students had higher numbers of these reinspections. Second pass reading At each grade level, poor readers made more and longer look backs than good readers. In 4th grade, poor readers had a higher number of look froms than good readers, but overall there were no other differences between good and poor students in the number or duration of look froms. Conclusions Patterns of eye-movements used by poor and good readers are remarkably consistent at these grade levels. Both reader types initially progress through a text with a similar number and duration of fixations. However poor readers engage in repeated reading at the intra-sentence, inter-sentence, and discourse levels. The consistency of this pattern across grades suggests that the difficulty in making connections and inferences faced by poor readers does not change as they gain more experience with reading.
Gareth Williams (Nottingham Trent University); Clare Wood - A study investigating the role of prosodic sensitivity in reading involving priming and low band pass speech.
Purpose Recent research has shown that prosody and word stress sensitivity play a role in early reading development and may be a difficulty faced by readers with dyslexia. The purpose of the study was to investigate the role that prosody and word stress sensitivity plays in reading for skilled readers. Method Skilled reading adult participants completed a priming task in which the target was a four-syllable word presented as text with low band pass speech primes. In one condition, the congruent prime was the same word as the target text, in a second condition the congruent prime was a word with a similar word stress pattern. Incongruent low band pass speech primes were also presented as was a control condition where the prime was clear speech. Participants were asked to judge whether the primes were similar. In addition, standardised measures of reading and comprehension were taken. Results Data collection is currently ongoing, preliminary results indicate that there is a difference in the way the participants respond to the prime conditions. The same prosody and clear speech prime pairs show a priming effect, a different process appears to take place when the word stress prime is similar, participants tend to respond to the incongruent primes faster than in the other conditions. Detection analysis indicates that participants are more sensitive to the same prosody than the similar prosody primes. Conclusions The results are to be discussed with relation to the role prosody plays in reading, comprehension and lexical decisions.
Joanna P. Williams (Columbia University);J.Grant Atkins; Jill G. Ordynans - Teaching three text structures within social studies instruction to at-risk second graders
PURPOSE: We developed and evaluated a comprehension intervention that teaches sequence, compare/contrast, and cause/effect, embedded within social studies lessons. METHOD: The Text Structure program focused on several strategies (clue words, generic questions, graphic organizers) plus the close analysis of well-structured paragraphs. It also included teacher read-alouds, discussion, vocabulary and writing activities. Classrooms were randomly assigned to the program; to a Comparison program that presented the same social studies content (same materials and same number of lessons) but without the text structure instruction; and a No-instruction control. Measures: subtests of the Woodcock-Johnson Mastery Test; DIBELS Oral Fluency Test, and researcher-developed tests: Comprehension outcomes, based on summaries and answers to questions: adequacy of main idea statements and identification and accurate reporting of structural information. Content outcomes: Recall of the content (information about the Sious, Colonists, and Pioneers) and vocabulary knowledge. RESULTS: Most measures showed the predicted effects: Comprehension measures: Text Structure students performed better than the other students. Content measures: Text STructure and Comparison students performed at the same level, and both groups performed better than the control students. CONCLUSIONS: 1. Young children at risk for reading failure can benefit from explicit comprehension instruction as early as the second grade. 2. Teaching text structure does not have to deprive students of content learning; combining reading comprehension and content instruction can be effective and efficient.
Shauna Wilson (Florida State University); Christopher J. Lonigan; Nicholas P. Allan - Relations between preschool temperament, inattention, and emergent literacy: a multi-method, multi-informant study
Purpose: Although research supports links between temperament, inattention, and academic achievement, there is not yet a clear picture of the mechanism underlying these links. Deater-Deckard et al. (2009) reported that elementary-age children with better effortful control scored higher on measures of reading, but this relation was present only among children displaying low positive reactivity. The aim of this study was to examine the relation between temperament and emergent literacy in preschool-age children and extend these findings from parent report of temperament to teacher report and direct measurement of temperament. Method: The Test of Preschool Early Literacy and an effortful control task called Sun-Moon Stroop were administered to 123 children (mean age = 4.25 years). Parents and teachers completed the Children's Behavior Questionnaire, and composites for positive reactivity, negative reactivity, and effortful control were calculated. Parents, teachers, and examiners rated children's inattention using the Conners' Rating Scale. Results: Preliminary analyses do not support a simple relation between preschool children's temperament and emergent literacy. The only significant predictor of children's emergent literacy skills in this study was examiner-rated inattention. Effortful control was positively correlated with inattention. Conclusion: Whereas results of these analyses replicated the common finding of a link between children's inattention and their reading-related skills, they call into question either the robustness of the link between temperament and reading-related skills or measurement of temperament constructs. Future research should examine first the construct validity of measures of temperament in this age group and subsequently the relations between temperament, inattention, and reading-related skills.
