Title: Production and perception of a temporal contrast by native and non-native speakers

Authors: Tessa Bent, Ann R. Bradlow, and Bruce Smith

In previous work, we found that both native and Chinese-accented talkers of English varied considerably in the extent to which they realized the duration contrast for vowels before voiced versus before voiceless consonants. The current study explored the perceptual consequences of this individual variability in production. Twenty native and 35 non-native listeners participated in a minimal pair identification task using stimuli from two native and two non-native talkers who differed substantially in the extent to which they realized the temporal contrast. Stimuli were consonant-vowel-consonant words that differed only in the voicing of the final consonant. Both native and non-native listeners exhibited sensitivity to variation in the production of this temporal contrast: greater identification accuracy was observed for relatively large duration differences. There was also a talker-listener interaction such that non-native listeners identified words produced by the non-native talkers more accurately than those produced by the native talkers. In a follow-up experiment, we investigated the perception-production relationship for this temporal contrast in the non-native listeners. Generally, non-native listeners who were more sensitive to this duration contrast in native-accented English showed a greater duration contrast in English production than non-native listeners who were less sensitive to this contrast. [Work supported by NIH-NIDCD.]

(word count: 200)