Are there independent lexical and nonlexical routes in word processing? An evaluation of the dual-route theory of reading

Predictive language processing: integrating comprehension and production, and what atypical populations can tell us

Predictive processing, a crucial aspect of human cognition, is also relevant for language comprehension. In everyday situations, we exploit various sources of information to anticipate and therefore facilitate processing of upcoming linguistic input. In the literature, there are a variety of models that aim at accounting for such ability. One group of models propose a strict relationship between prediction and language production mechanisms. In this review, we first introduce very briefly the concept of predictive processing during language comprehension. Secondly, we focus on models that attribute a prominent role to language production and sensorimotor processing in language prediction ("prediction-by-production" models). Contextually, we provide a summary of studies that investigated the role of speech production and auditory perception on language comprehension/prediction tasks in healthy, typical participants. Then, we provide an overview of the limited existing literature on specific atypical/clinical populations that may represent suitable testing ground for such models-i.e., populations with impaired speech production and auditory perception mechanisms. Ultimately, we suggest a more widely and in-depth testing of prediction-by-production accounts, and the involvement of atypical populations both for model testing and as targets for possible novel speech/language treatment approaches.

Shared and distinct routes in speech and gesture imitation: Evidence from stroke

Dual-route models of high-level (praxis) actions distinguish between an "indirect" semantic route mediating meaningful gesture imitation, and a "direct" sensory-motor route mediates meaningless gesture imitation. Similarly, dual-route language models distinguish between an indirect route mediating production and repetition of words, and a direct route mediating non-word repetition. Although aphasia and limb apraxia frequently co-occur following left-hemisphere cerebrovascular accident (LCVA), it is unclear which aspects of these functional-neuroanatomic dual-route architectures are shared across praxis and language domains. This study focused on gesture imitation to test the hypothesis that semantic information (and portions of the indirect route) are shared across domains, whereas two distinct dorsal routes mediate sensory-motor mapping. Forty chronic LCVA and 17 neurotypical controls completed semantic memory and language tasks and imitated 3 types of gesture stimuli: (1) labeled/"named" meaningful, (2) unnamed meaningful, and (3) meaningless gestures. The comparison of accuracy between meaningless versus unnamed meaningful gestures examined the benefits of semantic information, while the comparison of unnamed meaningful versus named meaningful imitation examined additional benefits of linguistic cueing. Mixed-effects models examined group by task interaction effects on gesture ability. We found that for patients with LCVA, unnamed meaningful gestures were imitated more accurately than meaningless gestures, suggesting that semantic information was beneficial, but there was no benefit of labeling. Reduced benefit of semantic information on gesture accuracy was associated with lesions to inferior frontal and posterior temporal regions as well as semantic memory performance on a pictorial (non-gesture) task. In contrast, there was no relationship between meaningless gesture imitation and nonword repetition, indicating that measures of direct route performance are not associated across language and action. These results provide preliminary evidence that portions of the indirect semantic route are shared across the language and action domains, while two direct sensory-motor mapping routes mediate word repetition and gesture imitation.

The evolution of cognitive models: From neuropsychology to neuroimaging and back

This paper provides a historical and future perspective on how neuropsychology and neuroimaging can be used to develop cognitive models of human brain functions. Section 1 focuses on the emergence of cognitive modelling from neuropsychology, why lesion location was considered to be unimportant and the challenges faced when mapping symptoms to impaired cognitive processes. Section 2 describes how established cognitive models based on behavioural data alone cannot explain the complex patterns of distributed brain activity that are observed in functional neuroimaging studies. This has led to proposals for new cognitive processes, new cognitive strategies and new functional ontologies for cognition. Section 3 considers how the integration of data from lesion, behavioural and functional neuroimaging studies of large cohorts of brain damaged patients can be used to determine whether inter-patient variability in behaviour is due to differences in the premorbid function of each brain region, lesion site or cognitive strategy. This combination of neuroimaging and neuropsychology is providing a deeper understanding of how cognitive functions can be lost and re-learnt after brain damage - an understanding that will transform our ability to generate and validate cognitive models that are both physiologically plausible and clinically useful.

Orthographic Analogies and Reading Development

Given the interest in the use of orthographic analogies in skilled reading, the role of analogies in reading development has received surprisingly little attention. The experiments presented here examine three important developmental issues: whether beginning readers can make orthographic analogies, how the consistency of spelling–sound relations affects this ability, and whether orthographic analogies are used in reading prose. It is concluded that orthographic analogies have an important role to play in reading development, and some suggestions are offered as to why this may be so.

