Author Information
Min-hsun Su, Ming Chuan University, TaiwanPey-chewn Duo, Ming Chuan University, Taiwan
Abstract
The emergence of Large Language Models (LLMs), such as ChatGPT, has transformed English as a Foreign Language (EFL) instruction by promoting adaptive and autonomous learning (Chapelle, 2024; Van Horn, 2024). This study investigates EFL college students’ Language Learning Strategies (LLS) in AI-supported contexts and examines how these differ from traditional strategy use. Adopting a mixed-methods design that combined questionnaire surveys and semi-structured interviews, the research was guided by the Strategy Inventory for Language Learning (SILL) and the Strategic Self-Regulation (S2R) model (Oxford, 1990, 2016; Dillon, 2024). Findings reveal a marked shift toward higher-order strategic engagement. Students reported increased use of metacognitive, compensation, and affective strategies, with compensation strategies being most prominent due to AI’s capacity for real-time feedback and adaptive scaffolding. Metacognitive strategies were reinforced through AI-assisted planning and monitoring, while affective strategies benefited from AI’s non-judgmental feedback, reducing anxiety and enhancing motivation. In contrast, social strategies remained less frequently applied, reflecting the limitations of AI in replicating authentic interpersonal interaction. These results indicate that AI integration propels EFL learning toward greater strategic autonomy and self-regulation. The study underscores the pedagogical need for prompt literacy training to empower learners to leverage AI effectively for sustained strategic development and self-efficacy.
Comments
Powered by WP LinkPress