Computational Modeling of Implicit and Explicit Ensemble Perception

We used computational modeling to systematically evaluate hypotheses about the mechanisms that support both implicit and explicit forms of ensemble perception. Ensemble perception refers to the ability to summarize properties of a collection of objects in a visual array. Implicit ensemble perception tasks ask participants to judge if a test object belongs to a studied visual array, with a bias towards the central tendency of the array used as a measure of an ensemble representation. Explicit ensemble perception tasks ask participants to compare a test object with the mean of a studied array, with comparison performance used as a measure of a representation of ensemble statistics. Our modeling framework tested competing hypotheses about how objects are perceptually encoded (assuming subsampling, noisy encoding, or memory strength), how ensembles are represented (prototype or exemplar), and how decisions in both implicit and explicit ensemble tasks are made. Models assuming imperfect perceptual encoding and exemplar representations of an ensemble jointly accounted for a central tendency bias and poor recognition of individual objects in implicit ensemble tasks as well as set size and variance effects in explicit ensemble tasks. While prototype models could account for implicit ensemble tasks and set size effects in explicit ensemble tasks, they could not account for variance effects under reasonable parameterizations.