The Effects Of Recasts And Metalinguistic Clues In Dyadic Interaction On English Article Acquisition

The present study investigates the effect of two types of corrective feedback (recasts and metalinguistic clues) on the acquisition of English articles in sophomores of high school and the extent to which individual differences may influence the effectiveness of CF. The individual difference factors are language proficiency and learners'attitudes towards error and error correction. The target structure is English definite and indefinite articles.This study lasts about ten weeks, and employs a pre-test, eight treatment sessions, an immediate posttest, a delayed post-test and two questionnaires design. The subjects in the present study are 44 students from High School Attached to Hunan University (in Grade2), including 29 boys and 15 girls. There are eight treatment sessions. Therefore, the 44 subjects are randomly assigned to 3 groups: the recast group (n=15, 10 boys and 5 girls); the metalinguistic group (n=14, 9 boys and 5 girls); the control group (n=15, 10 boys and 5 girls). After the immediate posttest, the entire subjects take a questionnaire designed to examine learners'attitudes towards error and error correction. After the delayed posttest, the two experimental groups take an exit questionnaire, which aims to exam whether the learners notice the CF that they have received.The SPSS (Statistical Package for the Social Sciences, version 13.0) is used as a statistical analytical instrument. Through the quantitative and qualitative analysis, the following conclusions are made: Firstly, the recast group and the metalinguistic group have much more gains in the use of English definite and indefinite articles than the control group who receives no feedback both in the immediate and delayed posttest, but there is no significant difference between the recast group and the metalinguistic group. Secondly, the potential benefit of recasts and metalinguistic clues to L2 learning is independent on the learners'proficiency level. That is, all high-proficiency learners and low-proficiency learners benefit equally from the two CF techniques. Besides, the immediate gains in the recast and metalinguistic group are positively correlated with learners'attitudes towards error and error correction. In contrast, the gain scores in the control group are not significantly related to learners'feelings about error and error correction.Based on the above conclusions, the following pedagogical implications for second language teaching and learning are explored. Teachers should attach more importance to the CF in foreign language learning and teaching, provide more systematic and consistent feedback, and the CF should be clear enough to be perceived.