Constraints on computational models of basic processes in reading

There are numerous reports in the visual word recognition literature that the joint effects of various factors are additive on reaction time. A central claim by D. C. Plaut and J. R. Booth (2000, 2006) is that their parallel distributed processing model simulates additive effects of stimulus quality and word frequency in the context of lexical decision. If correct, this success would have important implications for computational accounts of reading processes. However, the results of further simulations with this model undermine this claim given that the joint effects of stimulus quality and word frequency yield a nonmonotonic function (underadditivity, additivity, and overadditivity) depending on the size of the stimulus quality effect, whereas skilled readers yield additivity more broadly. The implications of these results both locally and more globally are discussed, and a number of other issues are noted. Additivity of factor effects constitutes a benchmark that computational accounts should strive to meet.     

Parallel distributed processing and lexical-semantic effects in visual word recognition: are a few stages necessary?

D. C. Plaut and J. R. Booth (2000) presented a parallel distributed processing model that purports to simulate human lexical decision performance. This model (and D. C. Plaut, 1995) offers a single mechanism account of the pattern of factor effects on reaction time (RT) between semantic priming, word frequency, and stimulus quality without requiring a stages-of-processing account of additive effects. Three problems are discussed. First, no evidence is provided that this model can discriminate between words and nonwords with the same orthographic structure and still produce the pattern of factor effects on RT it currently claims to produce. Second, the level of representation used by the model to make a lexical decision is inconsistent with what is known about how skilled readers with damage to their semantic system make word/nonword discriminations. Finally, there are a number of results that are difficult to reconcile with the single mechanism account. The authors' preference is to retain the stages-of-processing account.     

Priming and attentional control of lexical and sublexical pathways during naming

A modified priming task was used to investigate whether skilled readers are able to adjust the degree to which lexical and sublexical information contribute to naming. On each trial, participants named 5 low-frequency exception word primes or 5 nonword primes before a target. The low-frequency exception word primes should have produced a greater dependence on lexical information, whereas the nonword primes should have produced a greater dependence on sublexical information. Across 4 experiments, the effects of lexicality, regularity, frequency, and imageability were all modulated in predictable ways on the basis of the notion that the primes directed attention to specific processing pathways. It is argued that these results are consistent with an attentional control hypothesis.     

When children write nonwords to dictation

Skill in written spelling of simple, monosyllabic nonwords was investigated in 9- to 11-year-old English children. Two aspects of their spellings were of interest: first, could they spell these nonwords so that they sounded correct (nonword spelling accuracy), and second, did their spellings show evidence of biasing from words heard earlier in the test sequence? Nonword spelling was poorer for children of this age than for tested adults. Nevertheless, significant biasing occurred in these children's spellings, though not to the same extent as in adults' nonword spellings, and significant correlations emerged between reading age, nonword spelling skill, susceptibility to biasing, and real word spelling skill. Children with a reading age greater than 11 years showed biasing from word spellings that was within range of that reported for adults, and, for these more skilled readers, word spelling accuracy correlated significantly with both susceptibility to biasing and with nonword spelling accuracy. These children were not as accurate as tested adults at spelling nonwords. Children with a reading age below 11 years were poorer at nonword spelling and showed no overall biasing, yet they also showed a significant correlation between word spelling skill and nonword biasing. Together with evidence from the same task from adults with specific spelling disorders, these results suggest that word knowledge had a direct (biasing) and an indirect (general word spelling knowledge) effect on the performance of the nonword spelling task. But although skill in word spelling may be a necessary prerequisite for nonword spelling, it need not always be sufficient.     

Lexical access codes in visual word recognition: are the joint effects of context and stimulus quality diagnostic?

Is it possible to identify when lexical access in visual word recognition is based on an orthographic code and when it is based on a phonological code? Some researchers have argued that the joint effects of semantic context and stimulus quality in lexical decision are diagnostic. Their argument is that when phonological recoding is employed, it serves to keep the effects of stimulus quality and context from interacting. Semantic context and stimulus quality are therefore predicted to have additive effects on RT. In contrast, when lexical access is not mediated by a phonological recoding stage, then the effect of stimulus quality interacts with context, as commonly seen in the literature. A strong test of these claims was devised in which participants were forced to use phonological recoding for the purpose of lexical access. An interaction between context and stimulus quality was observed. This finding is taken as evidence that the joint effects of semantic context and stimulus quality are not diagnostic with respect to the nature of the lexical access code (i.e., orthographic versus phonological) for readers of English.     

