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.

What is the neural representation of a speech code as it evolves in time? A neural model simulates data concerning segregation and integration of phonetic percepts. Hearing two phonetically related stops in a VC-CV pair (V = vowel; C = consonant) requires 150 ms more closure time than hearing two phonetically different stops in a VC,-C2V pair. Closure time also varies with long-term stimulus rate. The model simulates rate-dependent category boundaries that emerge from feedback: interactions between a working memory for short-term storage of phonetic items and a list categorization network for grouping sequences of items. The conscious speech code is a resonant wave. It emerges after bottom-up signals from the working memory select list chunks which read out top-down expectations that amplify and focus attention on consistent working memory items. In VCi-C2V pairs, resonance is reset by mismatch of Cj with the C, expectation. In VC-CV pairs, resonance prolongs a repeated C. What is the nature of the process that converts brain events into behavioral percepts? An answer to this question is needed in order to understand how the brain controls behavior and how the brain is, in turn, shaped by environmental feedback that is experienced on the behavioral level. The nature of this connection also needs to be understood in order to develop neurally plausible connectionist models. Without it, a correct linking hypothesis cannot be developed between psychological data and the brain mechanisms from which they are generated.

This paper describes and motivates six principles for computational cognitive neuroscience models: biological realism, distributed representations, inhibitory competition, bidirectional activation propagation, errordriven task learning, and Hebbian model learning. Although these principles are supported by a number of cognitive, computational, and biological motivations, the prototypical neural network model (a feedforward backpropagation network) incorporates only two of them, and no widely used model incorporates all of them. This paper argues that these principles should be integrated into a coherent overall framework, and discusses some potential synergies and conflicts in doing so.

The effects of context on perceptual identification responses given without tinle pressure are well-described by classical models in which contextual and stimulus information exert independent effects. A recent Ilrticle by Massaro (1989) raises the possibility that interactive models, such as the TRACE model of speech perception, are inherently incompatible with these classical context effects. The present article shows that this incompatibility hypothesis can be rejected. Mathematical analysis and computer simulation methods are used to show that interactive models can exhibit the classical effects of context, if there is variability in the input to the network or if there is intrinsic variability in the network itself. A variety of interactive models which incorporate variability can aU produce the classical context effects, at least under some conditions; the conditions are rather general in the case of one of the variants. The findings suggest that interactive models should not be viewed as alternatives to classical accounts, but as hypotheses about the dynamics of information processing that lead to the global asymptotic behavior that the classical models describe.

This study demonstrates that listeners use lexical knowledge in perceptual learning of speech sounds. Dutch listeners first made lexical decisions on Dutch words and nonwords. The final fricative of 20 critical words had been replaced by an ambiguous sound, between [f] and [s]. One group of listeners heard ambiguous [f]-final words (e.g., [WI WItlo?], from witlof, chicory) and unambiguous [s]-final words (e.g., naaldbos, pine forest). Another group heard the reverse (e.g., ambiguous [na:ldbo?], unambiguous witlof). Listeners who had heard [?] in [f]-final words were subsequently more likely to categorize ambiguous sounds on an [f]–[s] continuum as [f] than those who heard [?] in [s]-final words. Control conditions ruled out alternative explanations based on selective adaptation and contrast. Lexical information can thus be used to train categorization of speech. This use of lexical information differs from the on-line lexical feedback embodied in interactive models of speech perception. In contrast to online feedback, lexical feedback for learning is of benefit to spoken word recognition (e.g., in

Almost all models of simple and choice response time (RT) employ a stochastic (i.e., variable within trial) accumulation decision process. In order to account for the relationship between correct and error choice RT, it has been found necessary to also include between trial variability in the starting point and/or the rate of accumulation, both in linear (Ratcliff & Rouder, 1998) and nonlinear (Usher & McClelland, 2001) stochastic models. We show that a ballistic (i.e., deterministic within trial) model using a simplified version of Usher and McClellands nonlinear accumulation process, and assuming only between trial variability in the rate and starting point of accumulation, is not only capable of accounting for the relationship between error and correct RT, but can also model other benchmark behavioural phenomena, such as RT distribution and speed-accuracy trade off. We successfully fit our ballistic model to Ratcliff and Rouders data, which exhibit many of the benchmark phenomena. Even for fast and easy decisions, a simple summation of sensory and motor transduction delays and conduction times in the nervous system cannot account for the duration and variability of reaction times. (Hanes & Schall, 1996, p.427). The slowness and variability of response time (RT) has been almost universally explained by decision processes involving stochastic accumulation of information. Stochastic models assume that the accumulated information varies randomly from moment to moment during the decision process. RT is relatively slow because a criterion amount of information must be accumulated before a response is made, and RT ...

this paper and the writing of the paper were supported, in part, by grants to Dominic W. Massaro from the Public Health Service (PHS R01 NS 20314), the National Science Foundation (BNS 8812728), a James McKeen Cattell Fellowship, and the graduate division of the University of California, Santa Cruz. Cohen & Massaro On the Similarity of Categorization Models 2

The computer metaphor of cognitivism that has had such a strong influence on cognitive science over recent decades seems to be confronted (again) by a competitor: the brain metaphor put forward by connectionism (e.g. [McClelland and Rumelhart 1986] and [Sejnowski et al. 1988]). Connectionism assumes that mental phenomena can be explained in terms of the parallel activation and interaction of a large number of units (model neurons). These units are linked by connections (artificial synapses) which modulate the transmitted activity. Knowledge is represented in these connections between the units and learning takes place by adjusting their strength. An important, and much emphasized, aspect of connectionist models is their emergent behaviour. The massive parallel interaction of a large number of simple units can lead to qualitatively different and more interesting forms of behaviour. Successes of this connectionist approach range from models of human

The influence of semantic context on word identification was examined using masked target displays. Related prime words enhanced a signal detection measure of sensitivity in making lexical decisions and in determining whether a probe word matched the target word. When line drawings were used as primes, a similar benefit was obtained with the probe task. Although these results suggest that contextual information affects perceptual encoding, this conclusion is questioned on the grounds that sensitivity in these tasks may be determined by independent contributions of perceptual and contextual information. The plausibility of this view is supported by a simulation of the experiments using a connectionist model in which perceptual and semantic information make independent contributions to word identification. The model also predicts results with two other analytic methods that have been used to argue for priming effects on perceptual encoding. The identification of objects or events in the environment, including written and spoken language, benefits from the availability of conceptually relevant information provided by the context in which an object or event occurs. For example, the identification of visually presented individual letters is enhanced when they are placed in the context of a familiar word (see, e.g., McClelland & Rumelhart,

Abstract: A naturalistic scheme of primitive conceptual representations is proposed using the statistical measure of mutual information. It is argued that a concept represents, not the class of objects that caused its tokening, but the class of objects that is most likely to have caused it (had it been tokened), as specified by the statistical measure of mutual information. This solves the problem of misrepresentation which plagues causal accounts, by taking the representation relation to be determined via ordinal relationships between conditional probabilities. The scheme can deal with statistical biases and does not rely on arbitrary criteria. Implications for the theory of meaning and semantic content are addressed. 1.