University physical education courses often face challenges in sustaining students’ learning engagement, particularly when practice tasks feel repetitive, progress is hard to perceive, and there is a lack of personalized feedback based on individual students’ immediate performance. Although AI (Artificial Intelligence)-based movement assessment can help address the need for instant, individualized feedback, maintaining students’ engagement over time remains a persistent challenge. In response, we developed a gamified AI movement assessment and feedback (G-AI-MAF) approach based on the ARCS motivational model (i.e., Attention, Relevance, Confidence, and Satisfaction), and adopted a quasi-experimental design to explore the impact of gamification mechanisms, such as levels, experience points, leaderboards, and player data, on student learning. The participants consisted of 80 students from two yogism classes at a university. One class served as the experimental group, adopting the G-AI-MAF approach, while the other class served as the control group, adopting the conventional AI movement assessment and feedback (C-AI-MAF) approach. The results showed that the G-AI-MAF approach significantly improved students’ yogism movement skills performance and learning engagement. In terms of behavioral patterns, the experimental group took tests and checked results more frequently than the control group. These findings suggest that the gamification mechanism motivated students to engage in learning, significantly enhancing their active learning behaviors and the depth of system usage.
Huang et al. (Tue,) studied this question.