An epenthetic vowel between consonantal sequences in perception and production by Japanese.

Abstract

Native Japanese have a strong tendency to epenthesize a vowel when producing consonantal sequences in English. This study investigated how Japanese learners of English perceive and produce word‐initial CCV versus CəCV contrasts (e.g., sport versus support). Two types of tasks were employed: a categorial ABX task for perception and a delayed imitation task for production. Nonsense words of the form /C1C2ani/ (e.g., spani) and /C1əC2ani/ (e.g., sepani) served as the stimuli where C1C2 combinations were /s‐p/, /s‐t/, and /s‐k/. In the ABX task, participants heard three short sentences that contained the target words and answered whether the third target word was the same as the first or the second one. In the delayed imitation task, participants heard the production of a native speaker in a carrier sentence (e.g., say sepani now) twice and produced the target word in isolation and then in the carrier sentence. Japanese participants made both perception and production errors. Interestingly, the majority of errors in the production task were vowel deletion for the CəCani contexts rather than vowel epenthesis for the CCani contexts. The relationship between perception and production among L2 learners as well as influence of L1/L2 phonotactic differences will be discussed.