Abstract
Humans can rapidly estimate the statistical properties of a group of stimuli, including their average, variability and even more complex aspects, such as their distributions. Studies of Feature Distribution Learning (FDL) have, for instance, shown that participants rapidly learn the full shape of a distractor distribution and can use it to improve visual search performance: response times (RT) are faster if the target-defining feature lies outside the previous distractor distributions. FDL is surprisingly rapid, requiring only a few trials, and markedly sensitive to different distribution types (e.g., Gaussian versus uniform). It is unknown, however, whether our perceptual system encodes feature distributions automatically through passive exposure —i.e., in the absence of an attentional task. In two experiments, we sought to answer this question. Participants performed blocks of trials with an initial exposure stage followed by a single search trial. In the exposure stage, they passively saw a series of displays of 36 lines that included one singleton (an oddly oriented line, Experiment 1) or no singletons (Experiment 2). In the search trial display, they had to report the location of an oddly oriented target. The orientations of the lines were determined either by a Gaussian or a uniform distribution. To measure FDL, we parametrically varied the orientation distance between the search target and the center of the exposed distractor distribution. We found evidence for FDL when search efficiency was high (e.g., RT<1 second) and the display contained a singleton (Experiment 1). Under these conditions, RT decreased as a function of the orientation distance between the target and the exposed distractor distribution. These results suggest that FDL can occur by passive exposure, provided an involuntary and task-irrelevant singleton selection process during exposure. This argues that FDL is not bound to the attentional selection process involved in an active visual search.