Guion, S.G., Flege, J.E., Akahane-Yamada, R., & Pruitt, J.C. (2000). An investigation of current models of second language speech perception: the case of Japanese adults’ perception of English consonants. Journal of Acoustical Society of America, 107 (5), 2711–2724.
This study aimed to test how well the two models, Perceptual assimilation model (PAM) and Speech learning model (SLM), can explain for cross-language speech perception and L2 phonetic learning. As both models incorporated the concept of perceived phonetic distance, the authors first conducted an identification and a rating experiment on near-monolingual Japanese listener to access the perceived relation between English and Japanese consonants. It was found that the distance between an L1 and L2 consonant was indeed a good predictor of the discrimination of a foreign sound. Later on, a perception experiment was executed on three groups of Japanese listeners differing in English proficiency and one group of English native listeners to examine whether learning could effect discrimination. The fact that some pairs such as the English-English contrast of /b/ and /v/, the Japanese-English contrast of /b/ and /v/, and the Japanese-English contrast of /ɾ/ and /ɹ/ could be discriminated by the high (and mid) proficiency speakers indicated that learning is effective to some extent. Its implications for the two models are as followed: first, in terms of PAM, which is a cross-language speech perception model, the results supported that it could be further expanded into a L2 acquisition model as near-monolingual Japanese listeners also performed as predicted. On the other hand, SLM was less fruitful in that it failed to be further extended to explain for less proficient L2 learners.
No comments:
Post a Comment