Effectiveness of AI-assisted game-based learning on science learning outcomes, intrinsic motivation, cognitive load, and learning behavior

This study aimed to investigate the effectiveness of using AI-assisted game-based learning on science learning outcomes, intrinsic motivation, cognitive load, and learning behavior. A total of 202 seventh graders were recruited and randomly assigned to the following three groups: (1) Game only (N = 70), (2) GameGPT (N = 63), and (3) GameGPT_examples (N = 69). The experimental groups received game-based learning with the assistance of ChatGPT with or without examples, while the control group received only game-based learning. The results showed that students in the GameGPT_examples group significantly outperformed those in the Game only group. Students in the GameGPT and GameGPT_examples groups reported significantly higher perceived competence than those in the Game only group. Furthermore, students in the Game only group reported a greater mental burden than those in the GameGPT_examples and GameGPT groups. The findings from learning behavioral analytics and interviews suggest that AI-assisted game-based learning can enhance students’ intrinsic motivation, reduce cognitive load, and promote effective learning behavior in science learning. This study has important implications for the design and implementation of AI in game-based learning environments that aim to improve students’ learning outcomes and motivation.