English vowel perception hierarchy of Korean learners: Effects of Residential Context*

The current study investigates the perceptual hierarchy of vowels among Korean speakers learning English to determine the influence of the residential context, (between Korea versus the United Kingdom), on this hierarchy. Sixty-five Korean participants (30 Korea-based, 35 UK-based) performed a vowel identification task and a speech-in-noise recognition task. Despite the two groups yielding highly comparable hierarchies (ρ=.883, ρ<.001), differences arose between the two groups in the form of the identification accuracy of /ɒ/, /u/, /aɪ/, and /aʊ , all of which tended to be higher in the Korea-based group. The poor performance of the UK-based group in identifying /ɒ/ suggests that the influence of British English plays a role here. A generalized linear mixed-effects logistic regression analysis found residential context to be a statistically significant correlate of identification accuracy even after controlling for individual differences. While native accent speech-in-noise recognition was a reliable predictor of participant performance, Korean accent speech-in-noise recognition was not. Confusion matrix analysis revealed three common trends: systematic confusion between the back vowels, bidirectional confusion between /i/ and /ɪ/, and dispersed confusion with /ʌ/. Together, these results indicate a significant L1 influence on Korean learners’ vowel perception, with residential context exerting selective rather than uniform effects on individual vowels.