AI-Mediated Language Learning and EFL Learners’ Self-Confidence, Self-Regulation, Well-Being, and L2 Motivation: A Mixed Method Study

Abstract

Artificial Intelligence (AI)-mediated language learning supports student-centered approaches in language education. Hence, using a mixed-method design, this study examined its effects on EFL learners’ self-confidence, self-regulation, well-being, and L2 motivation. A number of 75 EFL learners participated in the study, with 38 assigned to the control group (CG) and the remaining to the experimental group (EG). Four validated scales were administered to assess the dependent variables in both groups. The EG received AI-mediated instruction, while the CG followed traditional methods without AI support. After a three-month intervention, the same four questionnaires were distributed as post-tests. Additionally, semi-structured interviews were conducted with 14 EG participants to explore their perceptions of AI-mediated language learning. The quantitative findings indicated significant improvements in the EG compared to the CG across all measured variables. Interview responses further revealed that EG participants held positive attitudes toward the integration of AI in EFL classrooms, highlighting increased engagement and motivation. These results suggest that AI-mediated instruction can enhance learners’ psychological and motivational dimensions, contributing to more effective language learning. The study concludes with pedagogical implications and recommendations for integrating AI tools into EFL teaching practices. Future research could explore long-term impacts and examine specific AI applications across different educational contexts