Artificial Intelligence in L2 learning: A meta-analysis of contextual, instructional, and social-emotional moderators

Abstract

This study conducts a comprehensive meta-analysis to examine the effectiveness of artificial intelligence (AI) interventions in language learning and to explore the moderating effects of various contextual, instructional, and social-emotional factors. By synthesizing data from 49 studies comprising 79 reports, the analysis reveals that AI interventions have a significant positive impact on language learning outcomes. The study identifies several key moderators influencing AI effectiveness, including the learning environment, intervention duration, educational level, participants’ age, sample size, language skills targeted, interactivity types, and feedback sources. The findings indicate that online and blended learning environments, medium-duration interventions (6 weeks–6 months), higher education settings, and young adult learners show the highest effectiveness. Additionally, AI interventions are most effective in enhancing listening and speaking skills, with significant benefits also observed for writing and vocabulary acquisition. Social-emotional factors such as motivation, anxiety reduction, and willingness to communicate are positively influenced by AI, though engagement and satisfaction exhibit variability. The study underscores the importance of personalized and adaptive AI tools, leveraging both automated and human feedback to maximize learning outcomes. These insights contribute to the growing body of literature on AI in education, offering guidance for optimizing AI-enhanced educational tools to support diverse learners effectively. Future research should further investigate individual differences among learners to tailor AI interventions more precisely and enhance their educational impact.