Jordyn R. Wright¹ (), Sarah E. Purnell¹, Daniel D. Thayer¹, Thomas C. Sprague¹; ¹University of California, Santa Barbara

Statistical regularities in the environment can shape attentional priority in space (Theeuwes, 2022). For example, attentional selection of a target is enhanced when it appears at high probability locations (HPL), while interference from distractors is reduced when they are presented at HPLs (Ferrante et al., 2018). Although these statistical regularities can be learned even when task-irrelevant or without explicit awareness (Duncan & Theeuwes, 2020), some degree of top-down or bottom-up attention to stimuli containing these regularities may be necessary for learning to occur (Duncan et. al, 2025). Here, we tested whether learning statistical regularities depends on the top-down allocation of spatial attention to relevant locations (Golan et al., 2024). Participants covertly searched an 8-item array for a uniquely shaped target and indicated the orientation of a line within the target (horizontal or vertical). On most trials, a central cue indicated the spatially relevant hemifield of the search array (left or right; 4 of 8 items) in which the target would always appear. Half of all trials contained a task-irrelevant color singleton distractor, which could appear within or outside the cued hemifield. Critically, the distractor appeared more often at two HPLs within the array (one in each hemifield; left and right). Statistical regularities of distractor location interacted with spatial cueing to reduce interference from salient distractors, however, results varied depending on the spatial relevance of the HPL. These results suggest that learning distractor-location regularities may depend on whether those regularities are spatially relevant, and support future research on the flexibility of learning environmental regularities to guide visual search.

Acknowledgements: Funding: National Eye Institute R01-EY035300