The use of probabilistic lexicality cues for word segmentation in Chinese reading

In an eye-tracking experiment we examined whether Chinese readers were sensitive to information concerning how often a Chinese character appears as a single-character word versus the first character in a two-character word, and whether readers use this information to segment words and adjust the amount of parafoveal processing of subsequent characters during reading. Participants read sentences containing a two-character target word with its first character more or less likely to be a single-character word. The boundary paradigm was used. The boundary appeared between the first character and the second character of the target word, and we manipulated whether readers saw an identity or a pseudocharacter preview of the second character of the target. Linear mixed-effects models revealed reduced preview benefit from the second character when the first character was more likely to be a single-character word. This suggests that Chinese readers use probabilistic combinatorial information about the likelihood of a Chinese character being single-character word or a two-character word online to modulate the extent of parafoveal processing.     

Becoming a written word: Eye movements reveal order of acquisition effects following incidental exposure to new words during silent reading

We know that from mid-childhood onwards most new words are learned implicitly via reading; however, most word learning studies have taught novel items explicitly. We examined incidental word learning during reading by focusing on the well-documented finding that words which are acquired early in life are processed more quickly than those acquired later. Novel words were embedded in meaningful sentences and were presented to adult readers early (day 1) or later (day 2) during a five-day exposure phase. At test adults read the novel words in semantically neutral sentences. Participants’ eye movements were monitored throughout exposure and test. Adults also completed a surprise memory test in which they had to match each novel word with its definition. Results showed a decrease in reading times for all novel words over exposure, and significantly longer total reading times at test for early than late novel words. Early-presented novel words were also remembered better in the offline test. Our results show that order of presentation influences processing time early in the course of acquiring a new word, consistent with partial and incremental growth in knowledge occurring as a function of an individual’s experience with each word.     

Context effects on orthographic learning of regular and irregular words

The self-teaching hypothesis proposes that orthographic learning takes place via phonological decoding in meaningful texts, that is, in context. Context is proposed to be important in learning to read, especially when decoding is only partial. However, little research has directly explored this hypothesis. The current 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. Two experiments were conducted using regular and irregular novel words, respectively. Second-grade children were asked to learn eight novel words either in stories or in a list of words. The results revealed no significant effect of context for the regular items. However, in an orthographic decision task, there was a facilitatory effect of context on irregular novel word learning. The findings support the view that contextual information is important to orthographic learning, but only when the words to be learned contain irregular spelling-sound correspondences.     

The influence of visual contrast and case changes on parafoveal preview benefits during reading

Reingold and Rayner (2006) showed that the visual contrast of a fixated target word influenced its viewing duration, but not the viewing of the next (posttarget) word in the text that was shown in regular contrast. Configurational target changes, by contrast, influenced target and posttarget viewing. The current study examined whether this effect pattern can be attributed to differential processing of the posttarget word during target viewing. A boundary paradigm (Rayner, 1975) was used to provide an informative or uninformative posttarget preview and to reveal the word when it was fixated. Consistent with the earlier study, more time was spent viewing the target when its visual contrast was low and its configuration unfamiliar. Critically, target contrast had no effect on the acquisition of useful information from a posttarget preview, but an unfamiliar target configuration diminished the usefulness of an informative posttarget preview. These findings are consistent with Reingold and Rayner's (2006) claim that saccade programming and attention shifting during reading can be controlled by functionally distinct word recognition processes.     

