In:
The Journal of the Acoustical Society of America, Acoustical Society of America (ASA), Vol. 125, No. 4_Supplement ( 2009-04-01), p. 2757-2757
Abstract:
This study examines differences between confusions found in productions and perceptions of learners of English. Twenty Korean EFL learners engaged in three tasks involving obstruents placed in different prosodic positions: (a) identification of native English productions, (b) reading from orthographic prompts, and (c) mimicry of native English productions. Recordings of reading and mimicry were presented to 50 native English listeners for identification. This paper compares patterns of errors found for 10 intervocalic obstruents before and after a stress, since previous studies showed that Korean does not exhibit stress-induced differences in consonant allophones. Similarity estimates using Luce’s similarity choice model were regressed across the two intervocalic positions. We found robust correlations despite allophonic differences in English, suggesting a component of L1 transfer in all three tasks. Examining bias parameters, however, revealed systematic differences in the direction of the resulting errors, which is task-dependent. Identification and mimicry tended to underestimate allophonic shifts due to stress, and so to create more voiceless to voiced errors in poststress environments. Reading productions exhibited error directions in exactly the opposite directions, suggesting Korean learners produced the stress but not the corresponding allophonic variations. These patterns indicate very different error outcomes in production and perception.
Type of Medium:
Online Resource
ISSN:
0001-4966
,
1520-8524
Language:
English
Publisher:
Acoustical Society of America (ASA)
Publication Date:
2009
detail.hit.zdb_id:
1461063-2
