Search

Research into second language (L2) reading is an exponentially growing field. Yet, it still has a relatively short supply of comparable, ecologically valid data from readers representing a variety of first languages (L1). This article addresses this need by presenting a new data resource called "MECO L2" (Multilingual Eye Movements Corpus), a rich behavioral eye-tracking record of text reading in English as an L2 among 543 university student speakers of 12 different L1s. MECO L2 includes a test battery of component skills of reading and allows for a comparison of the participants' reading performance in their L1 and L2. This data resource enables innovative large-scale cross-sample analyses of predictors of L2 reading fluency and comprehension. We first introduce the design and structure of the MECO L2 resource, along with reliability estimates and basic descriptive analyses. Then, we illustrate the utility of MECO L2 by quantifying contributions of four sources to variability in L2 reading proficiency proposed in prior literature: reading fluency and comprehension in L1, proficiency in L2 component skills of reading, extralinguistic factors, and the L1 of the readers. Major findings included (a) a fundamental contrast between the determinants of L2 reading fluency versus comprehension accuracy, and (b) high within-participant consistency in the real-time strategy of reading in L1 and L2. We conclude by reviewing the implications of these findings to theories of L2 acquisition and outline further directions in which the new data resource may support L2 reading research.

We present a framework for conceptualizing and analyzing responses to serial naming (rapid automatized naming) tasks, in which participants sequentially name stimuli presented simultaneously in an array. We aim to understand how these tasks are processed and why they are associated with reading skills. We analyzed responses by 298 Greek children in Grades 1, 3, and 5 to serial and discrete naming of digits, dice, objects, number words, and words. We measured the durations of silent and speech intervals in each task and grade and tested predictions about their relations based on a hypothesis of two overlapping processing stages. We found that articulation times were longer in the serial tasks, modulated by task demands. Total times were faster for serial than for discrete tasks, and their differences (serial advantage) were increasingly associated with the duration of speech intervals, consistent with efficient scheduling. Serial naming rate approached or exceeded the limits imposed by processing time, consistent with increasing processing overlap. These patterns were primarily observed for digits, number words, and—to some extent—dice, after Grade 1. Object naming seemed to pose different cognitive demands, stably across grades. Word reading exhibited the greatest differences between grades, consistent with rapid development of automaticity. We interpret the findings within a cascaded processing framework, in which performance is determined by the efficiency of cognitive scheduling of successive operations, constrained by susceptibility to interference from adjacent items. We propose that reading fluency is predicted by serial naming because it is also governed by the same scheduling constraints. Educational Impact and Implications Statement: Serial naming tasks, in which children are asked to name aloud an array of digits, pictures, or other familiar items, are widely used in educational assessment because they are associated with academic skills and, in particular, predict the development of reading fluency. In this study, we analyze children's vocal responses to serial naming tasks breaking them down into the duration of the spoken words and of any silent intervals between them. We interpret these durations under a theoretical approach that focuses on the scheduling and control of cognitive processes, that is, how the mind organizes and monitors its functions while going through the array naming the items. We claim that this approach can explain the established associations between naming and reading and can also shed light on the development of reading fluency and potential obstacles to that. Seeing fluency as not only about the automaticity of reading individual words, but also as a problem of cognitive scheduling, can inform our approaches to instruction and assessment by drawing attention to issues of interference and control and directing future research into potential interventions in those domains. [ABSTRACT FROM AUTHOR]

Purpose: This longitudinal study examined the contribution of preschool morphological awareness to word reading skills and reading comprehension, as well as to the developmental change of reading ability beyond other well-established oral language and cognitive predictors. A distinction was made between the domains of inflectional and derivational morphology. Method: Two hundred and fifty-nine Norwegian-speaking children (46% female) with a mean age of 5.5 years were assessed in preschool on language measures and again in Grades 1 and 3 on measures of word reading accuracy and fluency and in Grades 3 and 4 on reading comprehension. We fit latent change score models with preschool predictors using parceling to control for measurement error. Results: We found a unique contribution of preschool morphological awareness to reading comprehension in Grade 3, but no unique contribution to Grade 1 decoding. Neither awareness of inflections nor awareness of derivations predicted developmental change in word reading fluency between Grades 1 and 3 or change in reading comprehension between Grades 3 and 4 beyond the effect of control variables. Conclusion: Our findings confirm the relevance of morphological awareness only for early attainment in reading comprehension and highlight the importance of accounting for measurement error in studying associations among variables aiming to discover specific contributions. [ABSTRACT FROM AUTHOR]

