Perception of Japanese temporally-cued phonetic contrasts by Japanese and American English listeners: Behavioral and electrophysiological measures.
Item
-
Title
-
Perception of Japanese temporally-cued phonetic contrasts by Japanese and American English listeners: Behavioral and electrophysiological measures.
-
Identifier
-
AAI3283197
-
identifier
-
3283197
-
Creator
-
Hisagi, Miwako.
-
Contributor
-
Advisers: Valerie L. Shafer | Winifred Strange
-
Date
-
2007
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Speech Communication | Psychology, Cognitive
-
Abstract
-
This study examined American listeners' perception of Japanese temporally-cued contrasts of vowels (kado vs. kaa do) and consonants (nifi vs. niffi). Japanese (JP) phonology includes distinctive contrasts of consonant and vowel "length" that are cued almost exclusively by duration differences at the word level. American English (AE) phonology, on the other hand, uses such temporal cues only as secondary phonetic features. Previous cross-language behavioral studies of temporally-cued contrasts produced mixed results, in part due to choice of stimuli and tasks. The present study examines the question by considering the implications of first-language (L1) perception as an "over-learned" automatic process of selective perception in adults. The notion of "highly over-learned processes" interfering with second-language (L2) speech perception can be related to models of L1 acquisition in which infants are claimed to learn to automatically focus on language-specific features of speech sounds. In this study, the event-related potential component, Mismatch Negativity (MMN), was used to index "pre-attentive" discrimination of temporally-cued vowel and consonant contrasts in a categorial oddball task by two group of listeners: AE learners with no experience in learning JP and native JP controls. The role of attention was examined in two experiments: Exp 1 Auditory-Attend, in which listeners attended to the auditory input (implicitly counted target deviants), and Exp 2 Visual-Attend, in which listeners attended to (implicitly counted) deviants in a simultaneously presented visual categorial oddball shape discrimination task. Follow-up behavioral tests indicated that AE listeners had more difficulty on the consonant contrast. Results showed effects of attentional focus for both language groups. For the easier vowel contrast, MMN amplitudes were smaller for AE listeners in the Visual-Attend experiment than JP controls and AE listeners in the Auditory-Attend experiment. For the more difficult consonant contrast, AE listeners produced very small MMNs in both attention experiments, while JP listeners produced larger MMNs in the Auditory-Attend experiment. This pattern of results supports the conclusion that native listeners show more robust indices of "pre-attentive" discrimination than do non-native listeners on both vowel and consonant contrasts; however, attentional resources may be required to process difficult consonant contrasts even in a listener's native language.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
