Abstract
Studies on attentional set and visual search have revealed that perception of a visual target is impaired by a temporally preceding distractor. The pattern of this impairment, known as the attentional capture, is contingent on the participant's attentional set. When searching narrowly for the specific target feature (the feature search mode), only the matching stimuli capture attention. When searching more broadly for an oddball (the singleton detection mode), any singleton captures attention. The present study examined which attentional set represents the “default” search mode. Unlike extant studies that provide participants with instructions to search for a specific target, the present study used operant conditioning to shape search behavior. In a training phase, participants receiving no instructions were presented with a red letter (target) embedded in a rapid stream of gray letters and heard a reward tone only when they hit a corresponding key. In this phase, observers could choose either the feature search or the singleton detection mode, because the target could be defined either as a red letter or as a singleton. In the test phase, the strategy used by observers was assessed by using peripheral distractors presented immediately before the target frame. The peripheral distractors contained one red (or green) and three gray symbols. The results showed that identification accuracy was impaired by both colors of distractors, indicating that participants adopted the singleton detection mode. Interestingly, approximately 90% of participants reported that they chose the feature search mode when asked explicitly afterwards. Identical patterns of results were obtained in follow-up spatial search experiments. These results suggest that the singleton detection mode represents a primary search mode. The dissociation between the behavioral results and the verbal reports implies that participants are not sensitive to the search mode they use, even though their choice of strategy seems highly conscious.