Action-feature integration blinds to feature-overlapping perceptual events: Evidence from manual and vocal actions

Previous studies showed that the identification of a left- or right-pointing arrowhead is impaired when it appears while planning and executing a spatially compatible left or right keypress (Müsseler & Hommel, 1997a). We attribute this effect to stimulus processing and action control operating on the same feature codes so that, once a code is integrated in an action plan, it is less available for perceptual processing. In three pairs of experiments we tested the generality of this account by using stimulus–response combinations other than arrows and manual keypresses. Planning manual left–right keypressing actions impaired the identification of spatially corresponding arrows but not of words with congruent meaning. On the contrary, planning to say “left” or “right” impaired the identification of corresponding spatial words but not of congruent arrows. Thus, as the feature-integration approach suggests, stimulus identification is impaired only with overlap of perceptual or perceptually derived stimulus and response features while mere semantic congruence is insufficient.

Graphic and Phonological Processing in Chinese Character Identification

A backward-masking procedure was used to examine the effect of graphic, phonological, and graphic-and-phonological information on Chinese character identification. Twenty-two Mandarin-speaking Taiwanese graduate students were asked to write down lists of paired characters presented sequentially in a tachistoscope. Analysis of variance performed on the target identification accuracy scores indicated that graphic information plays an essential role in Chinese character identification. Within the same time frame, phonological information, whether activated alone or in conjunction with graphic information, does not enhance the accuracy of identification. The present findings are discussed in relation to those of Perfetti and Zhang's (1991) Chinese character study and Perfetti, Bell, and Delaney's (1988) English word study.

Evidence for Superior Orthographic Skills in Dyslexics

The purpose of this study was to assess the development of both phonological and orthographic skills in normally achieving and dyslexic readers The subjects were 257 dyslexic and 342 normally achieving readers, matched at eight reading levels They were administered the Woodcock (1987) Word Attack Subtest, a measure of phonological skills requiring the reading of pseudowords, and an orthographic awareness task designed to measure awareness of the properties of English words and the probable sequences and positions of letters within words The dyslexics had significantly higher scores than the normally achieving readers on the orthographic awareness task However, the normally achieving readers had significantly higher scores on the Word Attack Subtest Therefore, the difficulties with phonological processing and the increased orthographic awareness of the dyslexics may indicate a reading strategy that relies more on the visual than the phonological features of words

Connectionist models of neuropsychological disorders

Artificial neural networks ('connectionist models') embody aspects of real neuronal systems. But does studying the breakdown of performance in such models help us to understand cognitive impairments in humans following brain damage? Here we review recent attempts to capture different neuropsychological disorders using connectionist models with simulated lesions. We show how such lesion studies can be used to evaluate some of the standard assumptions made in neuropsychological research, concerning both double dissociations and associations between patterns of impairment. We also illustrate how lesioned models, like humans, can sometimes be more impaired on the easier of two tasks and demonstrate that connectionist models can incorporate forms of internal structure. Finally we discuss the utility of the models for understanding and predicting the effectiveness of different rehabilitation strategies. Future questions concern the role and possible development of internal structure within these models, whether the models can be generalized to larger-scale simulations, and whether they can accommodate higher-order linguistic disorders.

Neuropsychological inference with an interactive brain: A critique of the “locality” assumption

When cognitive neuropsychologists make inferences about the functional architecture of the normal mind from selective cognitive impairments they generally assume that the effects of brain damage are local, that is, that the nondamaged components of the architecture continue to function as they did before the damage. This assumption follows from the view that the components of the functional architecture are modular, in the sense of being informationally encapsulated. In this target article it is argued that this “locality” assumption is probably not correct in general. Inferences about the functional architecture can nevertheless be made from neuropsychological data with an alternative set of assumptions, according to which human information processing is graded, distributed, and interactive. These claims are supported by three examples of neuropsychological dissociations and a comparison of the inferences obtained from these impairments with and without the locality assumption. The three dissociations are: selective impairments in knowledge of living things, disengagment of visual attention, and overt face recognition. In all three cases, the neuropsychological phenomena lead to more plausible inferences about the normal functional architecture when the locality assumption is abandoned. Also discussed are the relations between the locality assumption in neuropsychology and broader issues, including Fodor's modularity hypothesis and the choice between top-down and bottom-up research approaches.

On determining what is unconscious and what is perception

In sum, dichotic listening tasks are appropriate for testing whether perception requires attention. The separate issue of the possible presence of perception that can never be brought under the spotlight of awareness should indeed be investigated by paradigms such as masking. However, the ultimate criterion for availability to awareness must be phenomenal experience. The discrepancy between thresholds of different perceptual indices is an important empirical finding, but its theoretical interpretation is not straightforward. In addition, it is suggested that we worry about the possibility that so-called indirect evidence reflects side effects of perceptual processing rather than the contents of its final product. In that case, what are being observed are vestiges of the processing of stimuli that do not make it to awareness. Whether those stimuli are below the threshold for overt response is an open question.

Semantic activation without conscious identification in dichotic listening, parafoveal vision, and visual masking: A survey and appraisal

When the stored representation of the meaning of a stimulus is accessed through the processing of a sensory input it is maintained in an activated state for a certain amount of time that allows for further processing. This semantic activation is generally accompanied by conscious identification, which can be demonstrated by the ability of a person to perform discriminations on the basis of the meaning of the stimulus. The idea that a sensory input can give rise to semantic activation without concomitant conscious identification was the central thesis of the controversial research in subliminal perception. Recently, new claims for the existence of such phenomena have arisen from studies in dichotic listening, parafoveal vision, and visual pattern masking. Because of the fundamental role played by these types of experiments in cognitive psychology, the new assertions have raised widespread interest.

