Rong Liu PhD Candidate Second Language Acquisition and Teaching GIDP The American Association for Applied Linguistics (AAAL) 2008 Annual Convention March 29 to April 1, 2008 Washington, DC "Efficacy of computer assisted input-based language instruction on anaphora acquisition by adult English learners”

Previous studies of L2 anaphor acquisition have shown that it is difficult for learners to acquire the full properties of reflexives (e.g., Akiyama, 2002; Demirci, 2000; Thomas, 1989, 1993, 1995; White, 1998;). However, nearly all studies used offline measures for assessment of effects, which may just capture metalinguistic or conscious knowledge of the grammar that are not deployable during real-time processing (Doughty, 2004; VanPatten, 2002). This study investigated the efficacy of computer delivered processing instruction (PI) on anaphora acquisition with both online and offline measures. Numerous studies (e.g., Marsden, 2006; Sanz & morgan-Short, 2004; Takimoto, 2006: VanPatten & Sanz, 1995: VanPatten, 2005) have shown PI is effective. However, most studies used offline measures. This study used online methods, i.e., moving window self-paced reading tasks to investigate whether PI is as effective as its supporters have claimed. The role of feedback in CALL was also examined. The design was a pretest-treatment-posttest one. Forty intermediate ESL learners participated in the study. There were two treatment groups. One group received implicit feedback, i.e., yes/no feedback. The other group received explicit feedback, i.e., explicit grammar explanation and why wrong. All members received the same structured activities as stipulated by PI. Gains were assessed by sentence interpretation and production tasks as well as a self-paced reading task. Results showed that PI is effective in improving participants' ability to interpret and produce the target forms. However, it is only partially effective in alter learners' inefficient online processing during the self-paced reading tasks. No significant difference was found between implicit and explicit feedback groups. The study demonstrated the importance of the status of knowledge representation in L2 acquisition if we assume the ultimate goal of L2 learning and teaching is to cultivate the ability to use languages spontaneously and automatically.