Maximiliano Wilson (Centre de recherche de l'Institut Universitaire de gériatrie de Montréal, Canada); Fernando Cuetos; Rob Davies; Cristina Burani - Age of acquisition and frequency effects in Spanish visual word recognition
Purpose In literature, both frequency and AoA effects have been described to affect visual word recognition. Frequency effects have been reliably found in tasks that tap lexical access in transparent languages as Spanish. On the other hand, and according to the arbitrary mapping hypothesis (Ellis & Lambon Ralph, 2000; Zevin & Seidenberg, 2002), AoA effects should be found when input-output mappings are arbitrary. Since Spanish has very consistent orthography-phonology (O-P) mappings, little or no AoA effects are to be expected in word naming. Nevertheless, Cuetos & Barbón (2006) found that only AoA and not frequency affected word naming. The aim of the present study was to further investigate the effects of frequency and AoA in a transparent language as Spanish. Method Four fully factorial frequency x AoA experiments were administered to Spanish-speaking participants from the University of Oviedo, Spain. Results In Exp. 1 and 2 (immediate and speeded naming) only frequency affected naming. In Exp. 3 (lexical decision) and Exp. 4 (immediate naming with highly imageable items), both frequency and AoA affected latencies. Conclusions Results indicate that the semantic characteristics of the items (i.e., highly imageable and concrete words) induce AoA effects in word naming in Spanish. This supports the hypothesis that the possible locus for AoA effects is in the organization of semantic information (DRC model) or in the semantic - phonology mappings (arbitrary mapping hypothesis). Also, it suggests that the AoA effects found previously for Spanish may be due to item characteristics that induced semantic processing.
Heinz Wimmer (University of Salzburg, Austria); Matthias Schurz; Fabio Richlan; Jürgen Bergmann; Martin Kronbichler - On the Brain Reflection of Visual Word Processing in Dyslexic (German) Readers
Purpose: This presentation focuses on the role of the Visual-Word-Form Area (VWFA) in the left inferior occipitotemporal cortex for the reading fluency of impaired readers of German. The interesting feature of German is that it poses little difficulty for the acquisition of decoding accuracy so that a fluency impairment (in adolescent and young adults) cannot be seen as secondary to a preceding decoding problem. Methods: The results of several ERP and fMRI studies (adolescent and young adults) will be summarized. The common feature is that familiar vs. unfamiliar orthographic forms (e.g., brain vs. brane vs. brone) were presented in a phonological lexical decision task. Results: We found that dyslexic readers exhibited reduced activation of the VWFA and of the N400 amplitude. Furthermore, they failed to exhibit the familiarity related modulation of VWFA activity and of the N400 response of nonimpaired readers. The abnormalities in the VWFA stood in contrast to abnormally high activity in large premotor and motor regions of the left frontal cortex (and of subcortical and cerebellar regions). Conclusion: These dyslexic brain abnormalities will be related to our recent cognitive account of the dyslexic reading fluency deficit in terms of the dual-route model of visual word processing (Bergmann & Wimmer, 2008).
Heather Winskel (University of Western Sydney, Australia.) - Learning to spell in Thai: Some of the challenges posed by this distinctive orthography
Purpose The majority of research on literacy development has primarily been conducted on linearly arranged alphabetic orthographies, in particular Indo-European languages. Thai, a tonal language, forms an interesting comparison, as it has a distinctive alphabetic orthography with a non-linear arrangement of vowels that combine across the syllable to produce a large additional number of vowels. Thai orthography has shared properties with alphasyllabic scripts (due to common origins) and has implicit vowels for some consonants. Moreover, research has predominantly focused on reading rather than spelling development. In the current study, spelling development in Thai children is examined. A number of predictions about spelling development and types of errors are made based on previous research on other orthographies in conjunction with the characteristics of Thai orthography. Method Spelling of words was assessed in 60 Thai children ranging in age from 7 years to 9 years from Grade(s) 1, 2, and 3. Error analysis was also conducted. Results After 4 months of school, Grade 1 children achieved 32% correct, Grade 2 children 85% correct, and Grade 3 children 87% correct for word spelling. Spelling performance rapidly increased between the youngest Grade 1 children and the older children with relatively few errors made by the older children. Conclusions We found striking commonalities with other orthographies previously studied, but also orthography-specific characteristics influenced development. Homophonous consonants, consonant clusters, visually similar letters, vowel length, and other irregularities were sources of confusion for young learners. As predicted the complex vowel and tone systems also proved problematic.
Barbara Wise (University of Colorado- Boulder); Lynn Snyder; Tim Weston; Luann Sessions - Assessing reading comprehension and its related subskills with computerized informational passages
Purpose A computerized assessment system called ICARE aims to minimize two confounds of most comprehension assessments. Many comprehension tests gain reliability with many short items, but measure little beyond word reading. Others gain content validity with long passages, but are confounded by background knowledge. To meet both challenges, ICARE designed 42 2-3 paragraph "informational passages" about novel universes, and used 21 of them in this study. Method 469 2nd to 5th graders read 5-9 passages apiece before reaching ceiling; K-1st graders read fewer passages. All students also took the Woodcock Johnson Letter Word ID, Passage Comprehension, and Gates MacGinitie Comprehension tests. Children answered a main idea, major fact, near and far inference, and two vocabulary questions with multiple choice distracters. This study aims to replicate earlier results where 290 students read 2-4 passages apiece, and the test measured variance in Gates comprehension beyond word reading. Current analyses will also determine whether outstanding variance is carried by inference and challenge vocabulary questions. Results An average of 155 children read each passage from K to 5th grade, 102 for 6th, and 50 for 7th grade. This amount of data should be sufficient to confirm or disconfirm our hypotheses. We compare children of different abilities on near and far inferences and on multiple meaning vocabulary items. Conclusions This adaptive computer system provides a developmental way of assessing higher order skills in relationship to lower subskills and it should efficiently profile children's reading comprehension abilities.
