Zvia Breznitz

Asynchrony of visual-orthographic and auditory-phonological word recognition processes: An underlying factor in dyslexia

This study investigated whether asynchrony ofspeed of processing (SOP) betweenvisual-orthographic and auditory-phonologicalmodalities can account for word recognitiondeficits among dyslexic readers. SOP amongelementary school dyslexic readers was comparedto that of chronologically age-matched normalreaders. SOP was assessed using nonlinguisticand linguistic auditory and visual low-leveltasks and higher-level orthographic andphonological tasks. Behavioral andelectrophysiological (ERP) measures of SOP wereobtained. Data indicated that dyslexic readerswere significantly slower than control readersin most of the experimental tasks. Moreover,dyslexics revealed a systematic SOP gap betweenthe auditory-phonological and thevisual-orthographic modalities. This gap wasfound in both P200 and P300 latencies, andexplained most of the variance in wordrecognition. A theory is proposed suggestingthat asynchrony between the processing rates ofthe visual and the auditory modalities may bean underlying cause of dyslexia.

Assessment of Working Memory Components at 6 Years of Age as Predictors of Reading Achievements a Year Later.

The ability of working memory skills (measured by tasks assessing all four working memory components), IQ, language, phonological awareness, literacy, rapid naming, and speed of processing at 6years of age, before reading was taught, to predict reading abilities (decoding, reading comprehension, and reading time) a year later was examined in 97 children. Among all working memory components, phonological complex memory contributed most to predicting all three reading abilities. A capacity measure of phonological complex memory, based on passing a minimum threshold in those tasks, contributed to the explained variance of decoding and reading comprehension. Findings suggest that a minimal ability of phonological complex memory is necessary for children to attain a normal reading level. Adding assessment of phonological complex memory, before formal teaching of reading begins, to more common measures might better estimate childrens likelihood of future academic success.

Phonological and orthographic processing of Hebrew words: electrophysiological aspects.

Brain activity among 15 male, college-level, normal readers in Israel was examined during phonological and orthographic word-recognition tasks. Both electrophysiological (event-related potentials, or ERPs) and behavioral measures were obtained. Data indicated that (a) behavioral accuracy was almost perfect for all the experimental tasks, and (b) although P200 and N400 ERP components were elicited in the experimental tasks, the latencies of those components were significantly longer and their amplitudes significantly higher in the phonological task. Variations in vowel information had no effect on word recognition in either type of task. The results suggest that among skilled readers of Hebrew, phonological processing during word recognition may be more effortful and may demand greater cognitive resources than orthographic processing. Furthermore, the additional phonological information represented in vowels appears to contribute little to word recognition in this population. These findings support earlier research on skilled reading in Hebrew as well as current theoretical models of reading.

The Underlying Factors of Word Reading Rate

The discovery of the “acceleration phenomenon” and the contribution of reading rate as an independent variable to the quality of the reading process brought about a systematic research project that examined the underlying factors of word reading rate. Word reading requires processing information in the visual and auditory modalities and relies on word recognition skills such as phonological and orthographic decoding. Therefore, it is believed that speed of processing of some subset of the above sub-processes affects word-reading rate. By using experimental manipulations relying on both electrophysiological (ERPs) and behavioral measures, the sub-processes that are activated in word reading can be distinguished and shown to differentiate between the early perceptual, the central cognitive, and the motoric stages of activation. Results indicate that word reading rate is related to basic speed of processing in the modalities activated during the reading process. However, the processing level within each modality that contributes to self-paced and fast word reading rate differs according to age and the efficiency of reading skills.

The Effect of School Entrance Age on Academic Achievement and Social-Emotional Adjustment of Children: Follow-up Study of Fourth Graders.

One hundred and thirty-seven fourth-grade pupils participated in a follow-up study of the effect of school-entrance age on academic achievement and social-emotional functioning. The sample included 73 oldest pupils, born in the months of January to March, and 64 youngest pupils, born in October to December of the same calendar year. Results indicated that differences between the two groups in academic and social-emotional measures, found when the subjects were in first grade, persisted in part to the fourth grade. The youngest subjects continued to score lowest in mathematics and in oral and silent reading comprehension; they were slower readers; and they had higher trait anxiety scores. In addition, the differences between the groups in oral and silent reading comprehension performance were found to increase over the years.

Developmental associations between verbal and visual short-term memory and the acquisition of decoding skill

This study longitudinally investigated the relationships between verbaland visual short-term memory (STM) and the acquisition of decoding from thepre-reading through the early acquisition stages in 63 Hebrew-speakingchildren Test waves occurred in kindergarten, first grade and second grade.IQ, visual and verbal STM and decoding ability were assessed. The dataindicated that while both verbal and visual STM in kindergarten were significantlycorrelated to later decoding skill, pre-reading visual STM was a stronger predictor.The results further showed that pre-reading performance on the WISC-R BlockDesign test predicted later decoding ability, while performance on theWISC-R Vocabulary test did not. Lastly decoding skill in grade 1 was foundto predict only visual in grade 2. These results indicate that visualparameters may make a crucial contribution to the acquisition of decodingskills. The size of pre-reading visual STM capacity appears to play a rolein this process. The relationship between visual STM and decoding may bebidirectional, as learning to decode appears to develop visual STM. It issuggested that either language-related or task-related factors may accountfor these counter-to-mainstream results.

