Recasts, foreign language anxiety and L2 development during online mobile-mediated interaction

Despite the wealth of studies on corrective feedback (CF) and its relationship with individual learner factors, little is known about how foreign language (FL) anxiety moderates the effectiveness of recasts during mobile-mediated audio interactions. The present study thus examined the association between learners’ FL anxiety, the effectiveness of recasts, and learners’ responses to recasts during synchronous mobile-mediated interactions via audio call. Two intact classes of EFL (English as a Foreign Language) learners were assigned into a control group and an experimental condition. After taking pre-tests, the participants of the experimental condition participated in four sessions of mobile-mediated oral interaction with an interlocutor via WhatsApp and received recasts for their definite and indefinite article errors. The participants of the control group also participated in the mobile-mediated interactions but received no recasts for their errors. Learners’ improvement was measured on two occasions following the fourth treatment session. The participants’ anxiety was also measured as a continuous variable using a 5-point Likert scale. Mixed between-within group ANCOVA results and regression analysis provided evidence for the efficacy of recasts delivered during mobile-mediated interactions, as well as the significant role of learners’ anxiety as a predictor of the effectiveness of recasts. The results also indicated that learners with low anxiety were significantly more successful in modifying their incorrect forms following recasts compared to learners with higher anxiety during the mobile-mediated interactions.