Exploring the Relationship Between Challenge and Reward for Game Based Learning

This pilot study investigates how dynamically adjusted difficulty and rewards affect primary school teachers' motivation when learning a computer programming game, using Keller's ARCS model (Attention, Relevance, Confidence, Satisfaction) as the evaluative framework. The authors argue that game-based learning depends on balancing challenge and reward, and that a game adapting its difficulty and rewards to a player's in-game score should sustain motivation more effectively than a traditional linear design with fixed difficulty and rewards.

The study used a small sample of ten primary school teachers with one to fifteen years of experience, split equally between an experimental group that played the dynamically adjusting game and a control group that played the fixed, linear version. Participants completed an online feedback questionnaire whose items were mapped to the four ARCS elements and scored on Likert scales. Group differences were assessed with one-tailed t-tests at a significance level of 0.01 with nine degrees of freedom.

Results favoured the experimental group across all four ARCS elements. Its overall mean score was 4 out of 5, compared with 2.225 for the control group, with element scores of 4, 3.9, 4.1 and 4 versus 2.5, 2, 2.4 and 2 respectively. The reported t-values (4.09 for Attention, 4.68 for Relevance, 4.44 for Confidence and 4.04 for Satisfaction) all exceeded the critical value, and the authors identify satisfaction as the most prominent effect. The authors acknowledge that the sample is very small, and the findings should be treated as preliminary, a work-in-progress indication rather than a generalisable conclusion.