Enhancing Digital Game-Based Learning Through Reflective Game Design (RGD)

This study by Shaheen and Fotaris from the University of Brighton examines whether Reflective Game Design (RGD), a framework built on Kolb's experiential learning cycle, improves learning outcomes in digital game-based learning compared with conventional game design. The authors built two prototypes of a self-awareness serious game called "The Mend," a house renovation puzzle, one with explicit reflective features such as process displays, prompts, and diary-based reflection points, and one without. The central claim is that reflective features make digital games more effective as learning tools by promoting critical thinking, self-directed learning, and deeper engagement.

The experiment used a pre-test/post-test control-experimental design with two groups of 69 young adults aged 18 to 24, drawn mainly from undergraduate students and online gaming communities. Each group answered six multiple-choice questions before and after gameplay. Both groups improved, but the RGD group's mean post-test score rose from 2.44 to 4.98, compared with 2.83 to 3.42 for the non-RGD group. Paired t-tests found significant gains within each group (p < .001), and a comparison of learning gains between groups reported t(136) = 5.29, p < .01, favouring the RGD prototype. Qualitative responses from the experimental group highlighted stress relief from breathing exercises and value in the journaling activity, while also requesting more challenging puzzles and a more engaging narrative.

The study has notable limitations that bear on how its conclusions should be read. The sample was drawn from a single university and convenience sources, the outcome measure was a six-item instrument developed by the researchers rather than a validated scale, and the comparison relied on a non-random assignment of existing classes. Retention was not measured beyond the immediate post-test, despite retention being one of the stated objectives, and the authors acknowledge that further work is needed to generalise the framework.