Verbal Indicators of Depression

Speech content, voice quality, and temporal pacing of speech were evaluated for 11 well and 11 depressed women. Sadness was the dominant mood of the depressed womens speech, whereas happiness and mood neutrality characterized the well womens speech. The well womens tone of voice changed with the content of their speech. The well women spoke with a wide range of fundamental frequency, and the average fundamental frequency of their voices changed according to speech content. In contrast, the depressed women spoke with a narrow range of fundamental frequency, and the average fundamental frequency of their speech was unaffected by speech content. The depressed women spoke with longer pauses than the well women did. The results of this study contribute to the understanding of measures for assessing depression.

An error-detection mechanism in reading among dyslexic and regular readers--an ERP study.

OBJECTIVE As a continuation of a previous study which demonstrated the existence of the error-monitoring mechanism among regular readers, the current study attempts to characterize the error detection negativity (ERN/Ne) and the correct-related negativity (CRN) among dyslexics. More specifically, given the lexical and semantic processes involved in reading, the study attempts to reveal the reasons for differences between regular and dyslexic readers with respect to the ERN/Ne and N400 amplitudes and latencies. METHODS A visual lexical decision paradigm and event-related potential (ERP) analysis were used. RESULTS Dyslexics exhibited lower ERN/Ne amplitudes and later latencies in error responses as compared to correct responses. A smaller difference between the ERN/Ne and CRN amplitudes among dyslexics was observed. In addition, a later N400 component was elicited for error responses in both groups of readers. CONCLUSION A different error-detection activity level among dyslexic readers was found in reading. SIGNIFICANCE These results emphasize the differences in brain activity among regular and dyslexic readers. Our results point at a possible model of brain processing of incorrect reading among dyslexics. This model can be used to explain the reasons for repeated reading errors among adult compensated dyslexics despite extensive exposure to print over the years.

Brain Activity During Performance of Naming Tasks: Comparison Between Dyslexic and Regular Readers

This research was aimed at contributing to the current understanding of the underlying factors of naming speed and the causes of naming speed deficits. Forty regular readers and 40 dyslexic university students participated in the study. Electrophysiological (Event-Related Potentials [ERPs]) and behavioral measures were employed. Behavioral baseline tasks assessed general ability, reading skills, reading-related cognitive abilities, and standard Rapid Automatized Naming (RAN) and Rapid Alternating Stimulus tests. ERP tasks included letter and object naming tasks adapted for electrophysiological research presentation. The dyslexics were significantly slower and less accurate than the controls on most of the baseline measures. On all the naming tasks, the peak ERP latencies were elicited later, reaction times were longer, and the P300 latency width was wider among the dyslexics as compared to the regular readers. On the choice reaction time naming tasks, accuracy for both groups was almost perfect. When naming presentation time was controlled by the experiment, the dyslexics were significantly less accurate than the controls. Our data indicated that effective naming speed was related to an earlier P200 latency peak among regular readers and to an earlier P300 latency peak and narrower area component activation in the dyslexic group. The results from this study suggest that effective RAN speed among regular readers might be a result of efficient processing of RAN information at the input stage of stimulis perception and evaluation as well as of updating and processing the information in short-term memory among dyslexics.

Reading processes in L1 and L2 among dyslexic as compared to regular bilingual readers: behavioral and electrophysiological evidence

Abstract The present study examined differences between ‘regular’ and dyslexic adult bilingual readers when processing reading and reading related skills in their first (L1 Hebrew) and second (L2 English) languages. Brain activity during reading Hebrew and English words and pseudowords was studied using behavioral and ERP measures. Results indicated discrepancies in the processing profiles of dyslexic and regular bilinguals in both first and second languages. In general, dyslexic readers were significantly slower and less accurate compared to regular readers during processing of information in both L1 and L2. Furthermore, the latencies of the evoked potentials were later among dyslexic readers at all levels of the cognitive processing sequence in both languages, but were more pronounced in English (L2). In contrast, regular readers displayed either similar or significantly shorter ERP latencies in English compared to Hebrew on most of the experimental tasks. The data from the present study supports both the ‘Central Deficit’ and ‘Script Dependent’ Hypotheses by demonstrating universal deficits in L1 and L2 among dyslexic readers along with differential manifestations of these deficits as a function of specific orthographic features. The present results are also in line with the ‘Dyslexic Automatization Deficit Hypothesis’ which purports automatization deficits as a more general and more pervasive cause underlying dyslexic performance. Unlike dyslexic readers, regular bilingual readers not only have the capacity to reach automaticity in their second language, but can also exhibit more effective processing of reading material in English at different levels of the cognitive sequence, despite its irregular nature. Mastery of reading proficiency in English as L2 need not be lower than mastery of Hebrew as L1. Moreover, it may even be higher due to the unique features of the languages and orthographies involved.