Isomorphic Game Interaction Technique Patterns for Single Button: Definition and Evaluation

This work offers a pattern language for single-button interaction techniques in third-person humanoid-avatar games, addressing a gap in existing design pattern libraries, which have historically overlooked player interaction and action mapping. The authors derive eight isomorphic patterns by crossing three action characteristics, namely Effort, Duration, and Context, and illustrate each with examples from titles such as Darksiders 2 and Assassin's Creed 2. The framework is positioned as a reusable resource for game designers seeking consistent, low-complexity control schemes.

The central empirical contribution is a user study comparing nine candidate patterns across two prototypes, GCO and MWV, using a quantitative sensory evaluation. Eight of the nine patterns were confirmed by the results. The exception was the Quick Time Event pattern, which exhibited a 26% miss rate and a significance level of 0.130. Participants frequently gravitated toward lower-effort alternatives, such as Pump or Time Limited Hold, rather than the intended patterns, and regression analysis explained only about 41% of the variance in misses (adjusted R² = 0.4122), indicating that other factors also shaped performance.

A follow-up experiment with 46 volunteers, evenly split between those who missed and those who answered correctly, tested a rebuilt scene featuring longer event intervals, an additional event, and remodeled animations. Under these conditions, results favored the proposed single-button pattern, with a miss rate of 0.1555 and significance of 0.935 for the Defeat the Ghost King action. The authors conclude that inferring player behavior can help anticipate errors and improve enjoyment. Drawing on an inductive analysis of 54 games, with saturation reached around the twentieth title, they argue that the patterns are thematically neutral and broadly applicable, with occasional justified exceptions such as God of War's chest-opening mechanic.