Beyond AI Use: How AI Literacy Is Linked to Critical Thinking through AI-Assisted Problem-Solving Ability and AI Reliance

Abstract

With the widespread adoption of generative artificial intelligence in English as a Foreign Language (EFL) learning, helping learners use AI effectively to solve language-related problems while maintaining independent judgment and critical thinking has become a key issue in technologyenhanced language education. However, existing studies have largely focused on EFL learners’ acceptance of AI, intentions to use it, and usage behaviors, offering limited explanations of how AI literacy shapes learners’ interactions with AI and relates to critical thinking. To address this gap, the present study drew on self-regulated learning theory and adopted an explanatory sequential mixed-methods design. Questionnaire data from 450 university students with generative AIassisted EFL learning experience were analyzed using structural equation modeling and bootstrapping. Subsequently, 20 participants selected through stratified purposive sampling took part in semi-structured interviews, which were analyzed thematically. Quantitative results showed that AI literacy was positively associated with AI-assisted problem-solving ability and indirectly associated with critical thinking through AI-assisted problem-solving ability and AI reliance. Thematic analysis further revealed how learners translated AI-related knowledge and skills into effective problem-solving practices and how different levels of cognitive control were associated with distinct patterns of AI use and cognitive outcomes. This study deepens the understanding of the internal cognitive mechanisms underlying EFL learners’ adaptation to AI and provides practical implications for the systematic development of AI literacy and critical thinking in EFL education.