An Intelligent Serious Game for Digital Logic Education to Enhance Student Learning

Abstract

Contribution: A general-purpose model for integrating an intelligent tutoring system within a serious game for use in higher education. Additionally, this article also offers discussions of proper serious game design informed by in-classroom observations and student responses. Background: Personalized learning in higher education has become a key issue when working to improve student performance. By combining an intelligent tutoring system within a serious game, students can be engaged in their learning through gamified lessons while simultaneously receiving personalized and timely scaffolding to support their learning. Furthermore, related systems have not explored a general-purpose model for this type of system that can apply to any game or domain. Intended Outcomes: The combined intelligent tutoring system and serious game is well-received by students as determined by student surveys. Furthermore, students show better engagement in the given material and better performance on pre-post-intervention content tests. Application Design: The proposed system is a modular, general-purpose approach for integrating an intelligent tutoring system into any serious game for education. Using the machine learning paradigm of reinforcement learning, the system can adapt to student responses to improve future scaffolding. Findings: The results of the in-classroom testing are promising. Students who interacted with the intelligent game showed improved performance on content tests and positive responses on surveys regarding system usability and utility. This article also shows that students who used the intelligent game took less time and attempts to complete game sections, owing to the utility of the personalized support.