Purpose: Previous studies examining the inter-relations between serial and discrete naming with reading have found that the ability to efficiently process multiple items presented in a sequence (indexed by serial naming) is a unique predictor of word- and text-reading fluency. However, conclusions have been tempered by the concurrent nature of the available data and the uniformly low demands of the materials (words and texts). Here we go beyond previous studies by using more varied materials to examine the relations of serial and discrete naming with the discrete reading of words and the serial reading of word lists and connected text over time. Method: Two hundred and eight English-speaking Canadian children (51% female, Mage = 7.2 years) were followed from Grade 2 to Grade 5 and were assessed on serial and discrete digit naming and serial and discrete word reading at both measurement points. Results: Strong associations between discrete naming and discrete reading already from Grade 2 indicated that short and high-frequency words were processed in parallel early in development. By Grade 5, when word recognition was presumably automatized, serial naming accounted for unique variance in serial reading of word lists and connected texts after controlling for discrete word reading. More importantly, Latent Change Score modeling indicated that serial naming was the main predictor of growth in serial reading from Grade 2 to Grade 5. Conclusion: These findings suggest that, beyond individual word recognition, reading fluency development also requires efficient processing of multiple items presented in serial format (termed "cascaded processing"). [ABSTRACT FROM AUTHOR]

The serial advantage, defined as the gain in naming rate in the serial over the discrete task of the same content, was examined between grades and types of content in English and Greek. 720 English- and Greek-speaking children from Grades 1, 3, and 5 were tested in rapid naming and reading tasks of different content, including digits, objects, dice, number words, and words. Each type of content was presented in two presentation formats: multiple stimulus displays (i.e., serial naming) and isolated stimulus displays (i.e., discrete naming). Serial tasks yielded faster naming rates--irrespective of task content--in both languages. However, content-specific characteristics influenced the trajectory of the serial advantage between grades. Improvement in the serial advantage between grades was found to be greatest for word reading, which started off similar to object naming in Grade 1, but ended up similar to digit or dice naming by Grade 5. In addition, growth in serial advantage was found to be associated with growth in discrete naming rate only in grade level analysis. For individuals, greater serial advantage was found to rely on processing skills specific to serial naming rather than on differences in the rate of naming isolated items. Our findings suggest that group level findings may not generalize to individuals, and although practice and familiarity with the content on the naming/reading task may impact the development of serial advantage, isolated item identification processes contribute little to individual differences in the gain in serial naming rates.

This study aimed to examine (a) the developing interrelations between the efficiency of reading individually presented words (i.e., isolated word recognition speed) and the efficiency of reading multiword sequences (i.e., word list and text reading fluency); (b) whether serial digit naming, indexing the ability to process multi-item sequences, accounts for variance in word list and text reading fluency beyond isolated word recognition speed; and (c) if these patterns of relations/effects differ between two alphabetic languages varying in orthographic consistency (English and Greek). In total, 710 Greek- and English-speaking children from Grades 1, 3, and 5 completed a serial digit naming task and a set of reading tasks, including unconnected words presented individually, unconnected words presented in lists, and sentences forming a meaningful passage. Our results showed that the relation between isolated word recognition speed and both word list and text reading fluency gradually decreased across grades, irrespective of contextual processing requirements. Moreover, serial digit naming uniquely predicted both word-list and text reading fluency in Grades 3 and 5, beyond isolated word recognition speed. The same pattern of results was observed across languages. These findings challenge the notion that individual word recognition and reading fluency differ only in text-level processing requirements. Instead, an additional component of processing multi-item sequences appears to emerge by Grade 3, after a basic level of both accuracy and speed in word recognition has been achieved, offering a potential mechanism underlying the transition from dealing with words one at a time to efficient processing of word sequences.

