Doctoral Dissertation Research: The specificity and abstractness of allophonic variability and its effects on speech production and perception

University Of California-Los Angeles, Los Angeles CA

Investigators

Abstract

Humans have a remarkable ability to perceive spoken language. If one asks several different speakers to produce a given word, records their utterances, and then inspects the sound recording for each speaker, one will find that the phonetic signature (e.g., speech rate, frequency components, etc) varies greatly between speakers. In spite of this, most listeners have little trouble discerning which word was spoken, regardless of the speaker's age, gender, or familiarity to the listener. With support from the National Science Foundation, Dr. Patricia Keating and Ms. Kuniko Nielson will conduct research aimed at understanding how speech sounds are structured and represented in the human mind. Traditional accounts of speech perception assume mental representation of speech sounds to the abstract and invariant. On the other hand, recent studies suggest that traces of episodic memory are used in both speech perception and production; thus the mental representation of speech sounds is more plastic and variant than previously considered. In order to examine how short-term exposure to speech affects speech perception and production, two types of behavioral experiments will be conducted. One type of experiment investigates the degree of spontaneous phonetic imitation under different conditions, and its generalizability to novel stimuli; the other examines the relation between speech perception and production by comparing the imitation effect and perceptual adaptation in a single experiment. The intellectual merit of this project is that it will apply knowledge from linguistic phonetics to a current issue in psychology, by looking at both speech production and perception. Further, it will clarify how people's linguistic knowledge constrains the role of episodic memory in their use of spoken language. The broader impacts of the research include providing additional knowledge of theories of perception in general, and contributing to the field of language education (both first and second language) by demonstrating how experienced speech affects one's mental representation.