Can Australian English listeners learn non-native vowels via distributional learning?

Jia Hoong Ong, Josephine Terry, Paola Escudero

Research output: Chapter in Book / Conference PaperConference Paperpeer-review

Abstract

Inconsistent findings have been reported for distributional learning of vowels, possibly due to interference from learners' native phonological (L1) categories. Native Australian-English (AusE) listeners were exposed to unimodal and bimodal distributions of a continuum spanning Dutch /É‘/-/aË/, which is perceived moderately well by AusE listeners. Despite sustaining learners' attention during the training phase (c.f. passive training), the distribution groups did not differ in their pre-post vowel discrimination, suggesting a lack of distributional learning. Our results imply that learners do not benefit from such rapid learning of contrasts that are perceived with high accuracy due to learners' L1 categories.
Original languageEnglish
Title of host publicationProceedings of the Sixteenth Australasian International Conference on Speech Science and Technology, 6-9 December 2016, Parramatta, Australia
PublisherAustralasian Speech Science & Technology Association
Pages289-292
Number of pages4
Publication statusPublished - 2016
EventAustralasian International Conference on Speech Science and Technology -
Duration: 6 Dec 2016 → …

Publication series

Name
ISSN (Print)2207-1296

Conference

ConferenceAustralasian International Conference on Speech Science and Technology
Period6/12/16 → …

Keywords

  • vowels
  • second language acquisition
  • speech perception
  • Dutch language

Fingerprint

Dive into the research topics of 'Can Australian English listeners learn non-native vowels via distributional learning?'. Together they form a unique fingerprint.

Cite this