Listener perceptions of word-final /l/: Effects of listener dialect

This study examines the effect of listener dialect on perception of the /l/ sound. In some varieties of English, word-final /l/ is more vocalized than in others. The current study follows up on our previous study, which investigated Southern White English (SWE)-speaking listeners' perception of word-final /l/ produced by SWE speakers. The aim of the current study is to examine perception of listeners who are speakers Upper Midwest English (UME), a dialect that shows less word-final /l/ vocalization than SWE.

Listeners included 32 monolingual speakers of UME who judged the /l/-likeness of single words containing word-final /l/ produced by adult and child speakers of SWE using a multiple-choice task and a visual analog scale. Ratings were also compared with those of SWE listeners.

Like the SWE listeners, UME listeners judged SWE adult speakers' word-final /l/ predominantly as accurate, while those of children were judged as more vowel-like. Vowel context also influenced both UME and SWE listeners' judgments, with productions of /l/ following low back vowels generally judged as less accurate than those following high front vowels. Despite these similarities, there were subtle differences across the two dialect groups, particularly in terms of the effect of vowel context.

This demonstrates that despite subtle differences in perceptual ratings across dialect groups, speaker age and vowel context more substantially influence perceptual judgments. This suggests that it is important to consider the potential impact of vowel context and to include word-final /l/ in both high front and low back vowel contexts to prevent overidentification of errors in clinical settings.