Phonotactic regularities in the segmentation of spoken Italian

Five word-spotting experiments explored the role of consonantal and vocalic phonotactic cues in the segmentation of spoken Italian. The first set of experiments tested listeners’ sensitivity to phonotactic constraints cueing syllable boundaries. Participants were slower in spotting words in nonsense strings when target onsets were misaligned (e.g., lago in ri.blago) than when they were aligned (e.g., lago in rin.lago) with phonotactically determined syllabic boundaries. This effect held also for sequences that occur only word-medially (e.g., /tl/ in ri.tlago), and competition effects could not account for the disadvantage in the misaligned condition. Similarly, target detections were slower when their offsets were misaligned (e.g., cittá in cittáu.ba) than when they were aligned (e.g., cittá in cittá.oba) with a phonotactic syllabic boundary. The second set of experiments tested listeners’ sensitivity to phonotactic cues, which specifically signal lexical (and not just syllable) boundaries. Results corroborate the role of syllabic information in speech segmentation and suggest that Italian listeners make little use of additional phonotactic information that specifically cues word boundaries.     

The role of phonology in incidental vocabulary acquisition during silent reading

In two experiments, we investigated the role of phonology in learning new words incidentally during silent reading. Participants read sentence pairs containing novel or known words that varied in homophony (whether another word exists with an alternate spelling for the same pronunciation). In Experiment 1, we monitored readers' eye movements to investigate online processes involved in establishing meanings for novel words. In Experiment 2, participants completed cued recall and vocabulary recognition tasks after the reading session to assess the influence of phonological form on word learning. Eye movement results indicate that readers spent the most time reading novel homophones (e.g., skwosh) and surrounding context, indicating that phonological information is activated early during a reader's initial encounter with a new letter string. Retention measures suggest that readers were able to infer a meaning for each novel word type, despite the increased difficulty associated with reading novel words with familiar phonological forms, and that phonology aided the acquisition of orthography.     

The relationship between reading ability and lateralized lexical decision

Although lexical decision remains one of the most extensively studied cognitive tasks, very little is known about its relationship to broader linguistic performance such as reading ability. In a correlational study, several aspects of lateralized lexical decision performance were related to vocabulary and reading comprehension measures, as assessed using the Nelson-Denny Reading Test. This lateralized lexical decision task has been previously shown to demonstrate (1) independent contributions from both hemispheres, as well as (2) interhemispheric interactions during word recognition. Lexical decision performance showed strong relationships with both reading measures. Specifically, vocabulary performance correlated significantly with left visual field (LVF) word accuracy and LVF non-word latency, both measures of right hemisphere performance. There were also significant, though somewhat weaker, correlations between reading comprehension and RVF non-word latency. Lexicality priming, a measure of interhemispheric communication during lexical decision, was also correlated with reading comprehension. These results suggest that hemispheric interaction during word recognition is common, and that lexical processing contribution from the right hemisphere, something commonly taken as minor and inconsequential, can lead to significant performance benefits and to individual differences in reading.     

中文阅读中的字词激活模式:来自提示词边界延时效应的证据

研究采用眼动随动显示技术,通过操控提示正确/错误词边界线索的延迟时间,考察汉语阅读中汉字、词汇加工与词切分的时间特性。实验1发现,提示词n+1边界不影响总阅读时间,但能积极影响词兴趣区内的眼动数据,这种影响随着提示延迟时间的增加而逐步减弱;提示错误的词n+1的边界线索对总阅读时间的影响则随提示延迟时间的增加呈现倒"U"型的变化趋势。实验2发现,提示词n边界不影响总阅读时间,但却消极影响词兴趣区内的眼动数据,这种消极影响并没有随着提示延迟时间的增加而减弱;提示错误的词n边界对总阅读时间和眼动数据均产生消极的影响,且这种影响都随着提示延迟时间的增加而减弱。综合两项实验的结果可以推测,单一的向上反馈假设和整体假设都不能全面解释阅读中的字词加工过程,汉字加工与词汇加工间存在交互激活过程。     

The link between statistical segmentation and word learning in adults

Many studies have shown that listeners can segment words from running speech based on conditional probabilities of syllable transitions, suggesting that this statistical learning could be a foundational component of language learning. However, few studies have shown a direct link between statistical segmentation and word learning. We examined this possible link in adults by following a statistical segmentation exposure phase with an artificial lexicon learning phase. Participants were able to learn all novel object-label pairings, but pairings were learned faster when labels contained high probability (word-like) or non-occurring syllable transitions from the statistical segmentation phase than when they contained low probability (boundary-straddling) syllable transitions. This suggests that, for adults, labels inconsistent with expectations based on statistical learning are harder to learn than consistent or neutral labels. In contrast, a previous study found that infants learn consistent labels, but not inconsistent or neutral labels.     

