Abstract not found

Listeners are often active in conversation, and the feedback they provide speakers can improve the communication. To examine how feedback influences conversation, we had 76 speaker subjects watch a movie and then summarize it to one or two listeners. The- listeners provided varying amounts of feedback to the speaker. When two listeners were present, one could influence the speaker through feedback and the other could only eavesdrop on the conversation. When speakers received more feedback, their narratives were more comprehensible; that is, both listeners understood the movie better. In addition, feedback individuated communication; that is, the listener who provided the feedback understood the movie better than the eavesdropper who listened to the same conversation. In part, feedback produced these effects by coordinating what the speaker said with what the listener needed to know. Listener feedback signaled listeners ' prior knowledge of the movie, and speakers talked most efficiently about those sections of the movie about which listeners had prior knowledge. Traditional research on communication has often proceeded as if the variables affecting it were static and nonemergent, in the sense that most variables do not change during the course of the conversation and that changes that do occur are produced by something other than dynamic interaction between the participants. For example, work on attitude change has studied static characteristics of the source, message, and recipient (McGuire, 1969). Work in nonverbal communication, with a few notable exceptions (e.g., Goffman, 1971), also typically focuses on these characteristics by looking at personality differences (e.g., Rosenthal, Hall,

Abstract not found

Exemplar, prototype, and connectionist models typically assume that events constitute the basic unit of learning and representation in categorization. In these models, each learning event updates a statistical representation of a category independently of other learning events. An implication is that events involving the same individual affect learning independently and are not integrated into a single structure that represents the individual in an internal model of the world. A series of experiments demonstrates that human subjects track individuals across events, establish representations of them, and use these representations in categorization. These findings are consistent with ‘‘representationalism,’ ’ the view that an internal model of the world constitutes a physical level of representation in the brain, and that the brain does not simply capture the statistical properties of events in an undifferentiated dynamical system. Although categorization is an inherently statistical process that produces generalization, pattern completion, frequency effects, and adaptive learning, it is also an inherently representational process that establishes an internal model of the world. As a result, representational structures evolve in memory to track the histories of individuals, accumulate information about them, and simulate

Practical text processing systems need wide covering grammars. When parsing unrestricted language, such grammars often fail to generate all of the sentences that humans would judge to be grammatical. This problem undermines successful parsing of the text and is known as undergeneration. There are two main ways of dealing with undergeneration: either by sentence correction, or by grammar correction. This thesis concentrates upon automatic grammar correction (or machine learning of grammar) as a solution to the problem of undergeneration. Broadly speaking, grammar correction approaches can be classified as being either datadriven, or model-based. Data-driven learners use data-intensive methods to acquire grammar. They typically use grammar formalisms unsuited to the needs of practical text processing and cannot guarantee that the resulting grammar is adequate for subsequent semantic interpretation. That is, data-driven learners acquire grammars that generate strings that humans would jud...

computation. This paper argues that such a characterization is correct, and that fundamental properties of grammar can and should be understood in terms of restrictions on the complexity of possible grammatical computation, when defined in terms of generative capacity. More specifically, the paper demonstrates that the computational restrictiveness imposed by Tree Adjoining Grammar provides important insights into the nature of human grammatical knowledge. 2 1

of Education under Contract No. MS-NIE-C-400-76-0116.

Syntactic processes are an important component of the language processing system. While semantic and pragmatic information has been shown to influence the eventual interpretation of an utterance, psycholinguistic theories have not come to an agreement on how this information is combined with syntactic knowledge during the initial parsing process. This paper reviews recent psycholinguistic research on the effects of lexical information and discourse context on syntactic processing. The results of these studies are often contradictory, and do not allow to draw firm conclusions on how syntactic and non-syntactic processes interact. Nevertheless, it is suggested that the psychological evidence points towards a weakly interactive, parallel model of parsing. Methodological implications and further research directions are discussed, and an evaluation of text comprehension theories with respect to syntactic processing is attempted.

Language is the crowning achievement of the human species, and it is something that all normal humans can do. The average man is neither a Shakespeare nor a Caravaggio, but he is capable of fluent speech, even if he cannot paint at all. In fact, the average speaker produces approximately 150 words per minute, each word chosen from somewhere between 20,000 and 40,000 alternatives, at error rates below 0.1%. The average child is already well on her way toward that remarkable level of performance by 5 years of age, with a vocabulary of more than 6000 words and productive control over almost every aspect of sound and grammar in her language. Given the magnitude of this achievement, and the speed with which we attain it, some theorists have

Devitt (2006) makes a sustained critique of Chomskyan linguistics, articulating persistent complaints about the “psychological reality ” of generative grammars. I suggest these complaints are merely terminological and that Devitt fails to appreciate the status of Chomsky’s computational formalisms found elsewhere in cognitive science. Devitt ascribes an intentional conception of representations that Chomsky repudiates and that is independently implausible. I argue that Devitt’s proposed alternative “linguistic reality ” constituted by physical symbol tokens neglects the problems of tokens as opposed to types and he misses the force of Chomsky’s case against Behaviourism and nominalism. I suggest that Devitt’s case against intuitions as data misunderstands their standard role throughout perceptual psychology. I argue that Devitt’s position exemplifies pervasive errors concerning mental representation seen throughout cognitive science.