The impact of emotional intelligence on emotional engagement in GAI-assisted learning: a moderated chain mediation model

During generative artificial intelligence (GAI)-assisted learning, learners’ emotional information processing adapts to the technology-enhanced learning context. Understanding how learners’ emotional intelligence shapes emotional engagement in this context is vital for fostering effective learning. Therefore, this study constructed a moderated chain mediation model to examine the moderating effect of GAI usage experience incorporated into the developed relationship between emotional intelligence dimensions and emotional engagement. Data collected from 794 participants was analyzed. The results suggest that (1) clarity (β = 0.030, p < 0.01) and repair (β = 0.033, p < 0.001) mediate the linkage between attention and emotional engagement in a parallel rather than chain manner; (2) GAI usage experience serves only to moderate the impact of clarity on emotional engagement (β = 0.124, p < 0.001), highlighting the role of GAI tools in cognitive enhancement. These findings refine the theoretical understanding of emotional intelligence’s function in technology-enhanced learning and offer practical guidance for designing emotionally supportive GAI applications.