From personal best goals to engagement in AI-supported language learning: The achievement goal theory perspective

Drawing on Achievement Goal Theory, this study examined how personal best goals are associated with university students’ engagement in AI-supported language learning. It further investigated the mediating roles of motivation, flow, and self-efficacy, while also exploring multiple configurations leading to high engagement. A cross-sectional questionnaire survey was conducted with 687 university students who had prior experience using AI-enabled tools for language learning. Structural equation modelling (SEM) was first employed to test the direct and indirect relationships among the variables, and fuzzy-set qualitative comparative analysis (fsQCA) was then used to identify alternative pathways to high engagement. The SEM results showed that personal best goals were positively associated with motivation, flow, self-efficacy, and engagement. Motivation, flow, and self-efficacy were also positively associated with engagement and each mediated the relationship between personal best goals and engagement. The fsQCA results identified four sufficient configurations for high engagement, suggesting that students may become highly engaged through different combinations of achievement goals and psychological conditions. These findings extend research on AI-supported language learning by highlighting the role of self-referenced achievement striving in shaping learner engagement.