Evaluation of an AI-assisted Adaptive Educational Game System

Abstract As education continues to expand, both in outreach and in content, so too does the need for automated systems that can augment a student's educational process. This work builds on prior developments of a gamified adaptive tutoring system that automates and personalizes a student's learning process without instructor intervention. Our personalized learning system uses an augmented petri net graph structure to track student progress through the game, allowing us to enable or disable paths based on a student's performance. As an intelligent component, our system uses reinforcement learning agents to adaptively adjust system behavior based on student performance with the goal of optimizing student learning. The end result is a fully integrated game system that can measure student performance using integrated tests, leveraging that information to adjust game content, address learner misconceptions, and lead to a faster and more effective learning session. As part of continued research, we present data from pilot and comparison testing of our implemented game system. With our comparison testing, we show that the game provides greater educational utility for students compared to a standard lab. To verify improved educational utility, we present results from content tests given pre- and post-intervention. We further verify the game system's educational utility through an example case of the game adaptation, showing the full process of adapting to a student and providing educational assistance. By sharing our testing and verification, we demonstrate the effectiveness of our intelligent educational game system. In addition, we provide developmental insights for other researchers in this area who seek to implement or improve their own systems.