We implemented the connectionist model of social-pragmaticword learning (Caza & Knott, 2012) to test the hypothesis thatreduced joint attention between infant and mother wouldincrease the difference in acquisition between nouns andverbs as observed in Autistic Spectrum Disorder (ASD). Theratio of objects to actions in the observed event stream wasmanipulated to create an original noun-verb asymmetry. Tensimulations were run for each of the combinations of threeconditions of communicative reliability and two conditions ofunfiltered random associative learning, which is regarded bysome researchers as the primary mechanism of languagelearning in ASD. The simulations indicated that the reductionin the reliability of communicative actions does not lead toincreased noun-verb asymmetry within the originally plannedtraining epochs. A trend in the predicted direction appearedtoward the end of training, suggesting that further simulationsmay help resolve the issue within the current architecture.

Category learning is thought to be mediated—in at least somecategory structures—by hypothesis-testing processes. Verballabels for the stimuli and stimulus individuation have beenshown to facilitate the formation, testing, and application ofrules of category membership (Fotiadis & Protopapas, 2014).We sought to replicate the phenomenon of facilitation due toverbal names for the stimuli by training participants for twoconsecutive days to either learn new names for abstractshapes, or learn shape-ideogram pairings; a third group wasunexposed to the shapes. After training, participants weregiven a Type II categorization task—thought to be mediatedby verbal processes of rule discovery—utilizing the trainedshapes. We hypothesized that verbal labels for the shapes andshape individuation would provide facilitative effects inlearning to categorize. Results revealed no effect of trainingon categorization performance. This study suggests thatcaution should be taken when generalizing findings acrossperceptual modalities or different experimental paradigms.

We examined differences between the processing of inflectional versus derivational morphology, using Greek nounsand verbs with a primed lexical decision task. Previous work suggested that both noun and verb targets were significantlyprimed by the same grammatical class. However, when preceded by different grammatical class, verb but not noun targetsshowed priming. We attributed the asymmetrical priming to the materials used: noun stimuli were derived by their verbcounterparts, suggesting an important inherent asymmetry between nouns and verbs. To further investigate this suggestion,we used materials with the opposite asymmetry (verbs derived by nouns) expecting an asymmetry in the opposite direction toemerge for derivationally related words. A clear explanation of the asymmetry would allow us conclusions about the (debated)existence of differences in representation and processing between inflectional and derivational morphological relations and thusprovide evidence for or against a fully decompositional view of processing morphologically complex words.

Word list reading fluency is theoretically expected to depend on single word reading speed. Yet the correlation between the two diminishes with increasing fluency, while fluency remains strongly correlated to serial digit naming. We hypothesized that multi-element sequence processing is an important component of fluency. We used confirmatory factor analyses with serial and discrete naming tasks with matched items, including digits, dice, objects, number words, and words, performed by about 100 Greek children in each of Grades 1, 3, and 5. Separable serial and discrete factors emerged across grades, consistent with distinct skill dimensions. Loadings were greater for serial than discrete, suggesting that discrete processing does not fully determine serial processing. Average serial performance differed more than discrete between grades, consistent with improvement beyond single-item speed. Serial word reading aligned increasingly with the serial factor at higher grades. Thus, word reading fluency is gradually dominated by skill in simultaneously processing multiple successive items through different stages (termed "cascading"), beyond automatization of individual words.

We examined the manifestation of dyslexia in a cross-linguistic study contrasting English and Greek children with dyslexia compared to chronological age and reading-level control groups on reading accuracy and fluency, phonological awareness, short-term memory, rapid naming, orthographic choice, and spelling. Materials were carefully matched across languages in item properties and structure. English children with dyslexia were more impaired on reading accuracy and phoneme deletion but not on reading fluency, memory, naming, or orthographic choice. No differences in impairment were observed between words and pseudowords across languages. Orthographic tests targeted specific morphemes to examine the accessibility of functionally distinct word parts across languages. There were no differences in prefix and stem orthographic choice, but English children were less successful in spelling inflectional suffixes despite greater morphological richness in Greek, highlighting the need for additional considerations beyond grain size in cross-linguistic work.