There goes the neighbourhood: contextual control over the breadth of lexical activation when reading aloud

There are currently two computational accounts of how the time to read pseudohomophones (like BRANE) and their nonword controls (like FRANE) varies with changes in context. In Reynolds and Besner's (2005) account, readers vary the breadth of lexical activation in response to changes in context. A competing account proposed by Kwantes and Marmurek (2007) and independently by Perry, Ziegler, and Zorzi (2007) has readers varying their response criterion in response to changes in context. The present work adjudicates between these two accounts by examining how the effect of neighbourhood density changes as a function of list context when reading pseudohomophones aloud. The results of an experiment and simulations from a leading computational model support the lexical breadth account, but are inconsistent with the response criterion account.     

Evaluating current deficit theories of poor reading: role of phonological processing, naming speed, balance automaticity, rapid verbal perception and working memory

To clarify the nature of cognitive deficits experienced by poor readers, 9-10-yr.-old poor readers were matched against 9 chronological age and 9 younger reading age-matched controls screened and selected from regular classrooms. Poor readers performed significantly more poorly than chronological age-matched peers on digit naming speed, spoonerisms, and nonsense word reading. Poor readers were also significantly poorer than reading age-matched controls on nonword reading but were significantly better than reading age-matched controls on postural stability. Analyses of effect sizes were consistent with these findings, showing strong effects for digit naming speed, spoonerisms, and nonword reading. However, effect size analysis also suggested that poor readers experienced moderate difficulties with balance automatisation but did not show verbal speech perception deficits relative to either control     

Orthographic neighborhood effects in lexical decision: the effects of nonword orthographic neighborhood size

The effects of large neighborhoods (neighborhood size) and of higher frequency neighbors (neighborhood frequency) were examined as a function of nonword neighborhood size in lexical decision tasks. According to the multiple read-out model (J. Grainger & A. M. Jacobs, 1996), neighborhood size and neighborhood frequency effects should vary systematically as a function of nonword neighborhood size. In these experiments, the nonword context was more extensively manipulated than in previous studies, providing a more complete test of the model's predictions. In addition, simulations were conducted examining the model's ability to account for the facilitatory neighborhood size and neighborhood frequency effects observed in these experiments. The results suggest that the model overestimates the role of inhibition in the orthographic processing of English words.     

中文阅读中快速读者与慢速读者的词频效应

在熟练的中文阅读中,个体阅读效率越高,词汇加工效率是否越高,目前还缺少直接证据。本研究通过记录快速读者与慢速读者在阅读含有高频或低频目标词句子时的眼动轨迹来考察该问题。结果发现,在第一遍注视时间、回视路径时间和总注视时间上,快速读者对高频词与低频词的注视时间无显著差异,但都短于慢速读者;而慢速读者对低频词的注视时间显著长于高频词。即快速读者的早期词汇识别和后期词汇整合效率都较高,不受词频影响;而慢速读者的整个词汇加工效率较低,受词频影响较大。这些结果表明两组读者的词汇加工模式具有本质差异,符合个体差异性观点与词汇质量假说。     

词汇预测性对中文高、低阅读技能儿童眼动行为的影响

操纵目标词的预测性、词频和阅读技能水平,考察句子阅读中词汇预测性对高、低阅读技能儿童眼动行为的影响,揭示其在儿童阅读发展中的作用。结果显示：儿童对高预测词的跳读率更高、注视时间更短,且预测性与词频交互影响跳读率和注视时间;预测性对高阅读技能儿童早期的跳读率影响更大,而对低阅读技能儿童晚期的再阅读时间具有更大影响。结果表明：词汇预测性影响儿童阅读的眼动行为和词汇加工,且作用大小和发生时程受阅读技能调节。     

Children's reading of words, pseudohomophones, and other nonwords

In Experiment 1 children aged 8-9 and 9-10 years were tested for neighbourhood and pseudohomophone effects in nonword reading. Neighbourhood effects ( N effects) were robust irrespective of group or type of nonword. Pseudohomophones were read more accurately than other nonwords but this finding was robust only for the younger 8-9-year-olds. High-frequency words were read more accurately than low-frequency words, but the reverse applied to pseudohomophones based on high- and low-frequency words, although this was not robust. Error rates for the 9-10-year-olds in Experiment 1 were low, and so it was difficult to interpret the lack of a pseudohomophone advantage for reading nonwords in this age group. Experiment 2 was therefore carried out, which consisted of a replication of the first study with a further group of 9- 10-year-olds, but pronunciation latencies were measured, as well as accuracy. All the effects obtained in Experiment 1 were replicated but, in addition, an advantage for pseudohomophones in terms of pronunciation latencies was observed. The implications for accounts of reading development are discussed.     

