Abstract

Vowel epenthesis is a known phenomenon in Japanese where speakers insert a vowel ([u], by default) inside consonant clusters (e.g., in a VCCV sequence), due to the phonotactic constraint of the language. Dupoux et al. (1999) argue that vowel epenthesis occurs at a perceptual level. In their study, Japanese listeners perceived the epenthetic vowel [u] when the interconsonantal vowel [u] in the original stimuli was removed. In the present study, we investigate further whether (1) the perceptual system in Japanese listeners that causes vowel epenthesis is not only constrained by the phonotactic constraint (*VCCV) but also by the quality of the vowel that can be inserted, and (2) whether this constraint overrides existing acoustic information present in the input, e.g., the formant information of some other vowel. In our phoneme-detection experiment, we recorded a set of nonword stimuli (VCVCV) and gradually removed the interconsonantal vowel ([u] or other vowels). Some Japanese listeners perceived [u] not only when the originally present [u] was removed (as found in Dupoux et al.), but also when a vowel other than [u] was substituted in the interconsonantal position and that substituted vowel was partially removed, leaving only some remnant information.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call