How Cognitive Factors and Motivation Shape Learners' Engagement in AI ‐Supported Language Learning?: Insights From the Combined SEM and fsQCA Approaches
Artificial intelligence (AI) has increasingly been integrated into both classroom‐based and informal language learning, with tools such as AI chatbots, automated feedback systems and generative AI applications offering personalised, interactive and adaptive learning experiences. However, little is known about how learners' cognitive factors and psychological mechanisms are associated with engagement in AI‐supported learning environments. Guided by Social Cognitive Theory (SCT), this study investigated the roles of AI literacy, growth mindset and critical thinking disposition in predicting learners' motivation, self‐efficacy and engagement. A cross‐sectional design was employed, with 781 university students participating in structured questionnaires. Structural equation modelling (SEM) was used to examine direct and indirect effects, while fuzzy‐set qualitative comparative analysis (fsQCA) identified multiple configurations of cognitive and psychological factors associated with high engagement. SEM results indicated that all three cognitive factors are positively associated with motivation and self‐efficacy, which in turn significantly mediated engagement. The fsQCA findings demonstrated equifinality, showing that different combinations of cognitive and psychological factors can lead to similar high engagement outcomes. These findings provide theoretical support for SCT in technology‐mediated learning and suggest practical strategies for educators, including tailoring instructional design to learners' cognitive profiles, fostering motivation and self‐efficacy and leveraging AI tools to enhance active and sustained engagement in language learning.