Ulrika Wolff (University of Gothenburg) - Long term effects of a randomised reading intervention study on accuracy, speed and comprehension: an application of structural equation modeling
Previous research has demonstrated that intensive phonics-linked remedial intervention in a one-to-one setting is effective for word reading accuracy and reading comprehension, whereas reading fluency problems seem to remain (Torgesen, 2005). Most studies have been carried out in English speaking countries, where children learn to read and write in what is characterized as a deep orthography. However, it is often assumed that slow reading is a more pronounced problem among poor readers in more transparent orthographies (Wimmer, 1993). A Swedish theory based intervention program including phonics, fluency training and guided reading has been developed and implemented in a randomised intervention study for 9-year olds. Based on a screening battery (N=2 212) poor decoders were identified. The intervention group (n=58) received 45 minutes of one-to-one instruction per day for twelve weeks. The control group (n=55) took part in ordinary classroom activities, which in some cases included special education. Broad test batteries were performed at three occasions: pre-, post-, and follow-up test one year later. An application of SEM (Structured Equation Modeling) gives the opportunity to examine relations between background data as well as the latent variables phonological representation, phonological awareness, word decoding, text reading fluency, reading comprehension, listening comprehension, spelling, RAN and non-verbal cognitive skills between groups and over time. Preliminary analyses indicate significant effects of intervention on all measures but phonological awareness and listening comprehension. Results suggest that gains in word reading speed, text reading fluency and accuracy can be achieved by intensive phonics-linked instruction in combination with repeated reading.
Julie Wolter (Utah State University); Kenn Apel - Initial mental graphemic representation acquisition and later literacy achievement in children with language impairment: a longitudinal study
Purpose: The purpose of this study was to examine the relation between the ability to quickly acquire the orthographic skill of developing initial mental graphemic representations (MGRs) in kindergarten fourth grade literacy skills in children with typical language (TL) and children with language impairment (LI). Method: This study is a longitudinal extension of a study conducted by Wolter and Apel (in press) in which kindergarten children with LI and TL were administered a novel written pseudoword task that included measures of MGR learning (spelling and identification of target pseudowords). In the current study (four years later), we administered reading and spelling measures to 37 of the original 45 children (18 children with LI; 19 children with TL). Results: The children with LI performed significantly lower than their peers with TL on all later literacy measures. For both groups, kindergarten initial MGR acquisition ability significantly related to fourth grade real-word reading. For the children with LI, kindergarten initial MGR acquisition ability also related to fourth grade reading comprehension. Conclusions: Because MGR learning was significantly associated with word recognition skills, we propose a cascading effect in the development of MGRs; children who acquire more MGRs early on develop larger MGR banks than those who do not.
Simpson Wai Lap Wong ()Connie Ho; Bonnie Chow; Mary Waye; Dorothy Bishop - Models of speech perception and production, meta-linguistics skills and reading development in Chinese children learning to read English as a second language
Past studies have shown that phonological representations and conscious access to those representations are crucial in learning to read an alphabetic script. Our study has extended this line of research by investigating these links on second language acquisition. A sample of 294 Hong Kong Chinese children were tested on English tasks of phonological representations (speech perception and speech production), phonological awareness (rime detection, and syllable and phoneme deletion), and a English word reading measure. Phonological representations could have a) a direct impact on reading development, b) an indirect impact on reading development through the mediating effects of phonological awareness, or c) no relationship with reading skills. Results of Structural Equation Modelling suggested that the data was better fitted with the Indirect Model. Further examination of the beta weights on different pathways in the models showed that the quality of phonological representations was not sufficient to enhance reading abilities. Instead, good phonological representation was essential to the development of meta-linguistic skills which in turn fostered reading development in L2.
Anita Mei-Yin Wong (University of Hong Kong);Terry KF Au; Connie, SH Ho; Joanne C Kidd; Catherine CC Lam; Fanny WF Lam; Lesley, PW Yip - Word reading in Chinese kindergarten children with SLI
Dyslexia has been reported in Cantonese-Chinese-speaking primary school children with specific language impairment (SLI) (Wong et al., 2010). Do kindergarten children with SLI already show difficulties in word reading? Do they perform worse than their normal language (NL) age peers on the precursor cognitive skills? Sixty-nine 5-year-old kindergarten children with SLI and their NL age peers were classified as being at-risk (+), or not at-risk (-) for learning disabilities according to a parent-report checklist: 18 SLI+, 22 SLI-, 14 NL+, 18 NL-. The four groups did not differ in age, Ravens scores, parental education and reading time per week. The children completed a Chinese word-reading test, and cognitive tasks examining orthographic skills (Chinese radicals and lexical decision), phonological processing (working memory, syllable deletion, and rapid automatic naming of pictures (RAN-p)), and morphological awareness (identification and construction of word compounds). The SLI+ group, but not the NL+ group, did worse than the NL- group on word reading and identification of the semantic categories of Chinese radicals. The SLI+ group also did worse than the SLI- group on word reading, although the difference only approached significance. These results suggest the possibility of an earlier differential diagnosis of dyslexia for kindergartners with SLI. The two SLI groups scored lower than the NL groups on working memory and morphological awareness, suggesting that these two precursor reading-related skills have a strong basis on spoken language development. Connections between oral language and early word reading in Cantonese-Chinese-speaking children will be discussed.
