Design and Evaluation of an Adventure-Based Educational Game with Mamdani Fuzzy Logic-Based Adaptive Difficulty for Computer Systems Learning

Abstract

This study designed and evaluated an adventure-based educational game with a Mamdani fuzzy logic-based adaptive-difficulty mechanism for Grade VII computer systems learning. The proposed mechanism dynamically adjusts question difficulty using two student-performance indicators: the percentage of correct answers and average response time. These inputs are processed through nine Mamdani fuzzy rules to determine Easy, Medium, or difficult question levels. The game was developed using the Multimedia Development Life Cycle (MDLC) and evaluated using a design-and-development approach combined with a one-group pre-test–post-test design involving 40 Grade VII students. Evaluation included functional and algorithm testing, adaptive-difficulty scenarios, User Acceptance Testing (UAT), and learning-outcome analysis. The adaptive mechanism generated different difficulty decisions according to different student-performance conditions. Student UAT yielded an overall acceptance score of 90.5%, while the participating teacher provided a score of 95%. The mean learning score increased from 52.78 (SD = 15.52) in the pre-test to 80.98 (SD = 3.62) in the post-test, with a statistically significant within-group difference, t (39) = 10.84, p < 0.001, Cohen’s dz = 1.71, and a normalized gain of 0.60. These findings demonstrate the successful implementation of an interpretable adaptive-question mechanism and substantial learning-score improvement following game-based activities. However, causal effectiveness cannot be established because the study did not include a control group.