Date of Award
3-2024
Document Type
Thesis
Degree Name
Master of Science in Computer Science
Department
Department of Electrical and Computer Engineering
First Advisor
Mark G. Reith, PhD
Abstract
In an era of evolving warfare, the Department of Defense (DoD) recognizes the value of serious games as immersive tools for teaching critical concepts. This thesis introduces a pioneering framework tailored to enhance learning objectives through the presentation of educational game content. This addresses the unique needs of the DoD and other educators who use games by providing measurements to assess educational games. This researches investigates how instructors might assess games as potential teaching tools. It establishes a five-phase process, providing a framework to assess educational games against predefined learning objectives and informing future game development. This thesis demonstrates games strategically aligned to learning objectives, offering a contrast to games solely focused on winning. The framework leverages four types of AI agents with heuristics inspired by player profiles. Two agents represent a competitive agent and a random action agent. Another two agents guide players or other agents toward game states conducive to learning objectives. These agents avoid game termination unless all learning objective content has been presented. The framework is tested through an experiment of different two-player games and learning objectives where the first is identified as the learner and the second is identified as the opponent. The five-phase process was applied to playthroughs involving 80 combinations of game sizes and player AI agents. The results indicate that games with greater player autonomy and longer durations enhance the coverage of learning objective content. AI agents employing a competitive heuristic exhibit a higher win rate (averaging 61% across all playthrough types). However, depending on the specific game, they may not achieve a favorable coverage rate for the full set of learning objectives (observed in 32% of matches). AI agents with a heuristic to learn, show a higher coverage (40% of matches) and win more often as the learner player than as the opponent (56% as the learner vs 42% as the learner's opponent). After thousands of two-player matches with various game inputs, no set of game size or AI agent produces a rate of full coverage greater than 65%. This non-linear game experiment can be contrasted with a linear lecture format, which yields 100% coverage of material but lacks some of the benefits games may provide. This experiment explores the trade space between learning objective content coverage and incorporating game elements in learning activities. This research not only addresses the critical challenge of evaluating educational content within DoD serious games but also delivers tangible contributions. These contributions include a novel framework, applicable metrics, and a demonstration of concept implementation. These advancements aim to measure the suitability of games and enhance educational outcomes in game-related learning activities.
AFIT Designator
AFIT-ENG-MS-24-M-015
Recommended Citation
Hastriter, Michael L. Jr., "Measuring the Presentation of Supporting Content for a Set of Learning Objectives throughout the States of an Educational Game Tree" (2024). Theses and Dissertations. 7681.
https://scholar.afit.edu/etd/7681
Comments
A 12-month embargo was observed for posting this work on AFIT Scholar.
Distribution Statement A, Approved for Public Release. PA case number on file.