Fractal analysis of pathfinding aesthetics

Abstract

We study a new, fractal-based model of pathfinding aesthetics for videogames and other virtual worlds. This model fills a gap in previous pathfinding efforts that have studied mostly machine performance issues or relied on anecdotal arguments rather than metrics to hypothesize about and improve aesthetic outcomes. We show firstly that the fractal model consistently discriminates between paths that were generated with beautifying treatments versus control paths. We also report that the model reliably predicts player expectations of relative aesthetic values for pathfinding. These conclusions are supported by statistical analysis of model results and opinion survey responses.