Simple shapes guide visual attention based on their global outline or global orientation contingent on search goals.

Abstract

It is still unclear which features of a two-dimensional shape (e.g., triangle, square) can efficiently guide visual attention. Possible guiding features are edge orientations (single oriented shape edges; e.g., verticals during search for squares), global outlines (combination of the target edges; e.g., squares), or global orientations (specific orientations of global outlines; e.g., squares but not diamonds). Using a contingent-capture protocol, we found evidence for task-dependent guidance by the global shape outline and the global shape orientation. First, if participants searched for a shape (an equilateral triangle) independent of its pointing direction, cues with the same global shape outline as the target captured attention, even without sharing any edge orientations with the target. Second, however, if a shape's specific pointing direction was task-relevant, attentional guidance changed to the specific orientation of the global shape. Our results show that the global shape outline and the global shape orientation can both guide visual attention, contingent on the nature of the shape and the current search goals. We discuss differences between shapes (equilateral triangles and isosceles trapezoids) considering models of shape perception and conclude with a critical review of the contingent-capture protocol as a complementary method to visual search protocols. Public Significance StatementThis study shows that when we search for a two-dimensional object, its shape can be used to guide our visual attention so that we can efficiently find the object. In contrast, oriented edges of such a two-dimensional object are not sufficient to explain successful search for shapes. Whether the orientation of a shape is also used to guide attention depends on the shape itself and on the necessities imposed by the search context.

Little engagement of attention by salient distractors defined in a different dimension or modality to the visual search target

“…But then, proponents of this account would have to explain why it was not possible to suppress the latter type of distractor. For instance, why was it not possible to set up a negative template for “triangle” shapes, even though triangles are separable from squares based on possessing unique (oblique) side orientations (Buetti et al, 2019; Grüner et al, 2021; Wolfe & Horowitz, 2004, 2017; Xu et al, 2021). A likely explanation would have to involve assumptions similar to those central to the DWA/MWA, namely: the handling of intra‐dimensional distractors is inherently more difficult than the handling of cross‐dimension or cross‐modal distractors.…”

Top-down knowledge surpasses selection history in influencing attentional guidance

“…O’Keefe, 2007 ). The rational for using simulation to estimate achieved power is described in more detail in the appendix of Grüner et al ( 2021 ). In all experiments, we simulated reaction times from the actual reaction time distribution in each experiment.…”

Top-down suppression of negative features applies flexibly contingent on visual search goals

“…In addition, for each experiment, we used simulations (Forstinger & Ansorge, 2023 ; Grüner et al, 2021 ) to estimate the statistical power to detect significant validity effects in both the negative and the positive search tasks. Specifically, we aimed to simultaneously detect an inverse validity effect of −25 ms in the negative search task and a standard validity effect of 25 ms in the positive search task.…”