Theorizing positive transfer in cross-linguistic speech perception: The Acoustic-Attentional-Contextual hypothesis

Abstract

Can non-natives outperform natives on speech discrimination? Surprisingly, Cantonese listeners discriminated English stress more accurately than did English listeners. To ascertain its generalizability, I further ask whether this Cantonese advantage in English stress discrimination is equally potent across pitch accent and vowel reduction contexts. Sixty Cantonese and English listeners completed four blocks of English stress discrimination task with varying pitch accent and vowel reduction contexts. In the absence of rising pitch accent pattern and vowel reduction, the Cantonese listeners outperformed the English listeners on English stress discrimination. However, the Cantonese advantage disappeared when either rising pitch accent pattern or vowel reduction was present. When both rising pitch accent pattern and vowel reduction were present, the Cantonese listeners even performed poorer than the English listeners. The findings underscore two constraints of the Cantonese advantage in English stress discrimination—rising pitch accent pattern and vowel reduction. Based on collective research on non-native advantage in speech perception, the Acoustic-Attentional-Contextual hypothesis is proposed.