The effects of an AI-supported gamified learning program on elementary students' mathematics anxiety, learning motivation, and achievement

Although gamified learning can enhance motivation, many systems rely on uniform feedback and provide limited personalized and supportive interaction. Evidence regarding the incremental effects of AI-supported gamification in elementary mathematics remains limited. This study examined the effects of an AI-supported Gamified Learning Program (AI-GILP) on elementary students’ mathematics anxiety, learning motivation, and achievement. Using an experimental design with individual-level random assignment, 174 Chinese fifth-grade students were assigned to AI-GILP, gamified learning without AI support (GILP), or traditional lecture-based instruction (TL). The intervention lasted two weeks and was followed by a delayed posttest four weeks later. Mathematics anxiety, learning motivation, and achievement were assessed at pretest, posttest, and delayed posttest. Data were analyzed using analysis of covariance and mixed-design analyses of variance. AI-GILP produced significantly lower mathematics anxiety and higher learning motivation than both GILP and TL, with overall group effects of partial η 2 = .208 and .184, respectively. At the delayed posttest, the descriptive means for anxiety and motivation retained the same favorable ordering for AI-GILP, although pairwise significance at that assessment was not formally tested. For mathematics achievement, both gamified conditions outperformed TL, whereas AI-GILP did not significantly differ from GILP (Cohen’s d = 0.19). The motivational advantage of AI-GILP was concentrated in intrinsic motivation and perceived competence, with no significant difference in effort/importance. The findings suggest that gamification primarily accounted for the observed achievement gains, whereas the AI Learning Companion may have provided additional benefits for mathematics anxiety and selected dimensions of learning motivation. The proposed mechanisms involving psychological-need support, emotional regulation, adaptive feedback, and motivational internalization were not directly tested and require further process-oriented research.