Exploring Variability in Chinese EFL Learners’ Emotional Engagement in GenAI-Supported Language Classrooms

Abstract

While the cognitive and technical benefits of generative artificial intelligence (GenAI) in second language (L2) learning are well recognized, its emotional impact remains underexplored, especially regarding how emotional engagement (EE) evolves and varies among learners in GenAI-assisted environments. This longitudinal mixed-method study, grounded in the complex dynamic systems theory, examined both intra- and inter-individual variability in EE among Chinese L2 learners in a GenAI-supported classroom. Quantitative results showed an overall increase in EE, with two distinct trajectories: a majority subgroup maintaining consistently high EE with modest growth and a smaller subgroup starting low but improving substantially over time. Qualitative data further revealed that learners perceived GenAI as a non-judgmental, supportive collaborator that fostered creativity, emotional security, and agency. Teacher scaffolding and peer collaboration also enhanced learners’ EE by balancing challenge and support and promoting collective engagement regulation. Together, these findings suggest GenAI tools can function not only as cognitive aids but also as emotional supporters. The study advocates for differentiated pedagogical strategies tailored to learners’ EE profiles in GenAI-enhanced L2 contexts.