Use of acoustic cues in the perception of complex syllable structure

Abstract

This study examines second/foreign language learners’ (L2ers’) use of acoustic cues in the perception of syllable structure (onset clusters) that does not exist in the native language (L1, here, Korean). It investigates whether the target-like perception of syllable structure is facilitated by VOT ([stɪn] vs [sətɪn]) and partial devoicing of the liquid ([plim] vs [pəlim]) as compared to when either cue is absent ([blint] vs [bəlint]). This study also examined whether production errors can be predicted by perception errors. Twelve Korean-speaking English L2ers and 12 native English speakers completed AXB and production tasks in which they listened to and produced nonce words in the above three conditions. The AXB results showed significant effects of L1 and condition, but no interaction between the two. Participants were most accurate on the condition with VOT (English: 98%, Korean: 85%), followed by partial devoicing of the liquid (English: 96%, Korean: 79%) and without either cue (English: 94%, Korean: 74%). L2ers’ perception and production patterned alike, but no correlation was found between the two. These findings suggest that acoustic cues like VOT and partial devoicing of the liquid facilitate L2ers’ perception of new syllable structure, but do not play a direct role in its production.