Frequency and predictability effects in eye fixations for skilled and less-skilled deaf readers

The illiteracy rate in the deaf population has been alarmingly high for several decades, despite the fact that deaf children go through the standard stages of schooling. Much research addressing this issue has focused on word-level processes, but in the recent years, little research has focused on sentence-levels processes. Previous research (Fischler, 1985) investigated word integration within context in college-level deaf and hearing readers in a lexical decision task following incomplete sentences with targets that were congruous or incongruous relative to the preceding context; it was found that deaf readers, as a group, were more dependent on contextual information than their hearing counterparts. The present experiment extended Fischler's results and investigated the relationship between frequency, predictability, and reading skill in skilled hearing, skilled deaf, and less-skilled deaf readers. Results suggest that only less-skilled deaf readers, and not all deaf readers, rely more on contextual cues to boost word processing. Additionally, early effects of frequency and predictability were found for all three groups of readers, without any evidence for an interaction between frequency and predictability.     

On the joint effects of repetition and stimulus quality in lexical decision: looking to the past for a new way forward

Two experiments investigated the joint effects of stimulus quality and repetition in the context of lexical decision. Experiment 1 yielded an interaction between repetition and stimulus quality for words (but additive effects for nonwords) when the lag was short, replicating previous reports. Experiment 2, with a much longer lag than Experiment 1, yielded main effects of stimulus quality and repetition, but these factors no longer interact. The joint effects of stimulus quality and repetition for words as a function of lag can be understood in terms of two loci for repetition effects: one short-term and one long-term. The transient effect of repetition is on activation levels in the lexicons (and in which the input lexicon, but not beyond, is affected by stimulus quality), whereas the long-term effect is on the strength of two-way connections between lexical-lexical and lexical-semantic modules. These data and others, taken together with the account, provide a new way of thinking about a 30-year-old conundrum.     

Word,nonword, and visual paired associate learning in Dutch dyslexic children

Verbal and non-verbal learning were investigated in 21 8-11-year-old dyslexic children and chronological-age controls, and in 21 7-9-year-old reading-age controls. Tasks involved the paired associate learning of words, nonwords, or symbols with pictures. Both learning and retention of associations were examined. Results indicated that dyslexic children had difficulty with verbal learning of both words and nonwords. In addition, analysis of the errors made during nonword learning showed that both phonological errors and general learning errors were distributed similarly for the reading groups. This suggests that nonword learning in dyslexics is slower, but not qualitatively different from normal readers. Furthermore, no differences were found between the dyslexics and age-matched normal readers on non-verbal learning. Long-term retention of the learned visual-verbal associations (both words and nonwords) was not impaired in dyslexic children as compared to normal readers. Finally, phonological awareness ability was assessed. Dyslexics performed worse than age-matched normal readers, but similar to reading-age controls.     

On the joint effects of repetition and stimulus quality in lexical decision: Looking to the past for a new way forward

Two experiments investigated the joint effects of stimulus quality and repetition in the context of lexical decision. Experiment 1 yielded an interaction between repetition and stimulus quality for words (but additive effects for nonwords) when the lag was short, replicating previous reports. Experiment 2, with a much longer lag than Experiment 1, yielded main effects of stimulus quality and repetition, but these factors no longer interact. The joint effects of stimulus quality and repetition for words as a function of lag can be understood in terms of two loci for repetition effects: one short-term and one long-term. The transient effect of repetition is on activation levels in the lexicons (and in which the input lexicon, but not beyond, is affected by stimulus quality), whereas the long-term effect is on the strength of two-way connections between lexical–lexical and lexical–semantic modules. These data and others, taken together with the account, provide a new way of thinking about a 30-year-old conundrum.     

The impact of reader skill on phonological processing in visual word recognition

There has been much debate about the role of phonology in reading. This debate has been fuelled, in part, by mixed findings for phonological effects in lexical decision tasks. In the present research we investigated the impact of reader skill on three phonological effects (homophone, homograph, and regularity effects) in a lexical decision task and in a phonological lexical decision task. In both tasks, the more skilled readers showed different patterns of phonological effects from those of the less skilled readers; in particular, less skilled readers showed regularity effects in both tasks whereas more skilled readers did not. We concluded that more skilled readers activate phonology in these tasks but do so more efficiently, with less spurious phonological activation.     

