Abstract
Visual clutter holds great promise as a surrogate measure of set size effects in scenes, but just as all search objects are not equally distracting, clutter too may take different forms. We compared the effects of subjective clutter (determined by independent raters) and objective clutter (as quantified by edge count) on visual search using “evolving” scenes that varied systematically in clutter, yet maintained their semantic continuity. Observers searched for a visually-previewed target building in rural, suburban, and urban city scenes created using the game SimCity. Stimuli were 30 screenshots obtained for each scene type as the city evolved over time (e.g., urban scene-1 was a field with some roads; urban scene-30 was a bustling metropolis). Both subjective and objective clutter estimates were highest for urban scenes, intermediate for suburban scenes, and lowest for rural scenes. These identical relationships characterized the effect of clutter on RTs and search guidance (measured by scanpath ratio); RTs and guidance were fastest/strongest for rural scenes, slower/weaker for suburban scenes, and slowest/weakest for urban scenes. Importantly, subjective and objective within-city clutter estimates also increased as each city matured, but the effects of these clutter changes on search depended on the type of estimate. RTs and guidance correlated highly with subjective within-city clutter (as clutter increased, RTs were longer and gaze followed a less direct route to the target; R2 ranging from .52–.70), but within-city correlations between objective clutter and RTs (R2 .10–.31) and guidance (R2 .34–.40) were weaker. The fact that subjective estimates better predicted search than objective estimates suggests that within-city clutter may not be explained by low-level feature congestion alone; conceptual congestion (e.g., the number of different types of buildings in our scenes), part of the subjective clutter measure, may also be important in determining the effects of clutter on search.