Designing Immersive Fitness Environments: Integrating Generative AI and Scenario Creation in College Physical Education

Abstract

This practice-oriented study explores the integration of Generative Artificial Intelligence (GAI) into the design of immersive, scenario-based environments for university physical fitness education. We developed and implemented the GAI-driven Situational Creation for Fitness (GAI-SCF) model—a structured framework that leverages narrative and multimodal content generation to enhance spatial and experiential engagement in physical activity settings. Over a 12-week semester, 120 students participated in a mixed-methods study, with an experimental group (n=60) experiencing the GAI-SCF model and a control group (n=60) receiving traditional instruction. The intervention followed a four phase cycle (Analysis, Generation, Implementation, Evaluation), using a GAI platform to create personalized, thematic workout environments (e.g., “Cybernetic Rhythm Battle,” “Eco-System Rescue Mission”). Quantitative results showed that the experimental group achieved significantly greater improvements in skill performance (F(1, 117) = 28.74, p < .001, η² = .20), reported higher situational interest (t(118) = 3.89, p < .001, d = 0.71), and exhibited stronger learning motivation (t(118) = 4.56, p < .001, d = 0.83). Qualitative analysis revealed that the model fostered novel, autonomous, and socially connective learning atmospheres, effectively transforming the gym into a dynamic narrative space. This paper presents the GAI-SCF model as a replicable design framework for educators and designers seeking to use GAI to reimagine physical education environments as adaptive, engaging, and architecturally responsive experiences.