Clare Wood (Coventry University, Psychology Department); M. Luisa Tarczynski-Bowles - An evaluation of a prosody-based spelling intervention
Purpose: To consider whether encouraging adults to reflect on the prosodic characteristics of multisyllabic words will enhance their spelling performance. Method: A sample of adult participants (N=>60, data collection is ongoing) were recruited to a study which offered them the possibility of improving their spelling performance. Initially, all participants were assessed on a standardised spelling test (WRAT) and were also asked to spell a selection of high and low frequency four syllable words. Participants were then randomly allocated to one of two conditions. In the first (intervention) condition, they were directed to write out half of the four syllable words that they had just spelled in a way that enhanced the stress pattern of that word (strong syllables were written in capitals, and unstressed vowels were underlined. The participants in the other (control) condition were asked to complete a very similar exercise, but in this case they were directed to write out the words so that consonants were in capitals and vowels were underlined. In both conditions all the trained words were written out in this way three times. There was then short break, after which the participants were retested on all the four syllable words from the pre test stage (half of which were trained, half of which were untrained). Results: Data collection is ongoing, but based on the data collected to date, there is evidence of a significant effect of the prosodic training intervention on the spelling of low frequency words. Conclusions: If the spelling intervention is found to be effective, there is scope for using prosody-based training with adults to help them to improve their spelling of longer, more complex words. There is also the possibility that the same technique may be effective in supporting the spelling development of school age children, but further data would be needed to verify this.
Christiane Wotschack (Free University of Berlin);Reinhold Kliegl; - Reading-Strategy impacts on parafoveal-on-foveal effects in sentence reading
Purpose: The goal of the study was to investigate top-down influences of reading strategy related to comprehension depth on eye movement patterns in sentence reading. Especially differences between readers in the selectivity of fixated words may impact on effects sizes of foveal and parafoveal word characteristics. Method: In an age comparative eye movement study on isolated sentence reading, two groups of young and old readers read the identical 144 sentences for comprehension. Two reading strategies were indirectly induced by altering the difficulty and frequency of comprehension questions. Reader- and word-level effects on fixation duration were evaluated in corpus analyses using linear mixed models. Results: In both age groups, variation in comprehension demands led to slower reading rates. Old readers showed a lower skipping rate and a clear selectivity effect in single fixation cases: The amount of fixated function words increased when difficult questions were asked. Interestingly, a high prevalence of fixated function words led to strong evidence for preprocessing of word n+1 as reflected in a stronger word frequency effect of the upcoming word on single fixation duration. Conclusions: Results show clear evidence for strategic, top-down influences on eye movement behavior in sentence reading that are important for theories and models of eye movement control in reading. The increased parafoveal-on-foveal frequency effects are in line with the assumption of distributed processing within the perceptual span and its modulation by the difficulty of the fixated word, as predicted by the foveal-load hypothesis.
Taeko N. Wydell (Brunel University);Akira Uno; Noriko Haruhara; Masato Kaneko; Noriko Awaya; Naoko Shinya - The impact of phonological processing skills on writing in English and Kanji among Japanese secondary school children
Purpose: The aim of this study was to ascertain the impact of phonological processing skills of Japanese secondary school children on their spelling ability in English and their writing abilities in Japanese Kanji. Method: 162 Japanese secondary school children aged 14-15 from two state-run junior high schools in Japan were assessed on their IQ measured by RCPM (Ravens Coloured Progressive Matrices), other cognitive skills including phonological processing (e.g., ability to read English nonwords), and on the spelling/writing of English and Japanese Kanji words. Using a -1.5SD cut-off on English nonword reading scores, 22 children were identified as poor phonological recoders (PPRs). Results: While no significant difference was seen on the IQ measure between "normal" controls and PPRs, the PPRs revealed significantly poorer performance not only in the spelling of English words but also the writing of Kanji words. There were some qualitative differences in the errors made by the two groups of children. Conclusion: The current study revealed that the phonological processing deficit seen in the PPRs affected not only spelling in English but also writing in Japanese Kanji. The study thus suggests the importance of phonological processing skills for normal literacy development in alphabetic languages as well as non-alphabetic languages such as Japanese. This has not been shown before by any other studies
Shelley Xiuli Tong (Dalhousie University)Catherine McBride-Chang - Toward a fine-grained analysis of Chinese morphological awareness: The roles of morphemic and submorphemic sensitivity in Chinese word reading and vocabulary development