Parafoveal processing across different lexical constituents in Chinese reading

We report a boundary paradigm eye movement experiment to investigate whether the linguistic category of a two-character Chinese string affects how the second character of that string is processed in the parafovea during reading. We obtained clear preview effects in all conditions but, more importantly, found parafoveal-on-foveal effects whereby a nonsense preview of the second character influenced fixations on the first character. This effect occurred for monomorphemic words, but not for compound words or phrases. Also, in a word boundary demarcation experiment, we demonstrate that Chinese readers are not always consistent in their judgements of which characters in a sentence constitute words. We conclude that information regarding the combinatorial properties of characters in Chinese is used online to moderate the extent to which parafoveal characters are processed.     

Testing the Role of Phonology in Reading: Focus on Sentence Processing

Most reading research investigating the role of phonology in word recognition has focused on studies employing an individual word as the sole stimulus. The bulk of such research has offered support for the phonological recoding hypothesis, the conjecture that access to a printed word’s meaning requires activation of the word’s phonology (i.e., meaning is not typically activated via orthography alone). A criticism of such studies is that by presenting participants with only a single word on each experimental trial (a nonecological manipulation), participants may alter their typical strategy of reading in such a way as to artificially favor the phonological recoding hypothesis. The present study avoided a focus on single words by requiring participants to read sentences and paragraphs for comprehension. Experiment 1 showed that, in reading a paragraph of connected sentences, eliminating a letter in a word that altered the phonology was more deleterious than eliminating a letter that did. Experiment 2 focused on the reading of each sentence itself rather than on the paragraph and provided additional control conditions. The results were similar to those of Experiment 1, consistent with the phonological recoding hypothesis.     

Learning diphone-based segmentation

This paper reconsiders the diphone-based word segmentation model of Cairns, Shillcock, Chater, and Levy (1997) and Hockema (2006), previously thought to be unlearnable. A statistically principled learning model is developed using Bayes' theorem and reasonable assumptions about infants' implicit knowledge. The ability to recover phrase-medial word boundaries is tested using phonetic corpora derived from spontaneous interactions with children and adults. The (unsupervised and semi-supervised) learning models are shown to exhibit several crucial properties. First, only a small amount of language exposure is required to achieve the model's ceiling performance, equivalent to between 1 day and 1 month of caregiver input. Second, the models are robust to variation, both in the free parameter and the input representation. Finally, both the learning and baseline models exhibit undersegmentation, argued to have significant ramifications for speech processing as a whole.     

Orthographic learning at a glance: on the time course and developmental onset of self-teaching

Experiment 1 examined the time course of orthographic learning among Grade 3 children. A single encounter with a novel orthographic string was sufficient to produce reliable recall of orthographic detail. Moreover, newly acquired orthographic information was retained 1 month later. These data support the logistic learning functions featured in contemporary connectionist models of reading rather than a "threshold" model of orthographic learning. Experiments 2 and 3 examined self-teaching among novice readers. In contrast to the findings from less regular orthographies such as English and Dutch, beginning readers of a highly regular orthography (Hebrew) appear to be relatively insensitive to word-specific orthographic detail, reading in a nonlexical "surface" fashion. These results suggest fundamental differences between shallow and deep orthographies in the development of orthographic sensitivity.     

