A computer-based geometry game was adapted to allow for play using a conceptual rule or an arithmetic problem-solving mechanic. Participants (n = 91) from an urban middle school were randomly assigned to experimental conditions. Results suggest that play in the number condition was more situationally interesting than play in the rule condition. Participants in the rule condition were found to perform better in the game than those in the number condition. Learning outcome results suggest that in the number condition, but not the rule condition, playing more levels in the game diminishes the gain from pretest to posttest. For the design of games for learning, results highlight the importance of choosing a game mechanic that reflects the intended learning outcomes.
展开▼
