Category labels induce boundary-dependent perceptual warping in learned speech categories
Swan, Kristen A. and Myers, E. (2013) Category labels induce boundary-dependent perceptual warping in learned speech categories. Second Language Research 29 (4), pp. 391-411. ISSN 0267-6583.
Abstract
Adults tend to perceive speech sounds from their native language as members of distinct and stable categories; however, they fail to perceive differences between many non-native speech sounds without a great deal of training. The present study investigates the effects of categorization training on adults’ ability to discriminate non-native phonetic contrasts. It was hypothesized that only individuals who successfully learned the appropriate categories would show selective improvements in discriminating between-category contrasts. Participants were trained to categorize progressively narrow phonetic contrasts across one of two non-native boundaries, with discrimination pre- and post-tests completed to measure the effects of training on participants’ perceptual sensitivity. Results suggest that changes in adults’ ability to discriminate a non-native contrast depend on their successful learning of the relevant category structure. Furthermore, post-training identification functions show that changes in perceptual categories specifically correspond to their relative placement of the category boundary. Taken together, these results indicate that learning to assign category labels to a non-native speech continuum is sufficient to induce discontinuous perception of between- versus within-category contrasts.
Metadata
Item Type: | Article |
---|---|
Keyword(s) / Subject(s): | categorization, second language acquisition, speech perception |
School: | Birkbeck Faculties and Schools > Faculty of Science > School of Psychological Sciences |
Depositing User: | Administrator |
Date Deposited: | 11 Nov 2013 09:16 |
Last Modified: | 02 Aug 2023 17:08 |
URI: | https://eprints.bbk.ac.uk/id/eprint/8706 |
Statistics
Additional statistics are available via IRStats2.