After reviewing past efforts related to computer-based reading disability (RD) assessment, we present a fully automated screening battery that evaluates critical skills relevant for RD diagnosis designed for unsupervised application in the Greek educational system. Psychometric validation in 301 children, 8-10 years old (grades 3 and 4; including 288 from the general school population and 13 from a clinical sample), indicated that computer-based screening can detect children likely to be diagnosed with RD (with 80-86% correct classification), using a linear discriminant function derived from measures taken without supervision within a 30-minute "computer game" interaction. We conclude that automated screening solutions constitute a feasible option in the context of a shortage of expert personnel. (Contains 7 tables and 2 footnotes.)

Background: In languages with lexical stress, reading aloud must include stress assignment. Stress information sources across languages include word-final letter sequences. Here, we examine whether such sequences account for stress assignment in Greek and whether this is attributable to absolute rules involving accenting morphemes or to probabilistic lexical information. Methods: Pseudowords were constructed to not resemble particular words and were suffixed with derivational morphemes associated with specific stress patterns, to be read aloud, presented either without a stress diacritic or with a diacritic congruent or incongruent with the suffix. Morphemes differed in whether or not there were stress competitors ending in the same letter sequences. Results: Stress was assigned consistent with the suffix in the absence of the diacritic, more so when there were no stress competitors in the lexicon. Conclusions: Results suggest a lexically based probabilistic mechanism taking into account pattern distributions rather than absolute rules based on the morphological accenting.

Attempts to evaluate the cognitive-motivational profiles of students with reading comprehension difficulties have been scarce. The purpose of the present study was twofold: (a) to assess the discriminatory validity of cognitive, motivational, affective, and psychopathological variables for identification of students with reading difficulties, and (b) to profile students with and without reading comprehension difficulties across those variables. Participants were 87 students who scored more than 1.3 SD below the mean on a standardized reading comprehension battery and 500 typical students in grades 2 through 4. Results using linear discriminant analyses indicated that students with reading comprehension difficulties could be accurately predicted by low cognitive skills and high competitiveness. Using cluster analysis, students with significant deficits in reading comprehension were mostly assigned to a low skill/low motivation group (termed helpless) or a low skill/high motivation group (termed motivated low achievers). Based on these findings, it was concluded that motivation, emotions, and psychopathology play a pivotal role in explaining the achievement tendencies of students with reading comprehension difficulties. (Contains 5 figures, 5 tables and 3 notes.)

Although it is established that reading fluency is more strongly related to serial naming of symbols than to naming of isolated items ("serial superiority effect"), the reason for the difference remains unclear. The purpose of this study was to examine the role of executive functions in explaining the serial superiority effect. One hundred seven Grade 6 Greek children were assessed on serial and discrete naming (digits, objects, and words), executive (inhibition, shifting, and updating) and non-executive tasks (simple choice reaction), and on a serial Rapid Alternating Stimuli task. Reading fluency correlated more strongly with serial naming than with discrete naming, consistent with the serial superiority effect. In hierarchical regression analyses, executive measures failed to account for variance shared between serial naming and reading fluency. In confirmatory factor analyses, including a discrete and a serial factor for the naming tasks, variance in the executive tasks not shared with simple choice reaction was not associated with the serial factor. Thus, the executive tasks failed to account for the serial superiority effect. The high correlation between the simple choice factor and the discrete naming factor suggests that method variance partially underlies the observed relationship between executive function tasks and word reading. We argue that the distinction between serial and discrete dimensions indicates that internally regulated cognitive control is crucial for the serial superiority in naming symbols and words.

Effects of lexical and sublexical variables on visual word recognition are often treated as homogeneous across participants and stable over time. In this study, we examine the modulation of frequency, length, syllable and bigram frequency, orthographic neighborhood, and graphophonemic consistency effects by (a) individual differences, and (b) item repetition. A group of 129 participants performed lexical decision and naming, in counterbalanced order, using a set of 150 Greek words in which these variables were decorrelated. Frequency, length, and syllable frequency effects were reduced by a preceding task. Length effects were inversely related to years of education. Neighborhood effects depended on the metric used. There were no significant effects or interactions of bigram frequency or consistency. The results suggest that exposure to a word causes transient effects that may cumulatively develop into permanent individual differences. Models of word recognition must incorporate item-specific learning to account for these findings.

