Electronic Thesis and Dissertation Repository

Degree

Doctor of Philosophy

Program

Neuroscience

Supervisor

Dr. Marc Joanisse

Abstract

To date, theories of how humans recognize spoken words have yet to account for tonal languages such as Mandarin Chinese. One reason for this is that we know relatively little about how native speakers of tonal languages process spoken words in the brain. This dissertation addresses this problem by examining Mandarin spoken word processing in both adult native speakers and typically developing children. In adults, functional magnetic resonance imaging (fMRI) was used to assess the extent to which the brain regions involved in processing tonal information are distinct from those involved in vowel processing (Chapter 2), while event related potentials (ERPs) were used to investigate responses to different types of phonological competition in Mandarin (Chapter 3). In the fMRI experiment, subjects performed a passive listening task, in which they heard trains of Mandarin syllables consisting of a repeated standard followed by a deviant stimulus differing from the standard in either tone or vowels. Analyses revealed that the regions involved in processing tonal versus vowel deviants were not entirely overlapping. In the ERP experiment, subjects were presented with pictures of items and subsequently heard words that either matched or mismatched the pictures. Mismatches differed from expectations in different components of the Mandarin syllable, and analyses focused on ERP components associated with various stages of spoken word processing. A key finding from this study was that different cognitive processes underlie tonal versus phonemic (vowel) access. Chapter 4 extended these findings by studying the development of Mandarin spoken word processing. In this experiment, a group of typically developing children completed the same ERP picture-word matching task as a group of adults. It was observed that the children differed from the adults primarily in their responses to rhyme mismatches. On the basis of relevant findings, Chapter 5 puts forward several recommendations as to how current theories of spoken word recognition could be modified to account for tonal languages. This proposed theory is then used to explain the observed data from the adults and typically developing children. The dissertation then closes with a brief consideration of future research directions.


Share

COinS