Developmental and individual differences in performance on phonological synthesis tasks

This study was conducted in order to examine the role of individual differences in working memory and lexical access in accounting for age and reading skill related differences in performance on phonological synthesis tasks. The performance of 28 kindergarten, first-, and second-grade children with normal reading skills, as well as that of 28 reading-disabled second graders, was compared under four testing conditions. The testing conditions were formed by completely crossing rate of presentation of phoneme strings with type of stimulus to be blended (real or nonword). Both independent variables affected performance on the blending tasks, with better performance obtained at faster presentation rates and with real words. There was a developmental discontinuity in overall performance, with the kindergarten children obtaining substantially lower scores than the first or second graders. In the comparison of second-grade good and poor readers, there was a significant interaction between groups, presentation rates, and type of stimulus. The findings are discussed in terms of their implications for a general model of performance on the sound blending task, as well as their value for interpreting individual differences on the task.     

Does a sensory processing deficit explain counting accuracy on rapid visual sequencing tasks in adults with and without dyslexia?

The experiments conducted aimed to investigate whether reduced accuracy when counting stimuli presented in rapid temporal sequence in adults with dyslexia could be explained by a sensory processing deficit, a general slowing in processing speed or difficulties shifting attention between stimuli. To achieve these aims, the influence of the inter-stimulus interval (ISI), stimulus duration, and sequence length were evaluated in two experiments. In the first that used skilled readers only, significantly more errors were found with presentation of long sequences when the ISI or stimulus durations were short. Experiment 2 used a wider range of ISIs and stimulus durations. Compared to skilled readers, a group with dyslexia had reduced accuracy on two-stimulus sequences when the ISI was short, but not when the ISI was long. Although reduced accuracy was found on all short and long sequences by the group with dyslexia, when performance on two-stimulus sequences was used as an index of sensory processing efficiency and controlled, group differences were found with presentation of stimuli of short duration only. We concluded that continuous, repetitive stimulation to the same visual area can produce a capacity limitation on rapid counting tasks in all readers when the ISIs or stimulus durations are short. While reduced accuracy on rapid sequential counting tasks can be explained by a sensory processing deficit when the stimulus duration is long, slower processing speed in the group with dyslexia explains the greater inaccuracy found as sequence length is increased when the stimulus duration is short.     

Not all skilled readers have cracked the code: individual differences in masked form priming

This experiment investigated whether individual differences in written language proficiency among university students predict the early stages of lexical retrieval tapped by the masked form priming lexical decision task. To separate the contributions of sublexical facilitation and lexical competition to masked form priming, the effects of prime lexicality were directly compared for both transposed-letter (TL) primes (e.g., sung SNUG; salb SLAB) and neighbor primes (e.g., snag SNUG; sleb SLAB) in a sample of 100 university students assessed on measures of reading, spelling and vocabulary. The data for the whole sample showed facilitation from nonword primes, but inhibition from word primes. Linear mixed models including the individual difference measures showed that higher scores on a principal component that captured the shared variance among reading, spelling, and vocabulary were associated with both stronger inhibition from TL word primes and stronger facilitation from neighbor nonword primes. These findings are consistent with the lexical quality hypothesis of reading that predicts that skilled readers vary in the extent to which they have developed precisely specified orthographic representations.     

Dynamic relationships between phonological memory and reading: A five year longitudinal study from age 4 to 9

We reconcile competing theories of the role of phonological memory in reading development, by uncovering their dynamic relationship during the first 5 years of school. Phonological memory, reading and phoneme awareness were assessed in 780 phonics‐educated children at age 4, 5, 6 and 9. Confirmatory factor analyses demonstrated that phonological memory loaded onto two factors: verbal short‐term memory (verbal STM; phonological tasks that loaded primarily on serial order memory) and nonword repetition. Using longitudinal structural equation models, we found that verbal STM directly predicted early word‐level reading from age 4 to 6, reflecting the importance of serial‐order memory for letter‐by‐letter decoding. In contrast, reading had no reciprocal influence on the development of verbal STM. The relationship between nonword repetition and reading was bidirectional across the 5 years of study: nonword repetition and reading predicted each other both directly and indirectly (via phoneme awareness). Indirect effects from nonword repetition (and verbal STM) to reading support the view that phonological memory stimulates phonemically detailed representations through repeated encoding of complex verbal stimuli. Similarly, the indirect influence of reading on nonword repetition suggests that improved reading ability promotes the phoneme‐level specificity of phonological representations. Finally, the direct influence from reading to nonword repetition suggests that better readers use orthographic cues to help them remember and repeat new words accurately. A video abstract of this article can be viewed at https://www.youtube.com/watch?v=70LZfTR0BjE .