Abstract
Contextual cues guide attention towards targets. Can context also be used to prevent the capture of attention from salient but irrelevant distractors? In two experiments, we explored this question using the additional singleton paradigm. On each trial, a background scene (a city or forest) presented shortly before stimulus onset indicated whether there was a high (80%) or low (20%) likelihood that a salient distractor would appear on the upcoming trial. The location of the distractor, if it was present, was unpredictable across all conditions, and the color of the distractor was fixed. Results revealed a significant reduction in attentional capture — in measures of both reaction time and error rates — when the background cued a high versus low probability of distractor presence. In a second experiment, we modified the color of the distractor such that it was unpredictable from one trial to the next. Still, we replicated the results of the first experiment — when the background indicated a high probability of distractor presence, attentional capture was reduced. Taken together, these data suggest that when participants could anticipate that a distractor was going to appear because of contextual cues, they were able to reduce the impact of that distractor. Participants were largely unaware of the relationship between background and distractor probability, indicating that this is largely an implicit learning process. Furthermore, these cues were randomly intermixed, meaning that selection history effects cannot explain the reduction of capture in the high probability condition. These data support the idea that implicit contextual learning can facilitate the suppression of salient distractors. Furthermore, this contextual suppression effect does not require the location or color of the distractor to be fixed. Instead, our results suggest that contextual cues can trigger second-order suppression (e.g., Won et al., 2019) in visual search.