Interhemispheric cooperation and non-cooperation during word recognition: evidence for callosal transfer dysfunction in dyslexic adults

Participants report briefly-presented words more accurately when two copies are presented, one in the left visual field (LVF) and another in the right visual field (RVF), than when only a single copy is presented. This effect is known as the 'redundant bilateral advantage' and has been interpreted as evidence for interhemispheric cooperation. We investigated the redundant bilateral advantage in dyslexic adults and matched controls as a means of assessing communication between the hemispheres in dyslexia. Consistent with previous research, normal adult readers in Experiment 1 showed significantly higher accuracy on a word report task when identical word stimuli were presented bilaterally, compared to unilateral RVF or LVF presentation. Dyslexics, however, did not show the bilateral advantage. In Experiment 2, words were presented above fixation, below fixation or in both positions. In this experiment both dyslexics and controls benefited from the redundant presentation. Experiment 3 combined whole words in one visual field with word fragments in the other visual field (the initial and final letters separated by spaces). Controls showed a bilateral advantage but dyslexics did not. In Experiments 1 and 3, the dyslexics showed significantly lower accuracy for LVF trials than controls, but the groups did not differ for RVF trials. The findings suggest that dyslexics have a problem of interhemispheric integration and not a general problem of processing two lexical inputs simultaneously.     

Recognition of partial regularity in words and sentences

The involvement of morphological representations in reading Swedish and Dutch was explored using a lexical decision task and a sentence continuation task. In Experiment 1, letter strings were studied which were composed of a root and derivational suffix, a nonroot and suffix, a root and nonsuffix, or a nonroot and nonsuffix. Latencies in identifying nonword items increased as the letter strings became morphologically regular. In Experiment 2, subjects chose among pairs of continuations to a sentence. One choice offered at a given point was acceptable in context, while the other was acceptable, or contained an inappropriate stem, an inappropriate inflection, or both together. The latency to choose a correct continuation increased as the foil offered became morphologically appropriate. The results show that use of morphemic representations in reading as reported by Caramazza et al. (1988) extends from the identification of single morphs to the integration of information across phrases.     

A Bayesian framework for word segmentation: Exploring the effects of context

Since the experiments of Saffran et al. [Saffran, J., Aslin, R., & Newport, E. (1996). Statistical learning in 8-month-old infants. Science, 274, 1926-1928], there has been a great deal of interest in the question of how statistical regularities in the speech stream might be used by infants to begin to identify individual words. In this work, we use computational modeling to explore the effects of different assumptions the learner might make regarding the nature of words - in particular, how these assumptions affect the kinds of words that are segmented from a corpus of transcribed child-directed speech. We develop several models within a Bayesian ideal observer framework, and use them to examine the consequences of assuming either that words are independent units, or units that help to predict other units. We show through empirical and theoretical results that the assumption of independence causes the learner to undersegment the corpus, with many two- and three-word sequences (e.g. what’s that, do you, in the house) misidentified as individual words. In contrast, when the learner assumes that words are predictive, the resulting segmentation is far more accurate. These results indicate that taking context into account is important for a statistical word segmentation strategy to be successful, and raise the possibility that even young infants may be able to exploit more subtle statistical patterns than have usually been considered.     

Lexical processing while deciding what task to perform: reading aloud in the context of the task set paradigm

The results of two experiments provide the first direct demonstration that subjects can process a word lexically despite concurrently being engaged in decoding a task cue telling them which of two tasks to perform. These results, taken together with others, point to qualitative differences between the mind‘s ability to engage in lexical versus sublexical processing during the time they are engaged with other tasks. The emerging picture is one in which some form of resource(s) plays little role during lexical processing whereas the need for some form of resource(s) during sublexical processing serves to bottleneck performance.     

Symbol-string sensitivity and children's reading

In this study of primary school children, a novel 'symbol-string' task is used to assess sensitivity to the position of briefly presented non-alphabetic but letter-like symbols. The results demonstrate that sensitivity in the symbol-string task explains a unique proportion of the variability in children's contextual reading accuracy. Moreover, developmental dyslexic readers show reduced sensitivity in this task, compared to chronological age controls. The results suggest that limitations set by visuo-spatial processes and/or attentional iconic memory resources may constrain children's reading accuracy.