Imitation-A Tool for Studying Speech Perception

Haskins Laboratories, Inc., New Haven CT

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Despite the idealization of much linguistic analysis, speakers of the same language speak alike only roughly. There is dialectal and idiolectal variation, stylistic variation, rate variation and more. Recent linguistic and psycholinguistic research shows clearly that this variation is not treated as noise by listener/speakers. In this research proposal, we are interested in one use that variation in talk can serve: It permits language users to sound more and less like one another. That is, it permits imitation, where imitation serves a useful role, for example, to mark group affiliation or cooperatively. We have five specific aims, all relating to linguistic variation and imitation. First, we propose to test whether imitation occurs in cooperative joint activities but not in less cooperative, more competitive, settings. Second, we propose to look at effects of imitation over the longer term of language change, that is periods spanning months and periods spanning years. Our third aim is to test the role of familiarity on perception and imitation of speech in other languages and dialects. Earlier research has shown that imitation is shaped by relevant exemplar memories (Goldinger, 1998). Accordingly, the familiarity of an utterance affects its identifiably, for example. Our fourth aim is to find evidence for or against one possible mechanism underlying imitation: the motor-perception link proposed as the motor theory of speech perception (e.g., Liberman & Mattingly, 1985). Finally, our fifth aim is to develop both computational and behavioral models of the roles of variability and imitation in speech production development, in development of phonological inventories, and in development of language itself.