Language is one of the most fundamental human cognitive abilities. It plays an important role in normal development, and is the major means for acquiring information in many domains. A number of psychological disorders (e.g., schizophrenia, autism) can disrupt the normally impressive functioning of this system, greatly exacerbating the negative consequences of these disorders. Psycholinguists have made significant progress in specifying the structures and processes that underlie language comprehension. Over the course of the last two decades, our laboratory has learned a great deal about how words are represented in a person’s “mental lexicon”. Our research has begun to clarify how the presentation of a spoken word leads to a particular lexical representation becoming activated, and what the effects of such activation are (e.g., what effect one active lexical representation has on others, and on units at other levels of representation). The current proposal continues our investigation of spoken language processing, with a focus on the dynamic properties of lexical and sublexical systems: Through perceptual learning, there is continuous redevelopment of the sublexical representations that are used to access the lexical entries. The proposed project includes a large set of theoretically-driven empirical studies of sublexical perceptual learning. The experiments use a range of different methodologies, in order to assure correct theoretical inferences through converging operations. The set of studies on perceptual learning will provide a detailed understanding of how the perceptual system restructures its sublexical representations. Our work to date in this area has shown that the system is appropriately conservative about such restructuring, only changing the representations to accommodate phonetic variation when the variation is characteristic of a speaker (the Conservative Adjustment/Restructuring Principle, or “CA/RP”). The proposed experiments will track the representational shift, across time, across number of exposures to a variant, and across the phonetic details of the variants. The product of the proposed research will be a much better understanding of the architecture of the system that accomplishes language comprehension. These studies will provide a much more detailed picture of how lexical and sublexical representations change over both the short term, and the longer term. Such an understanding is critical to our understanding language processing. Because language is such a fundamental cognitive ability, progress in describing language processing will enhance our understanding of human cognition, under both normal and disordered conditions.

Language is one of the most fundamental human cognitive abilities. It plays an important role in normal development, and is the major means for acquiring information in many domains. A number of psychological disorders (e.g., schizophrenia, autism) can disrupt the normally impressive functioning of this system, greatly exacerbating the negative consequences of these disorders. The proposed research will provide important information about how variation in a person’s language environment leads to changes in the representations of speech sounds. In previous studies, these changes have been impressively stable, indicating that the learning can last for a very long time. This durability suggests that presenting listeners with an appropriate pattern of speech variation could be used to reshape phonetic category boundaries that have been negatively impacted by partial hearing loss, or by stroke. Given the importance of communication to normal functioning, any such remediation would have significant public health implications. The kinds of phonetic variation used in our studies are similar to variations that are present in the accents of individuals for whom English is a second language. Given the increasingly multilingual society that we live in, understanding how such variation affects speech perception, and how it may affect the representation of language, will be increasingly important in assuring normal language development. The proposed research will provide a much more detailed understanding of these factors than is currently available.