A parallel distributed processing model of visual word recognition and pronunciation is described. The model consists of sets of orthographic and phonologlc ~ units and an interlevel of hidden units. Weights on connections between units were modified during a training phase using the back-propa-gation learning algorithm. The model simulates many aspects of human performance, including (a) differences bet~n~.'n words in terms of processing difficulty, (b) pronunciation of novel items, (c) differences between readers in terms of word recognition skill, (d) transitions from beginning to skilled reading, and (e) differences in performance on lexieal decision and naming tasks. The model's behavior early in the learning phase corresponds to that of children acquiring word recognition skills. Training with a smaller number of hidden units produces output characteristic of many dys-lexic readers. Naming is simulated without pronunciation rules, and lexical decisions are simulated without accessing word-level representations. The performance of the model is largely determined by three factors: the nature of the input, a significant fragment of written English; the learning rule, which encodes the implicit structure of the orthography in the weights on connections; and the architecture of the system, which influences the scope of what can be learned. The recognition and pronunciation of words is one of the cen-

We develop a connectionist approach to processing in quasi-regular domains, as exemplified by English word reading. A consideration of the shortcomings of a previous implementation (Seidenberg & McClelland, 1989, Psych. Rev.) in reading nonwords leads to the development of orthographic and phonological representations that capture better the relevant structure among the written and spoken forms of words. In a number of simulation experiments, networks using the new representations learn to read both regular and exception words, including low-frequency exception words, and yet are still able to read pronounceable nonwords as well as skilled readers. A mathematical analysis of the effects of word frequency and spelling-sound consistency in a related but simpler system serves to clarify the close relationship of these factors in influencing naming latencies. These insights are verified in subsequent simulations, including an attractor network that reproduces the naming latency data directly in its time to settle on a response. Further analyses of the network's ability to reproduce data on impaired reading in surface dyslexia support a view of the reading system that incorporates a graded division-of-labor between semantic and phonological processes. Such a view is consistent with the more general Seidenberg and McClelland framework and has some similarities with---but also important differences from---the standard dual-route account.

Ambiguity resolution is a central problem in language comprehension. Lexical and syntactic ambiguities are standardly assumed to involve different types of knowledge representations and be resolved by different mechanisms. An alternative account is provided in which both types of ambiguity derive from aspects of lexical representation and are resolved by the same processing mechanisms. Reinterpreting syntactic ambiguity resolution as a form of lexical ambiguity resolution obviates the need for special parsing principles to account for syntactic interpretation preferences, reconciles a number of apparently conflicting results concerning the roles of lexical and contextual information in sentence processing, explains differences among ambiguities in terms of ease of resolution, and provides a more unified account of language comprehension than was previously available. One of the principal goals for a theory of language compre- third section we consider processing issues: how information is hension is to explain how the reader or listener copes with a processed within the mental lexicon and how contextual inforpervasive ambiguity problem. Languages are structured at mation can influence processing. The central processing mechmultiple levels simultaneously, including lexical, phonological, anism we invoke is the constraint satisfaction process that has morphological, syntactic, and text or discourse levels. At any been realized in interactive-activation models (e.g., Elman &

Abstract not found

In contrast to expectation-based, predictive views of discourse comprehension, a model is developed in which the initial processing is strictly bottom-up. Word meanings are activated, propositions are formed, and inferences and elaborations are produced without regard to the discourse context. However, a network of interrelated items is created in this manner, which can be integrated into a coherent structure through a spreading activation process. Data concerning the time course of word identification in a discourse context are examined. A simulation of arithmetic word-problem under-standing provides a plausible account for some well-known phenomena in this area. Discourse comprehension, from the viewpoint of a computa-tional theory, involves constructing a representation of a dis-course upon which various computations can be performed, the outcomes of which are commonly taken as evidence for com-prehension. Thus, after comprehending a text, one might rea-sonably expect to be able to answer questions about it, recall or summarize it, verify statements about it, paraphrase it, and SO on.

