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

Abstract

This study investigates the effect of phonetic similarity between L1 and L2 sounds on the perception of L2 contrasts. It employs hypotheses from Flege’s speech learning model as well as two types of assimilation patterns from Best’s perceptual assimilation model. L1 L2 phonetic similarities were measured perceptually and acoustically prior to the experiments. The first experiment examined the perception of the Korean voiceless sibilant contrast /s-s*/ by American English L2 learners. The second experiment examined the perception of the English voiceless sibilant contrast /s-■/ by Korean L2 learners. The findings of the first experiment show that a higher degree of phonetic similarity did not always result in good identification of L2 contrasts by American English advanced learners of Korean; overall, however, more similar L2 sounds were perceived better than less similar L2 sounds. In the second experiment, L2 contrasts that are acoustically and perceptually different from each other were perceived by L2 learners of Korean just as well as native speakers of American English perceived them. The results were not compatible with the predictions of either Best’s or Flege’s models, but the second result partially supported Best.