Purpose: We examined whether both morphemic and submorphemic sensitivity uniquely explain Chinese word reading and vocabulary knowledge with development in children. Method: Morpho-lexical compounding (involving the ability to combine morphemes orally -morphemic sensitivity) and radical-semantic relatedness (focused on the semantic cuing function of radicals-submorphemic sensitivity), along with other standard measures of Chinese character recognition, vocabulary knowledge, phonological awareness and nonverbal IQ, were administered to 453 Cantonese-speaking kindergartners, second graders, and fifth graders (151 per grade). Results: The unique associations of morphemic sensitivity and submorphemic sensitivity and their interaction effects with grade were evaluated in linear regression equations. Morphemic sensitivity and submorphemic both emerged as independent predictors of Chinese character recognition (β=.10, p <.05; β=.19, p <.001), and vocabulary knowledge (β=.63, p <.001; β=.43, p <.01), even after controlling for phonological awareness, nonverbal IQ and either word reading or vocabulary knowledge. Furthermore, the contribution of submorphemic sensitivity to Chinese character recognition decreased from kindergarten (β=.53, p<.001) to second (β=.30, p<.001) and fifth grade (β=.18, p<.05) in this sample. In contrast, the effect of morphemic sensitivity on vocabulary knowledge was weaker in kindergartners (β=.09, p=.27) than in second (β=.33, p<.001) and fifth graders (β=.34, p<.05). Conclusion: Morphemic and submorphemic sensitivity were both associated with word reading and vocabulary knowledge across grade levels. There exists a developmental difference in sensitivity to morphemic and submorphemic sensitivity in learning to read and comprehend words. This underscores the uniqueness of submorpheme to Chinese, and supports the importance of morpheme across scripts.
Rachael Xuereb (Communication Therapy Division, Institute of Health Care University of Malta, Mater Dei Hospital, Tal-Qroqq, Malta.)Introduction: Malta presents an interesting situation in terms of bilingual research and literacy skills. Children attend 3 different types of school (religious, state and independent) depending largely on parental socio-economic status and geographical area. Literacy instruction varies across school type and within each type, reading instruction in the two national languages (Maltese and English) also vary. Purpose: The theoretical aim of the research is to look at the effect of bilingualism on reading ability and disability in this population in relation to the orthographic depth of the two languages. The clinical aim is to develop a sensitive comprehensive diagnostic assessment battery for reading disorders. Method: 549 children (8 - 11 years) attending different school systems were tested on reading, spelling, and phonological processing measures. Parental questionnaires investigated language and literacy practices in the home. Teachers were asked to complete a questionnaire on reading instruction and language use in the classroom. Results: Preliminary findings show that children read better in their first language. Children with a primary orientation to English and who typically attend independent schools can be administered phonological awareness tests to assess their English reading proficiency skills. On the other hand, rapid naming tests are more indicative of reading proficiency in children with Maltese as a first language who typically attend state schools. - Literacy in Maltese-English bilingual children
Introduction: Malta presents an interesting situation in terms of bilingual research and literacy skills. Children attend 3 different types of school (religious, state and independent) depending largely on parental socio-economic status and geographical area. Literacy instruction varies across school type and within each type, reading instruction in the two national languages (Maltese and English) also vary. Purpose: The theoretical aim of the research is to look at the effect of bilingualism on reading ability and disability in this population in relation to the orthographic depth of the two languages. The clinical aim is to develop a sensitive comprehensive diagnostic assessment battery for reading disorders. Method: 549 children (8 - 11 years) attending different school systems were tested on reading, spelling, and phonological processing measures. Parental questionnaires investigated language and literacy practices in the home. Teachers were asked to complete a questionnaire on reading instruction and language use in the classroom. Results: Preliminary findings show that children read better in their first language. Children with a primary orientation to English and who typically attend independent schools can be administered phonological awareness tests to assess their English reading proficiency skills. On the other hand, rapid naming tests are more indicative of reading proficiency in children with Maltese as a first language who typically attend state schools. Conclusion: These findings indicate that the literacy skills of Maltese children are not being assessed in their entirety and according to the orthographic depth of the language in question.
Chihyu Yang ()Hwawei Ko; Yujen Chang; Minglei Chen - Memory span and the contextual effects on the processing of homonymy and polysemy during Chinese sentence reading：evidences from eye movements
This study examined the effects of using context to solve the lexical ambiguity in reading sentences by recording participants' eye movement. Variables such as types of the ambiguous word (homonymy and polysemy), the location of disambiguating context (preceding and following context) and the participants' memory span were included in the study. The purpose is to investigate how memory span influences reader's reading while they read relatedness or unrelatedness of meaning of lexical ambiguity in a context supported sentence. 32 readers were grouped as high memory span group (HG) and low memory span group (LG). They were asked to read ambiguous sentences which was consistent with the subordinate meaning, meanwhile, their eye movements were recorded by SR EyeLink Ⅱ.The results of the eye movements showed that the online processing of ambiguous word will be effected by the types of lexical ambiguity and the contextual location. In the context-before condition, participants fixated longer on ambiguous word but is similarity between homonymy and polysemy. However, in the context-after condition, participants fixated longer on disambiguating context in homonymy condition. In addition, there were no differences in all eye movement indicators between HG and LG. Based on these results, researcher speculated the order of disambiguating clauses affect LG participants' lexical ambiguity resolution. They took advantage of the contextual location clues to help their reading as effectively as HG ones. Furthermore, the relatedness degree of meanings also had powerful influence on the processing of ambiguous word during sentence reading.
