Abstract
The visual simplicity of an image (e.g. a real world scene or a pattern of blocks) is a characteristic we seem to quickly extract. This ability begs the question of which perceptual dimensions observers may use to quickly organize some pattern as a whole. In this work, we aim to propose an analytic definition of visual simplicity (and its counterpart, visual complexity) and test its relevance for perceptual organization. Simplicity is a “complex” perceptual dimension involving a set of perceptual properties (e.g. quantity of objects, symmetry, density). The current model focuses on implementing algorithms of these perceptual properties, and evaluating the visual simplicity of patterns by comparing a model's judgments of simplicity with those of human subjects. The results suggest that symmetry (i.e., the invariance of a pattern under a group of transformations), and lacunarity (i.e., the degree of spatial homogeneity and segregation of objects at different scales), are the most important properties underlying an analytic definition of simplicity.
M.L.M. was funded by a research assistantship from an NSF-IGERT training grant (DGE-0114378). This research was funded by a NIMH grant (1R03MH068322-1) awarded to A.O.