Normative data for an expanded set of stimuli for testing high-level influences on object perception: OMEFA-II.

We present normative data for an expanded set of stimuli designed to investigate past experience effects on object detection. The stimuli are vertically-elongated "bipartite" displays comprising two equal-area regions meeting at an articulated central border. When the central border is assigned to one side, a shaped figure (i.e., an object) is detected on that side. Participants viewing brief masked exposures typically detect figures more often on the critical side of Intact displays where a common ("familiar") object is depicted than on a matched critical side of Part-Rearranged (PR) displays comprising the same parts arranged in novel configurations. This pattern of results showed that past experience in the form of familiar configuration rather than familiar parts is a prior for figure assignment. Spurred by research implicating a network involving the perirhinal cortex of the medial temporal lobe in these familiar configuration effects, we enlarged the stimulus set from 24 to 48 base stimuli to increase its usefulness for behavioral, neuropsychological, and neuroimaging experiments. We measured the percentage of participants who agreed on a single interpretation for each side of Intact, Upright PR, and Inverted PR displays (144 displays; 288 sides) under long exposure conditions. High inter-subject agreement is taken to operationally define a familiar configuration. This new stimulus set is well-suited to investigate questions concerning how parts and wholes are integrated and how high- and low-level brain areas interact in object detection. This set also allows tests of predictions regarding cross-border competition in figure assignment and assessments of individual differences. The displays, their image statistics, and normative data are available online (https://osf.io/j9kz2/).

False pop out.

A single, unique target often pops out quickly and efficiently from a field of homogenous distractors in visual search. Pop out has helped shape theories of visual attention and feature integration as well as to identify basic features in human vision. Here we report a new phenomenon, false pop out, wherein one of the homogenous distractors competes with the singleton target to pop out, perhaps by breaking an overall grouping or pattern emerging from the display. We show the effect occurs with more than 1 type of stimulus, and we discuss the implications of such a counterintuitive finding for theories of visual search.