Minwha Yang (Kookmin University); Seungsoo Yeo - Interactive learning process of spelling knowledge in Korean Beginning Learners
Purpose This study was designed to find a statistical model to best explain whether the process of Korean children's spelling acquisition. The specific interest was whether the Korean learner's spelling development is hierarchical or interactive in learning phonetic principles and more abstract (pattern/morphemic) principles. In most orthography that was studied, a hierarchical development, in which phonetic knowledge is acquired prior to more complex orthographic rules was reported. The authors hypothesized that Korean spellers also may learn phonetic spelling rules first, and then move to learning more abstract spelling rules. Method A total of 35 kindergartners from a public kindergarten in Seoul participated in this study. The children were asked to spell the same ten words in four different time periods throughout a year. The spelling words were chosen based on how clearly the word represents a phonetic feature, pattern feature, or morphemic feature. The test scores for each child represented how many phonetic features, pattern features, and morphemic features he/she spelled correctly. Latent curve modeling(Bollen & Curran, 2006) was conducted to test the authors' hypothesis. Results The results showed that the hypothesis needs to be rejected and that Korean spelling acquisition is not hierarchical, in that the sequential learning process from phonetic spelling to abstract spelling is not observed. Rather, the interactive model explains Korean beginning spellers' learning process better. Conclusion The interactive model revealed that young Korean children learn to spell phonetic features and abstract features at about the same time. Korean children's knowledge of abstract features helps them to spell phonetic features more accurately, just as their knowledge of phonetic features helps them learn more about abstract spelling features.
Li Yin (Tsinghua University)Rebecca Treiman - Young Chinese children differentiate between drawing and writing
Purpose: Children under the age of 3 ½ or so are commonly thought to produce the same types of scribbles for writing and drawing. We tested this idea by examining whether Chinese children aged 2 through 6 make distinctions between drawing and writing in their productions and whether these distinctions make sense to adults. Method: 109 native Chinese children aged 2 through 6 in Beijing, China, participated in the study. They were asked to write and draw four targets. In Experiment 1, 16 native Chinese college students were shown, individually, several writings and drawings from a child and asked to determine whether a new product from the same child was writing or drawing. In Experiment 2, another 16 college students were shown two sets of productions from each child and asked to assign the productions to the categories of writing and drawing. Results: Results show that Chinese children as young as 2 produce distinctions between writing and drawing in ways that adults can appreciate. The adult judges performed significantly above the level expected by chance in distinguishing the written and drawn productions of children from all age groups in both experiments. Importantly, adults performed above the level of chance even with the productions of children aged 2 to 2 ½ and even with the productions of children who could not read any characters. Conclusions: Distinguishing writing from drawing in production, a first step in literacy learning, appears to take place earlier than previously believed.
Astri Yulia (); R. Malatesha Joshi; Zainab Allaith - Does the spoken language help or hinder the spelling in the second language? A case of Indonesian and Acehnese language
Purpose The purpose of this study was to examine the influence of the spoken form of a second-language on children's spelling of Acehnese words. Acehnese is a regional language spoken in Aceh, Indonesia and taught as a second-language at schools in Aceh. Method The participants were two groups of Indonesian children from grade 4: Group 1 was exposed to the spoken Acehnese (parents spoke Acehnese at home) and Group 2 was not exposed to spoken Acehnese. Their spellings of 50 Acehnese words were compared. The selected words consisted of five target phonemes which exist in both Acehnese and Indonesian but represented by different graphemes. The comparison was conducted on the spelling accuracy and the nature of the spelling errors. The spelling of graphemes was scored using a rubric with the following categories: spelling accurate Acehnese graphemes, replacing the target Acehnese graphemes with Indonesian graphemes, spelling the target phonemes with other plausible phonological representation, and spelling the target phonemes with other phonologically non-related graphemes. Results The exposure to spoken Acehnese was found to interfere with the children's spellings in Acehnese words. The children who were not exposed to spoken Acehnese made fewer spelling errors than those who spoke Acehnese at home. Further, the spelling errors committed by the children in both groups indicated the influence of Indonesian graphemes on spelling in Acehnese. Conclusions The phonology influences the children's spelling performance in the second-language. Additionally, children tend to use first-language graphemes to represent similar phonemes in a second-language.
Maaike Zeguers (University of Amsterdam)Patrick Snellings; Hilde M. Huizenga; Maurits W. van der Molen - Does visual word recognition in developing readers involve automatic phonological code activation?