The concept of Matthew effects in reading development refers to a longitudinally widening gap between high achievers and low achievers. Various statistical approaches have been proposed to examine this idea. However, little attention has been paid to psychometric issues of scaling. Specifically, interval-level data are required to compare performance differences across performance ranges, but only ordinal-level data are available with current literacy measures. To demonstrate the interpretability problems of contrasting growth slopes, we use data from a longitudinal study of literacy development. We explore the possibility of comparing across ages, matched for performance, and we examine the consequences of nonlinear growth, temporal lag estimates, and individual differences in developmental progression. We conclude that, although conceptually appealing, the widening gap prediction is not empirically testable.

Dyslexic difficulties in lexical stress were compared to difficulties in segmental phonology. Twenty-nine adolescents with dyslexia and 29 typically developing adolescents, matched on age and nonverbal ability, were assessed on reading, spelling, phonological and stress awareness, rapid naming, and short-term memory. Group differences in stress assignment were larger than in segmental phonology in reading and spelling pseudowords but not words, indicating a fragility of explicit processes that manipulate stress representations. Despite impaired stress performance in dyslexia at the group level, individual variability failed to reveal evidence for a stress-specific deficit or for a distinct stress-impaired subgroup.

In this study we propose a classification system for spelling errors and determine the most common spelling difficulties of Greek children with and without dyslexia. Spelling skills of 542 children from the general population and 44 children with dyslexia, Grades 3-4 and 7, were assessed with a dictated common word list and age-appropriate passages. Spelling errors were classified into broad categories, including phonological (graphophonemic mappings), grammatical (inflectional suffixes), orthographic (word stems), stress assignment (diacritic), and punctuation. Errors were further classified into specific subcategories. Relative proportions for a total of 11,364 errors were derived by calculating the opportunities for each error type. Nondyslexic children of both age groups made primarily grammatical and stress errors, followed by orthographic 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, especially in inflectional suffixes. Dyslexic children differed from nondyslexic ones in making more errors of the same types, in comparable relative proportions. Spelling profiles of dyslexic children did not differ from those of same-age children with poor reading skills or of younger children matched in reading and phonological awareness. In conclusion, 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.

The simple view of reading posits that reading comprehension can be decomposed into a print-specific component (concerning decoding and sight word reading) and a language comprehension component (concerning verbal and metalinguistic skills not related to print). One might properly consider lexical skills, indexed by vocabulary measures, part of the language component; however, vocabulary measures end up taking up substantial amounts of print-dependent reading comprehension variance, presumably because of the interrelations among semantic, orthographic, and phonological specification of lexical entries. In the present study we examined the role of vocabulary in the prediction of reading comprehension by testing alternative formulations within the context of the simple view. We used cross-sectional and (1-year) longitudinal data from 436 children in Grades 3-6 attending regular classrooms. We quantified the proportion of variance accounting for reading comprehension that could be attributed to vocabulary measures. We then tested a latent variable model positing a mediating position for vocabulary against a model with lexically based covariation among the simple view components. We discuss the results in an attempt to bring together the simple view with the lexical quality hypothesis for reading comprehension. (Contains 8 tables, 4 figures, and 1 footnote)

Rapid automatized naming (RAN) is strongly correlated with reading fluency. A substantial part of this correlation is ascribed to the serial nature of the task. In this study, we tested the hypothesis that the left-to-right and downward scanning direction of reading and RAN may partially underlie their relationship. 107 Grade 6 Greek children were assessed on RAN digits and objects, a reversed-direction version of RAN digits, and word- and passage-reading fluency. The correlations of regular RAN with reading were not larger than those of RAN backward, ruling out visual scanning as an explanation of the relation between RAN and reading.

Labels for the categories have been found to facilitate learning by boosting accuracy. According to the label-feedback hypothesis, this facilitation is due to a mechanism selectively sensitising perceptual dimensions. To further investigate the label-facilitation phenomenon, one group of participants in our study learned both named and hard-to-name artificial categories, in a novel, within-subjects design. Another group of participants was administered a—highly similar—paired-associate task purportedly not involving sensitization of dimensions. Results showed that labels boosted accuracy during learning, but only when learning to categorise—not when learning to associate. The label-feedback hypothesis posits that labels exert an influence also after new categories have been learned. To test for sustained effects of labels, we administered a post-learning visual discrimination task while monitoring participants' eye movements and analysing dwell time on the trained shapes. There was some indication of sustained effects of labels for newly-learned categories, but there was no effect following learning to associate. Our results suggest that labels for newly learned categories have immediate effects during learning and that the effects of labels may also be sustained during post-learning processing. [ABSTRACT FROM AUTHOR]

