Effects of Adaptive and Non-Adaptive Serious Games on Secondary Students' Mathematics Learning Gains
Adaptivity has emerged as a promising design feature in serious games, enabling learning experiences to be tailored to individual learner needs, performance, and progression. Despite its growing adoption, empirical evidence regarding the educational benefits of adaptive serious games remains limited, particularly in secondary school mathematics contexts. This study investigates the effectiveness of adaptivity in serious games by evaluating MathMystery, a mastery-driven adaptive mathematics game that integrates Bayesian Knowledge Tracing (BKT) and Generative AI (GenAI)-supported scaffolding. Using a comparative design, the adaptive version was evaluated against a non-adaptive version with identical learning content and core game mechanics among 254 secondary school students. A quasi-experimental pretest–post-test design was conducted to examine the impact of personalized adaptation on mathematics learning gains, with data analyzed using a mixed repeated-measures ANOVA. The findings indicate that students who engaged with the adaptive serious game achieved significantly greater learning gains than those who used the non-adaptive version. Further analysis indicates that students with lower prior knowledge benefited more from the adaptive condition than their higher-knowledge peers. These findings advance understanding of how knowledge tracing and GenAI can be integrated into adaptive serious games to improve learning outcomes.
Authors
- Cheng Yee Yap
- Tomás Maul (ORCID: https://orcid.org/0000-0002-3563-1115)
- Kher Hui Ng (ORCID: https://orcid.org/0000-0003-2743-3431)
Institutions
- University of Nottingham Malaysia Campus (MY)
Publication Details
- Journal
- International Journal of Education in Mathematics Science and Technology
- Published
- 2026-11-22
- DOI
- https://doi.org/10.46328/ijemst.8665
- Primary Topic
- Educational Games and Gamification
- Type
- article
- Field-Weighted Citation Impact
- 0.00