Purpose - A key question in the literature on visual wordrecognition is whether phonological activation is late, strategic and rule based or rather occurs quick, automatic and general. Masked priming studies yield important insights into the timecourse of phonological and orthographic code activation during the initial, prelexical phases of visual word recognition (Ferrand and Grainger, 1993; Grainger and Holcomb, 2009). Surprisingly, little research has focused on children. Thus far, results on whether developing readers can activate phonology fast and automatically are inconsistent (Booth, Perfetti & MacWhinney, 1999; Davis, Castles & Iakovidis, 1998). Therefore, the current study will adopt a comprehensive masked priming paradigm to chart the time course of phonological and orthographic activation during visual word recognition at different phases of reading development. Method - Forward masked priming was used in a lexical decision task. Targetwords were preceded by one of three nonword primes (phonological, orthographic or unrelated control prime) and primes were presented for either 33 ms, 50 ms, 67 ms or 83 ms. 338 children from 2nd (n=109), 4th (n=105) and 6th (n=124) grade participated. Results - Preliminary analyses are suggestive of both phonological and orthographic priming effects. Results suggest that priming effects occur in all three grade levels, but that the size of the effects depends on the prime duration. Subsequent analyses will focus on developmental changes in phonological and orthographic priming and on determinants of individual differences in priming effects. Conclusions - Results will be interpreted in light of theories on automatic activation of phonological processing in developing readers. Results will enhance our understanding of lexical access processes in children and will contribute to refinement of current theories on word recognition.
Pascal Zesiger (FPSE - University of Geneva); Jacqueline Lurin; Anne Soussi - Prediction of literacy skills of at-risk adolescents in a French-speaking environment
Purpose Recent large scale studies have shown that a number of children reaching the end of primary school still struggle with written text comprehension and production, in particular in languages possessing an opaque orthographic system. This study aims at gaining a better understanding of the variables explaining these difficulties in 7th grades, French speaking, monolingual and bilingual students with low academic performance. Method The impact of several attributes (linguistic, cognitive, and meta-cognitive skills) on the reading and writing skills is studied in two cohorts of adolescents who have been identified as low academic performers by the end of primary school. Each cohort contains 30 students; one is French monolingual and the other one is bilingual. The students have been tested in the middle of their 7th year of schooling. Results The variance analyses show that monolingual students outperform bilingual students in vocabulary (mostly production), in written word identification, in spelling and in written comprehension. Furthermore, regression analyses indicate that performance in literacy is predicted by skills in vocabulary, grammar and by decoding skills (reading fluency and accuracy) and non-verbal intelligence. However, the predicting variables are partly different for monolingual and for bilingual students. Conclusions These results suggest that literacy difficulties of at-risk adolescents are mostly attributable to their linguistic skills, and that bilingual students are particularly vulnerable for arrears in this domain.
Jason D. Zevin (Sackler Institute, Weill Medical College of Cornell University) - Ad hoc functional units in spelling to sound: Grain size issues in cross-linguistic modeling of reading
Yuping Zhang (); Hua Shu; Hong Li; Jing'er Pan; Catherine McBride-Chang - Socioeconomic status modulates phonological and reading development
Purpose: Two main questions would be addressed: 1) how socioeconomic status (SES) modulates the growth of phonological processing ability in early childhood; 2) whether the SES and reading relation is mediated by early phonological ability and the nature of Chinese character reading is also discussed. Method: From the Chinese longitudinal study of dyslexia，we now report the data including 262 children (119 girls, 143 boys) since 3-year-old to discuss SES, early phonology and reading relation. Results: SES contributes to early phonological acquisition with a clear increasing tendency: when children were only 3-year-old, their phonological achievement was not significantly influenced by SES (△r2 = .01, p > .05). While as children growing up, the importance of SES index increased remarkably and the explainable unique variance reached 6.1% and 9.3% respectively for age 4 and age 5. The relation between SES and Chinese character reading is totally mediated by phonological processing ability, and the parameter estimate for the SES effect reduced to 1.67 (from 3.42). The bootstrapping estimated indirect effect is 1.76 with a standard error of 0.56. The relation between SES and reading comprehension is partially mediated by phonological processing ability, and the parameter estimate for the SES effect reduced to 0.55 (from 0.83). The bootstrapping estimated indirect effect is 0.28 with a standard error of 0.08. Conclusions: SES contributes to individual phonology development gradually. The relation between SES and reading was mediated by phonological processing ability.
