Experience-driven suppression of irrelevant distractor locations is context dependent.

Humans can learn to attentionally suppress salient, irrelevant information when it consistently appears at a predictable location. While this ability confers behavioral benefits by reducing distraction, the full scope of its utility is unknown. As people locomote and/or shift between task contexts, known-to-be-irrelevant locations may change from moment to moment. Here we assessed a context-dependent account of learned suppression: can individuals flexibly update the locations they suppress, from trial to trial, as a function of task context? Participants searched for a shape target in displays that sometimes contained a salient, irrelevant color singleton distractor. When one scene category was presented in the background (e.g., forests), the distractor had a greater probability of appearing in one display location than the others; for another scene category (e.g., cities), we used a different high-probability location. Results in Experiments 1 and 2 (and in the Online Supplementary Material) failed to show any context-dependent suppression effects, consistent with earlier work. However, in Experiments 3 and 4, we reinforced the separation between task contexts by using distinct sets of shape and color stimuli as well as distinct kinds of reported features (line orientation vs. gap judgment). Results now showed robust task-dependent signatures of learned spatial suppression and did not appear to be tied to explicit awareness of the relationship between context and high-probability distractor location. Overall, these results reveal a mechanism of learned spatial suppression that is flexible and sensitive to task contexts, albeit one that requires sufficient processing of these contexts.