Modeling AI trust and AI self-efficacy in AI-assisted EFL classrooms: the role of basic psychological needs in classroom engagement

The increasing use of artificial intelligence (AI) in English as a foreign language(EFL) classroom has raised questions about how learners' perceptions of AI relate to their classroom engagement. Against this backdrop, this study examined the relationships among AI trust, AI self-efficacy, basic psychological needs (BPNs) satisfaction, and classroom engagement in an AI-assisted EFL context. A total of 1,002 Chinese university students completed a questionnaire measuring these constructs. Data were analyzed using confirmatory factor analysis and structural equation modeling (SEM). The results showed that both AI trust and AI self-efficacy were positively associated with learners' satisfaction of autonomy, competence, and relatedness. BPNs satisfaction, in turn, was strongly related to classroom engagement. However, the direct effects on classroom engagement were not significant. Instead, both variables showed significant indirect associations with engagement through BPNs satisfaction, indicating a pattern consistent with statistical mediation. These findings suggest that learners' AI-related beliefs are related to classroom engagement primarily through their associations with motivational experiences rather than directly driving engagement behaviors. This study contributes to understanding the relationships between learners' perceptions of AI and classroom engagement. The study also highlights the importance of need-supportive instructional design in AI-assisted EFL classrooms.