A Study on the Factors Influencing User Experience of AI Pose Recognition Feedback Systems in Ballet-Class Contexts

With advances in artificial intelligence and computer vision, pose recognition-based feedback systems are increasingly being introduced into dance classes to support movement understanding and error correction. However, how learners interpret and adopt such feedback during actual classroom use, and how this process shapes their intention to continue using the system, remains insufficiently understood. This study develops and validates a three-layer “System–Psychological–Experience” model, with learning experience positioned as the key mechanism linking antecedent factors to usage intention. A sequential mixed-methods design was employed. In Phase 1, interviews were conducted with 20 dance majors to identify factors relevant to classroom integration. In Phase 2, structural equation modeling was performed using 398 valid survey responses. In Phase 3, explanatory interviews with five students and five teachers were conducted to further interpret the underlying mechanisms. The results showed that learning experience was the strongest direct predictor of usage intention (β = 0.409, p < 0.001). At the system layer, perceived usefulness and perceived ease of use contributed to usage intention mainly by strengthening learning experience. At the psychological layer, system trust and perceived value showed the same indirect pattern through learning experience. Teaching context also played an important role. Teaching compatibility directly enhanced learning experience, while teacher influence both strengthened learning experience and indirectly contributed to usage intention by increasing system trust and perceived value. Overall, the findings suggest that learners’ intention to continue using pose recognition-based feedback systems is closely associated with their overall learning experience in teacher-mediated classroom practice, and the study offers implications for classroom-oriented design.