Affective processes in perceived inclusive EFL classrooms: Linking self-efficacy, engagement, resilience, and attitudes toward generative AI

Inclusive education, extending beyond classroom access to emphasize affective support, equitable participation, and learner diversity, frames this study of university English as a foreign language (EFL) learning in the age of artificial intelligence (AI). We examined affective processes linking academic self-efficacy, classroom engagement, resilience, and attitudes toward generative AI within learners' perceived inclusive EFL classroom contexts. A cross-sectional survey was administered to 345 undergraduates from four comprehensive universities in China. Structural equation modeling revealed that self-efficacy significantly predicted both engagement and resilience, with engagement partially mediating this relationship. Resilience was positively associated with learners' attitudes toward generative AI, whereas AI attitudes did not significantly moderate the other structural paths. The model explained 29% of the variance in engagement, 58% in resilience, and 63% in AI attitudes. These findings highlight the role of learners' affective resources in relation to engagement, resilience, and evaluations of AI-assisted learning, offering a delimited, affect-focused account rather than evidence of broad inclusion or AI-driven equity effects.