The purpose of this paper is to show that this enthusiasm may be premature. Analysis of the three new lines of evidence for semantic activation without conscious identification leads to the following conclusions. (1) Dichotic listening cannot provide the conditions needed to demonstrate the phenomenon. These conditions are better fulfilled in parafoveal vision and are realized ideally in pattern masking. (2) Evidence for the phenomenon is very scanty for parafoveal vision, but several tentative demonstrations have been reported for pattern masking. It can be shown, however, that none of these studies has included the requisite controls to ensure that semantic activation was not accompanied by conscious identification of the stimulus at the time of presentation. (3) On the basis of current evidence it is most likely that these stimuli were indeed consciously identified.

Individual cognitive training of reading disability improves word identification and sentence comprehension in adults with mild mental retardation

Reading therapy has been shown to be effective in treating reading disabilities (RD) in dyslexic children, but little is known of its use in subjects with mild mental retardation (MR). Twenty adult volunteers, with both RD and mild MR, underwent 60 consecutive weeks in a cognitive remediation program, and were compared with 32 untreated control subjects. The experimental group showed a significant improvement in word identification, as measured by oral production (p=0.0004) or silent reading (p=0.023), and sentence comprehension (p=0.0002). Adults with MR appear to benefit from new approaches in the field of RD.

Cross-task strategic effects

When easy and difficult items are mixed together, their reading aloud latencies become more homogeneous relative to their presentation in unmixed ("pure") conditions (Lupker, Brown, & Colombo, 1997). We report two experiments designed to investigate the nature of the mechanism that underlies this list composition, or blocking, effect. In Experiment 1, we replicated Lupker et al.'s (1997) blocking effect in the reading aloud task and extended these findings to the visual lexical decision task. In Experiment 2, we found that blocking effects generalized across tasks: The characteristics of stimuli in a visual lexical decision task influenced reading aloud latencies, and vice versa, when visual lexical decision and reading aloud trials were presented alternately in the same experiment. We discuss implications of these results within time-criterion (Lupker et al., 1997) and strength-of-processing (Kello & Plaut, 2000, 2003) theories of strategic processing in reading.

Lexical access and phonological decoding in adult dyslexic subtypes

Lexical access and phonological decoding were tested in 100 normal adult readers and 21 adult dyslexic individuals. Within the dyslexic sample, 11 dysphonetic dyslexic and 10 dyseidetic dyslexic participants were classified on the basis of spelling patterns. In the 1st experiment, adult dyseidetic readers showed a marked deficit on the lexical-access decision task in comparison with adult dysphonetic readers. In the 2nd experiment, the phonological-decoding decision task did not separate the subtypes. A lexical-access deficit in adult dyseidetic dyslexia cannot be explained in terms of a developmental delay. A phonological-decoding deficit in adult dyseidetic dyslexia may be explained by increased involvement of the lexical procedure in phonological assembly under an analogy strategy.

Spatiotemporal dynamics of modality-specific and supramodal word processing

The ability of written and spoken words to access the same semantic meaning provides a test case for the multimodal convergence of information from sensory to associative areas. Using anatomically constrained magnetoencephalography (aMEG), the present study investigated the stages of word comprehension in real time in the auditory and visual modalities, as subjects participated in a semantic judgment task. Activity spread from the primary sensory areas along the respective ventral processing streams and converged in anterior temporal and inferior prefrontal regions, primarily on the left at around 400 ms. Comparison of response patterns during repetition priming between the two modalities suggest that they are initiated by modality-specific memory systems, but that they are eventually elaborated mainly in supramodal areas.

Spatiotemporal maps of past-tense verb inflection

Does the brain inflect verbs by applying rules, by associative retrieval of the inflected form, or both? We used whole-head magnetoencephalography to spatiotemporally map the brain response underlying verb past-tense inflection. Placing either regular or irregular verbs into the past tense sequentially modulates the bilateral visual, left inferotemporal, posterior superior temporal (Wernicke's area), left inferior prefrontal (Broca's area), and right prefrontal cortices. Although irregular and regular verb inflection evokes similar cortical response patterns, differences in specific frontotemporal regions are observed. At approximately 340 ms, irregular verbs evoke greater response modulation in left occipitotemporal cortex. This modulation occurs when widespread areas are simultaneously active, suggesting that it reflects associative activation necessary for generation of past-tense forms. Subsequently, regular verbs show increased response at approximately 470 ms within left inferior prefrontal regions associated with rule-based inflection. Increased right dorsolateral prefrontal response at approximately 570 ms may represent directed/effortful retrieval of irregular past-tense forms. Thus, the brain inflects verbs by dynamically modulating different functional divisions of an integrated language system.