Cognitive control has been typically examined using single-item tasks. This has implications for the generalizability of theories of control implementation. Previous studies have revealed that different control demands are posed by tasks depending on whether they present stimuli individually (i.e., single-item) or simultaneously in array format (i.e., multi-item). In the present study, we tracked within-task performance in single-item and multi-item Stroop tasks using simultaneous pupillometry, gaze, and behavioral response measures, aiming to explore the implications of format differences for cognitive control. The results indicated within-task performance decline in the multi-item version of the Stroop task, accompanied by pupil constriction and dwell time increase, in both the incongruent and the neutral condition. In contrast, no performance decline or dwell time increase was observed in the course of the single-item version of the task. We interpret these findings in terms of capacity constraints on cognitive control, with implications for cognitive control research, and highlight the need for better understanding of the cognitive demands of multi-item tasks. Public Significance Statement: Although cognitive control is an essential part of our everyday functioning with control failures leading to maladaptive behaviors, laboratory tasks used for the investigation of cognitive control are often limited to single-item presentation posing presumably minimal control requirements on the individuals. Having as starting point the observation that real-life situations are more complex and often demand the parallel execution of different tasks (i.e., multi-tasking), we evaluated and compared within-task performance in otherwise similar tasks differing only on stimuli presentation (one-by-one or simultaneously), while at the same time, we took advantage of pupillometry and gaze measures as additional indexes of processing demands. Altogether, our results show that cognitive control implementation depends highly on the task, with more complex tasks leading to performance deterioration, and suggest that the use of primarily simple, laboratory tasks might restrict our understanding of cognitive control implementation in our everyday life. [ABSTRACT FROM AUTHOR]

Learning nonnative speech contrasts in adulthood has proven difficult. Standard training methods have achieved moderate effects using explicit instructions and performance feedback. In this study, the authors question preexisting assumptions by demonstrating a superiority of implicit training procedures. They trained 3 groups of Greek adults on a difficult Hindi contrast (a) explicitly, with feedback (Experiment 1), or (b) implicitly, unaware of the phoneme distinctions, with (Experiment 2) or without (Experiment 3) feedback. Stimuli were natural recordings of consonant-vowel syllables with retroflex and dental unvoiced stops by a native Hindi speaker. On each trial, participants heard pairs of tokens from both categories and had to identify the retroflex sounds (explicit condition) or the sounds differing in intensity (implicit condition). Unbeknownst to participants, in the implicit conditions, target sounds were always retroflex, and distractor sounds were always dental. Post-training identification and discrimination tests showed improved performance of all groups, compared with a baseline of untrained Greek listeners. Learning was most robust for implicit training without feedback. It remains to be investigated whether implicitly trained skills can generalize to linguistically relevant phonetic categories when appropriate variability is introduced. These findings challenge traditional accounts on the role of feedback in phonetic training and highlight the importance of implicit, reward-based mechanisms. (Contains 4 tables and 3 figures.)

The simple view of reading admits two components in accounting for individual differences in reading comprehension: a print-dependent component related to decoding and word identification, and a print-independent one related to oral language comprehension. It has been debated whether word or nonword reading is a better index of the print-dependent component and whether vocabulary measures fit within the print-independent component or constitute an additional factor. Here we apply a confirmatory factor analysis on a set of relevant measures from 488 Greek children in Grades 3-5 independently of reading comprehension. The results indicate that word and nonword reading do not constitute distinct factors but covary along the same two dimensions of accuracy and fluency. Oral vocabulary measures group with listening comprehension, resulting in excellent model fits. Strong correlations were observed between the latent factors of the purported print-dependent and print-independent components, consistent with an approach that focuses on the strong relations among semantic, orthographic, and phonological aspects of word representations. (Contains 3 tables and 4 figures.)

