Attentional repulsion effect despite a colour-based control set

Abstract

A salient distractor can have a twofold effect on concurrent visual processes; it can both reduce the processing efficiency of the relevant target (e.g., increasing response time) and distort the spatial representation of the display (e.g., misperception of a target location). Previous work has shown that knowledge of the key feature of visual targets can eliminate the effect of salient distractors on processing efficiency. For instance, knowing that the target of interest is red (i.e., having an attentional control set for red) can eliminate the cost of green distractors on the speed of response to the target. The present study shows that the second mark of irrelevant salient distractors, i.e., distortions in spatial representation, is resistant to such top-down control. Using the attentional repulsion effect, we examined the influence of salient distractors on target localization. Observers had a colour-based control set and the distractors either matched or mismatched with the control set. In the first two experiments, we found systematic mislocalization of targets away from the peripheral distractors (i.e., an attentional repulsion effect). Critically, the effect was caused by distractors that both matched and mismatched the control set. A third experiment, using the same stimuli, found that processing efficiency was perfectly resistant to distractors that did not match the control set, consistent with previous work. Together, the present findings suggest that although top-down control can eliminate the cost of a salient distractor on processing efficiency, it does so without eliminating the distractor's influence on the spatial representation of the display.

Keywords:

• Attentional repulsion effect
• Space representation
• Visual attention

Acknowledgements

This work was supported by the Natural Sciences and Engineering Research Council of Canada in the form of a CGS scholarship to DGG and a Discovery Grant to JP. The authors would like to thank Charles Folk, Mei-Ching Lien, and two anonymous reviewers for their helpful suggestions in improving earlier versions of the manuscript. We also thank Nadine Jagdeo and Sasha Rennick for collecting the data.

Notes

¹Each trial involved presentation of a pair of cues along one diagonal (e.g., top-left and bottom-right), and since observer's task in the ARE paradigm is localization of the top Vernier line relative to the bottom line, it is assumed that capture of attention by the top (e.g., resulting in rightward repulsion of the above Vernier) or bottom (e.g., resulting in leftward repulsion of the bottom Vernier) cues would have the same effect on the spatial bias. We remain agnostic as to whether on each trial attention was captured by both peripheral cues or by one only. Note, however, that Suzuki and Cavanagh (1997, p. 445) found stronger ARE with double cues compared to single cues, which is consistent with a capture-by-both-cues view (cf. Yantis & Johnson, 1990).