L1 PHONOLOGICAL TRANSFER IN L2 ENGLISH PERCEPTION: A MIXED STUDY OF CANTONESE-SPEAKING UNDERGRADUATES

Abstract

This study examines the extent to which the phonological system of Cantonese influences the perception of English segmental and suprasegmental features among second language (L2) learners. The research is motivated by persistent perceptual challenges faced by Cantonese-speaking learners, particularly concerning phonemes absent in their L1 and prosodic mismatches between tonal and stress-based systems. Grounded in two theoretical frameworks—the Revised Speech Learning Model (SLM-r) and the Perceptual Assimilation Model for L2 learners (PAM-L2)—this study adopts a mixed-methods approach, combining a phoneme perception questionnaire with follow-up interviews. Data were collected from 60 Cantonese-speaking undergraduates in Guangdong, China. Results indicate systematic segmental confusion (e.g., /θ/ → /s/, /v/ → /w/, /æ/ → /ɛ/) and suprasegmental misperception stemming from tonal transfer, such as the reliance on pitch and syllable duration to identify English lexical stress. Lexical familiarity was found to modulate perception accuracy, with unfamiliar words prompting increased L1-based inference. These findings confirm that L2 speech perception is heavily filtered through L1 categories and align with assimilation types predicted by SLM-r and PAM-L2. The study underscores the need for perception-based instruction tailored to L1-specific constraints and contributes empirical evidence to the growing body of research on cross-linguistic phonological transfer in tonal-language learners.