Juan Zhang (Chinese University of Hong Kong); Catherine McBride-Chang - Predictors of Reading Comprehension Difficulties in Chinese Children
Purpose: Early predictors of reading comprehension difficulties in Chinese children beyond word reading were investigated. Method: Based on a test of sentence-level reading comprehension at age 8, 30 children in the bottom 18% of a sample of 163 Hong Kong Chinese children from a longitudinal study were chosen as the poor comprehension group, and they were matched on concurrent Chinese word reading, maternal education and age to another 30 good comprehenders, whose reading comprehension scores ranged from 30 -99.8%. These two groups were compared on a battery of skills across ages 5-8, i.e., morphological awareness, phonological awareness, vocabulary, rapid automatized naming, sentence imitation and , speed of processing (assessed at ages 5-7), listening span (tested at ages 5 and 6) and receptive grammar (measured at age 5). Results: Good comprehenders performed better than poor comprehenders on Chinese word reading and receptive grammar at age 5, and listening span at ages 5-6. The groups also differed significantly on vocabulary, phonological awareness, morphological awareness and speed of processing across ages 5-7, but only on vocabulary knowledge at age 8. Conclusions: Early metalinguistic skills, grammatical knowledge, short-term memory and processing speed distinguish those with and without early reading comprehension difficulties in Chinese children. In addition, it is possible that very early delayed word reading might be an additional marker of subsequent poor reading comprehension, even when this difficulty subsequently is ameliorated. Key Words: reading comprehension; cross-linguistic, comprehension difficulty, longitudinal
Jing Zhao (Sun Yat-sen University)Blanca Quiroz; L. Quentin Dixon; R. Malatesha Joshi - The effect of bilingualism on English spelling: A meta-analytic review
Purpose - To investigate the effects of English language learner (ELL) status on English spelling, we examined results on two spelling outcomes: real word spelling (lexical) and non-word spelling (sub-lexical). We also investigated systematic sources of variability in effect sizes, including reading level and first language (L1) background. Method -Studies published in peer-reviewed journals between January 1990 and November 2009 were retrieved, resulting in 841 abstracts from which 16 quantitative studies using English L1 (EL1) control groups were examined. The study-level variables and characteristics of 25 effect sizes across 15 studies were analyzed using the Q-method. Results - A positive bilingual effect (ELLs vs. EL1s) was found for lexical spelling (g=1.09) and a negative bilingual effect was found for sub-lexical spelling (g=-0.25). Moderator analysis using the Qb and Qw method (analogous to ANOVA) showed that ELLs classified as at-risk readers performed statistically significantly better than at-risk EL1s on lexical spelling; however, there was a smaller effect favoring at-risk ELLs over at-risk EL1s (g=.71) compared to the not-at-risk groups (g=1.30). The same procedure revealed statistically significant differences in the average effect sizes favoring L1-alphabetic ELLs over L1-non-alphabetic ELLs on lexical spelling. Conclusions - By virtue of their smaller oral vocabularies, perhaps ELLs are learning to spell from print, leading to higher accuracy for real words, but little accuracy with nonwords. The disadvantage associated with at-risk ELLs suggests their lexical spelling success is related to their reading success. Instruction in phoneme-grapheme correspondence may bring benefits for ELLs, particularly those whose L1 is distant from English.
Yan Ling Zhou (The Hong Kong Institute of Education);Cathy Fong; Catherine McBride-Chang - A comparison of phonological awareness, lexical compounding, and homophone training for Chinese word reading in Hong Kong kindergartners
Purpose-We investigated the effects of three training programmes on Chinese reading development among Hong Kong kindergartners (age 6). Methods-The "phonological" group focused on phonological structures of Chinese words, the "compounding" group received training in analyzing the lexical compounding structures of words, and the "homophone" group was trained on homophone recognition across words. All groups received two 20-minute training sessions by classroom teachers with materials from experimenters each week for 6 weeks. Training focused on oral language skills only for the first three weeks and subsequently on combined oral and print skills. Compared with the control group, the phonological and homophone groups showed improved phonological skills. Results- The compounding awareness and homophone groups both improved in both morphological awareness and Chinese character reading skills. The homophone group also improved in vocabulary knowledge Conclusions-Results underscore the importance of morphological awareness for early reading skills in Chinese.
Johannes Ziegler (CNRS and Aix-Marseille University)Jonathan Grainger - What letters, numbers and symbols tell us about dyslexia
We used a Reicher-Wheeler 2AFC paradigm with strings of letters, digits and symbols to adjudicate between competing theories of dyslexia. A comparison of performance in dyslexics and normally developing readers revealed global deficits for letters and numbers but not for symbols, which rules out low-level visual and/or attentional explanations of dyslexia. Moreover, the serial position functions of dyslexics for letters and numbers were identical to those of controls, which rules out theories that attribute dyslexia to increased crowding or lateral masking. The fact that numbers showed the same deficit as letters rules out backwards causality whereby lack of reading would cause poor letter processing rather than the other way around. Our results suggest that dyslexia is the consequence of deficits in basic mechanisms involved in processing alphanumeric stimuli, such as the mapping of position-coded elements onto componential phonology.
Naomi Zigmond (University of Pittsburgh); Amanda Kloo - What state tests reveal about achievement and growth of students with disabilities
Purpose - This study explores one state's efforts to understand achievement levels and growth patterns in students with disabilities to identify a target population of students with persistent academic difficulties considered eligible for an alternate state test based on modified academic achievement standards, or "the 2% test". Method - Hierarchical Linear Modeling and logistic regression were used to analyze longitudinal state accountability assessment data for students with disabilities to determine identifiable trends in achievement and growth as well as variables predictive of achievement such as demographics, disability category and instructional setting. Results - Preliminary analyses suggest that performance is static across grade levels. Students who fail to reach proficiency thresholds in third grade continue to perform well below grade level standards in subsequent grade levels no matter their demographics, disability type or instructional setting. Conclusions - Classifying students annually into proficiency categories based on statewide assessment performance tells us very little about actual student achievement and growth. Because students who persistently fail to make enough annual progress to achieve proficiency status are now considered eligible for an alternate assessment based on modified achievement standards, we need to know considerably more about those students before modifying achievement expectations. Even more importantly, we must consider what these results tell us about the effectiveness of inclusive educational practices intended to increase students' access and success in the general education curriculum.