The Effects of AI-Assisted Dance Learning on Embodied Awareness and Creative Self-Efficacy

Abstract

Artificial Intelligence (AI) in dance education has begun to change the teaching, practice and absorption of human movement in a new way. This paper investigates the two sides of AI-assisted dance classes on students' sense of embodiment and creative self-efficacy. Integrate phenomenological frameworks of the moving body with Bandura's social cognitive theory to explore how AI-driven computer vision and real-time pose estimation technologies alter dancers' sense of proprioception. Research on the shift from older mirror-based externalisation to new algorithmic systems for self-monitoring has been conducted. Although AI feedback has improved the precision of kinematics, it may also harm the all-encompassing phenomenology of the 'lived body' if not used in teaching thoughtfully. In addition, the paper also studies how adaptive, confidence-weighted AI feedback loops affect creative self-efficacy, that is, individuals' sense of their own ability to create new works of art. Based on the above results, AI should be used as a cooperative learning partner rather than an inflexible test-taker to improve both the accuracy of movement and the self-confidence needed for creative exploration. Finally, this paper proposes a human-AI cooperative model for dance education that combines the accuracy of algorithms with the expressiveness of human art.