Assessment of lexical/semantic knowledge is performed with a variety of tests varying in response requirements. The present study exemplifies the application of modern statistical approaches in the adaptation and assessment of the psychometric properties of the "Peabody Picture Vocabulary Test--Revised" (PPVT-R) Greek. Confirmatory factor analyses applied to data from a large sample of elementary school students (N = 585) indicated the existence of a single vocabulary dimension and differential item functioning procedures pointed to minimal bias due to gender or ethnic group. Rasch model-derived indices of item difficulty and discrimination were used to develop a short form of the test, which was administered to a second sample of 900 students. Convergent and discriminant validity were assessed through comparisons with the "Wechsler Intelligence Scales for Children"--III Vocabulary and Block design subtests. Short- and long-term stability of individual scores over a 6-month period were very high, and the utility of the test as part of routine educational assessment is attested by its strong longitudinal predictive value with reading comprehension measures. It is concluded that the Greek version of the PPVT-R constitutes a reliable and valid assessment of vocabulary for Greek students and immigrants who speak Greek. (Contains 4 tables and 6 figures.)

The presence of Matthew effects was tested in students of varying reading, spelling, and vocabulary skills. A cross-sequential design was implemented, following 587 Grade 2 through 4 students across five measurement points (waves) over 2 years. Students were administered standardized assessments of reading, spelling, and vocabulary. Results indicated that the hypothesized fan-spread pattern for Matthew effects was not evident. Low and high ability groups were formed based on 25th and 75th percentile cutoffs on initial measures of spelling, reading accuracy and fluency, vocabulary, and reading comprehension. Multilevel modeling suggested that low and high ability groups had significantly different starting points (intercepts) and their pattern of growth on passage comprehension did not indicate that the gap would increase over time. Instead, some analyses, especially of the youngest cohorts, showed significant convergence. However, there was no evidence of eventually closing the gap. Thus, although the poor students may not be getting poorer, they do not get sufficiently richer either. (Contains 3 tables and 5 figures.)

Learning a second language as an adult is particularly effortful when new phonetic representations must be formed. Therefore the processes that allow learning of speech sounds are of great theoretical and practical interest. Here we examined whether perception of single formant transitions, that is, sound components critical in speech perception, can be enhanced through an implicit task-irrelevant learning procedure that has been shown to produce visual perceptual learning. The single-formant sounds were paired at subthreshold levels with the attended targets in an auditory identification task. Results showed that task-irrelevant learning occurred for the unattended stimuli. Surprisingly, the magnitude of this learning effect was similar to that following explicit training on auditory formant transition detection using discriminable stimuli in an adaptive procedure, whereas explicit training on the subthreshold stimuli produced no learning. These results suggest that in adults learning of speech parts can occur at least partially through implicit mechanisms. (Contains 8 figures.)

In Greek orthography, stress position is marked with a diacritic. We investigated the developmental course of processing the stress diacritic in Grades 2 to 4. Ninety children read 108 pseudowords presented without or with a diacritic either in the same or in a different position relative to the source word. Half of the pseudowords resembled the words they were derived from. Results showed that lexical sources of stress assignment were active in Grade 2 and remained stronger than the diacritic through Grade 4. The effect of the diacritic increased more rapidly and approached the lexical effect with increasing grade. In a second experiment, 90 children read 54 words and 54 pseudowords. The pattern of results for words was similar to that for nonwords suggesting that findings regarding stress assignment using nonwords may generalize to word reading. Decoding of the diacritic does not appear to be the preferred option for developing readers. (Contains 7 tables and 3 footnotes.)

To assign lexical stress when reading, the Greek reader can potentially rely on lexical information (knowledge of the word), visual-orthographic information (processing of the written diacritic), or a default metrical strategy (penultimate stress pattern). Previous studies with secondary education children have shown strong lexical effects on stress assignment and have provided evidence for a default pattern. Here we report two experiments with adult readers, in which we disentangle and quantify the effects of these three potential sources using nonword materials. Stimuli either resembled or did not resemble real words, to manipulate availability of lexical information; and they were presented with or without a diacritic, in a word-congruent or word-incongruent position, to contrast the relative importance of the three sources. Dual-task conditions, in which cognitive load during nonword reading was increased with phonological retention carrying a metrical pattern different from the default, did not support the hypothesis that the default arises from cumulative lexical activation in working memory.