On the perception of Korean and English sibilants in second language acquisition

Abstract

This paper presents the findings of two perception experiments that examine both the identification of Korean contrast (/s-s*/) by American English learners of Korean (Experiment 1) and the identification of English contrast (/s-š/) by Korean ESL learners (Experiment 2). Two prominent L2 models that are primarily based on the perception of L2 segments are the Speech Learning Model (SLM) and the Perceptual Assimilation Model (PAM). From a PAM perspective, the result of Experiment 1 did not support the prediction of the Category Goodness (Korean /s-s*/) contrast, which expected a better identification performance than chance accuracy. The result of Experiment 2 supported the prediction of the Two Category (English /s-š/) contrast, which expected an excellent identification performance. From a SLM perspective, on the other hand, it was predicted that American English advanced learners of Korean perceived the Korean /s/ accurately. But their overall mean identification scores were slightly above the chance level. Korean advanced learners of English perceived the English contrast accurately, as expected. In sum, the results of each experiment were partially compatible with the predictions of the two models.