Mission Profiles and Interaction Patterns in Gravitas: Evidence from a Simulation-Based Physics Learning Game

Abstract

Digital games and simulation-based environments are recognized as promising approaches for enhancing conceptual understanding and motivation in physics education, particularly with regard to persistent misconceptions in mechanics. Gravitas is a simulation-based learning game for lower-secondary students that employs a cyclical design of diagnosis, cognitive conflict, simulation, and reflection to address misconceptions about gravity and forces. In a quasi-experimental field study, 272 students selected one of four self-chosen mission profiles (Adventurer, Researcher, Tactician, Tinkerer). Log data and reflection-based performance measures were analyzed, with profiles serving as a categorical grouping variable. Interaction metrics (simulation time, reset frequency, use of conceptual supports, reflection duration) and reflection scores were normalized per completed level and examined using Kruskal-Wallis tests and Spearman correlations. Results revealed no significant differences between profiles in game progress, exploration intensity, support usage, reflection duration, or reflection performance (all ε² ≈ 0). Reset frequency showed a marginal profile effect (χ²(3) = 7.59, p = .055, ε² = .02) and a small negative correlation with reflection performance (ρ = −.18, p = .006). Overall, the findings indicate that self-selected mission profiles had limited explanatory value for learners’ interaction patterns and in-game conceptual performance in Gravitas. This suggests that adaptive design in simulation-based learning games may benefit from relying more strongly on dynamic, behavior-based indicators rather than on static self-categorizations.