The overall objective of the proposed research is to increase our understanding of spoken language comprehension. The research focuses specifically on the very early stages of comprehension, during which the listener analyzes the speech signal so as to recover the prelexical linguistic structure of the utterance the sequences of phonetic segments (consonants and vowels) that comprise the lexical items of a language. The fundamental issue addressed is how listeners internally represen the phonetic categories of their language, and how they map the incoming speec signal onto theses categorical representations during processing. Over the years considerable emphasis has been placed on the abstract nature of the categorical representations of speech. Indeed, a widely held assumption in the literature has been that during the course of processing listeners derive an abstract phonetic representation and, in the course of doing so, discard information about the fine grained detail of the speech signal. However, newer research is showing that the representations of speech are much richer than this emphasis on abstract entities would suggest, and that listeners retain in memory of substantial amount of fine grained acoustic phonetic information. On line of evidence for the richness of phonetic representations comes from research showing that members of a given phonetic category are not perceptuall equivalent, but rather that phonetic categories are internally structures in a graded fashion, with some members perceived as better exemplars than others. Our recent research has begun to explicate the nature of this graded internal category structure and its role in processing, and we propose a series of multifaceted experiments to continue this line of investigation. The proposed research is centered on four specific aims: (1) Investigate the constraints on how context shapes the graded internal structure of phonetic categories; (2) Investigate the role of specific language experience in shaping this structure (3) Investigate the role of talker differences in shaping this structure; (4) Investigate the role of graded internal category structure in online speech processing.
