Abstract
Identifying the larger (or smaller) digit of a pair takes longer when its physical and numerical size are incongruent (e.g., physically large 2) compared to congruent (Henik & Tzelgov, 1982). In such number comparison tasks, the size congruity effect (SCE) is associated with judgments of either physical or numerical size. However, in visual search, the SCE is more reliable when participants search based on numerical compared to physical size, and occurs only when the target’s specific identity changes across trials (Sobel & Puri, 2018). Because the SCE is larger when targets vary in size and identity (or shape), we examined the degree to which selection history, or priming of attributes associated with recent targets (Awh et al., 2012), contributes to the SCE in visual search. Participants searched within digit arrays for small (2 or 3) or large (8 or 9) targets that were congruent or incongruent with respect to physical and numerical size, and reported the side of the screen containing the target. Reaction times (RTs), accuracy, and eye movements were recorded. RTs were longer for incongruent compared to congruent trials, replicating the SCE in visual search. Furthermore, longer time to first fixation (TFF) to incongruent targets suggests that delayed eye movements to targets on incongruent trials contribute to the RT effect. Regarding selection history, RTs were faster when the physical size of the current target matched that of the previous, but the SCE was similar across match conditions. With respect to shape, RTs and TFF were again faster for match compared to nonmatch trials, although this effect was driven by physically large targets. The SCE was present across match conditions, suggesting that in this task, the SCE does not rely on intertrial priming effects. Additional analyses will examine eye movements to distractors as a function of selection history.