Supplementary file 1_The impact of intention to adopt generative AI for exercise information on exercise adherence among Chinese college students: the mediating role of autonomous motivation and network analysis.docx

BackgroundInsufficient exercise adherence among college students is a common health issue, while generative artificial intelligence (AI) provides a new approach for personalized exercise guidance. However, the extent to which the intention to adopt generative AI for exercise information affects exercise adherence, and the role of autonomous motivation in this process, remains underexplored in empirical research.
MethodsThis study employed a cross-sectional survey design and administered a questionnaire to 1,878 Chinese undergraduates via stratified random cluster sampling. The core variables were measured using the self-developed “Intention to Adopt Generative AI for Exercise Information Scale,” the Chinese version of the “Autonomous Motivation Scale,” and the “Exercise Adherence Scale.” Mediation effects were tested using the PROCESS macro, and network analysis was performed in R to examine the complex interactions among variables.
ResultsThe intention to adopt generative AI for exercise information was significantly positively correlated with exercise adherence among college students (β = 0.299, p < 0.001), and it produced a significant indirect effect by enhancing autonomous motivation (effect size = 0.116, 95% CI [0.094, 0.139]), with the mediation effect accounting for 27.78% of the total effect. Network analysis further identified “intrinsic motivation” and “behavioral habits” as the core nodes with the most significant influence on the overall psychological-behavioral network.
ConclusionThis study explores the associations and potential mechanisms by which generative AI may relate to exercise adherence via autonomous motivation, supporting a theoretical pathway of “technology adoption—motivation internalization—behavior persistence.” The findings offer a novel perspective on the theoretical associations underlying AI-enabled health behaviors and provide preliminary correlational evidence to inform the future design of generative AI applications for health interventions.