A Gamified Approach to Teaching Behavior Tree Execution Semantics

Behavior Trees (BTs) are a central model in game AI and robotics, but their execution semantics are challenging for novices to grasp. Traditional educational methods fail to bridge the gap between static structure and dynamic behavior. To address this, we developed an educational game that provides real-time visual feedback on BT execution within a structured, level-based learning environment. Following a Design-Based Research methodology, the system was refined through iterative expert evaluation. Analytical assessments, including heuristic evaluation and content validity analysis, demonstrate that the refined system significantly improves usability, pedagogical alignment, and support for higher-order cognitive tasks. The results indicate that interactive, feedback-driven visualization can effectively illustrate complex AI reasoning structures for novices. This work contributes a validated framework for teaching BTs and offers design insights applicable to other complex logic-learning domains.