Previous work has shown how a back-propagation network with recurrent connections can successfully model many aspects of human spoken word recogni-tion (Norris, 1988, 1990, 1992, 1993). However, such networks are unable to revise their decisions in the light of subsequent context. TRACE (McClelland & Elman, 1986), on the other hand, manages to deal appropriately with following context, but only by using a highly implausible architecture that fails to account for some important experimental results. A new model is presented which displays the more desirable properties of each of these models. In contrast to TRACE the new model is entirely bottom-up and can readily perform simulations with vocabularies of tens of thousands of words. 1.

Deep dyslexia is an acquired reading disorder marked by the occurrence of semantic errors (e.g., reading RIVER as "ocean"). In addition, patients exhibit a number of other symptoms, including visual and morphological effects in their errors, a part-of-speech effect, and an advantage for concrete over abstract words. Deep dyslexia poses a distinct challenge for cognitive neuropsychology because there is little understanding of why such a variety of symptoms should co-occur in virtually all known patients. Hinton and Shallice (1991) replicated the co-occurrence of visual and semantic errors by lesioning a recurrent connectionist network trained to map from orthography to semantics. While the success of their simulations is encouraging, there is little understanding of what underlying principles are responsible for them. In this paper we evaluate and, where possible, improve on the most important design decisions made by Hinton and Shallice, relating to the task, the network architecture, the training procedure, and the testing procedure. We identify four properties of networks that underly their ability to reproduce the deep dyslexic symptom-complex: distributed orthographic and semantic representations, gradient descent learning, attractors for word meanings, and greater richness of concrete vs. abstract semantics. The first three of these are general connectionist principles and the last is based on earlier theorizing. Taken together, the results demonstrate the usefulness of a connectionist approach to understanding deep dyslexia in particular, and the viability of connectionist neuropsychology in general.

Many theorists assume that the cognitive system is composed of a collection of encapsulated processing components or modules, each dedicated to performing a particular cognitive function. On this view, selective impairments of cognitive tasks following brain damage, as evidenced by double dissociations, are naturally interpreted in terms of the loss of particular processing components. By contrast, the current investigation examines in detail a double dissociation between concrete and abstract word reading after damage to a connectionist network that pronounces words via meaning and yet has no separable components (Plaut & Shallice, 1993). The functional specialization in the network that gives rise to the double dissociation is not transparently related to the network's structure, as modular theories assume. Furthermore, a consideration of the distribution of effects across quantitatively equivalent individual lesions in the network raises specific concerns about the interpretation of...

this article, in conjunction with those developed previously (Plaut et al., 1996; Seidenberg & McClelland, 1989), illustrate how connectionist computational principles---distributed representation, structure-sensitive learning, and interactivity---can provide insight into central empirical phenomena in normal and impaired lexical processing. Moreover, they make it clear that distinctions in the function of the lexical system---as manifest in the behaviour of experimental subjects--- need not re#ect corresponding distinctions in the structure of the system. Thus, networks exhibit word-frequency effects and word/nonword discrimination without word representations, and spelling --sound consistency effects without separate mechanisms for regular and exception items. In this way, gaining insight into the structure and function of the cognitive system by observing its normal and impaired behaviour ---the central goal of cognitive psychology and neuropsycho logy---may depend critically on developing theories and explicit simulations in the context of a speci#c computational framework that relates structure to function

We explore representation of 3D objects in which several distinct 2D views are stored for each object. We demonstrate the ability of a two-layer network of thresholded summation units to support such representations. Using unsupervised Hebbian relaxation, the network learned to recognize ten objects from different viewpoints. The training process led to the emergence of compact representations of the specific input views. When tested on novel views of the same objects, the network exhibited a substantial generalization capa- bility. In simulated psychophysical experiments, the network's behavior was qualitatively similar to that of human subjects.