Limitations on Animacy Categorization in Ensemble Perception

People can rapidly and efficiently categorize the animacy of natural objects. Does this imply that visual system has an unlimited capacity for processing across entire field? To test this, we generated a set morphed “animacy continua” between animate inanimate silhouettes. Observers were shown different ratios these items (2:1, 10:3) judged whether was more animate/inanimate. The distribution either “segmentable” (including only 100% animate/inanimate items) or “non-segmentable” (50% ambiguous morphs also presented). In Experiments 1 2 (N=61), observers failed to integrate from multiple items, as they showed very poor performance not sensitive type, despite almost perfect ability individual same manipulation with color category-defining feature elicited both good ensemble categorization performances segmentability effect. Experiment 3, parametrically manipulated discriminability in ensure our results based on discriminability. We always presented distributions 2:1 numeric ratio used two levels (easy hard – 30˚ 15˚ apart along CIELab wheel) three (hard objects, easy but non-morphed original silhouettes real objects). again bad at even high rather task low (although perfectly accurate features). Thus, alone cannot explain differences dimensions. Overall, conclude does allow people build strong representations, which suggests limited perception.