Abstract
Observers can use explicit foreknowledge of a feature of an upcoming target to guide search. However, little is known about observers' use of explicit foreknowledge that a specific feature will not match the upcoming target. In a series of experiments, we presented observers with either “ignore” cues (cf. Munneke et al., 2008) that validly indicated the color of a nontarget item (rather than the color of the target item) in the upcoming display, or “neutral” cues. Surprisingly, observers were unable to use “ignore” cues to speed search; instead, knowing the color of a nontarget item on the upcoming display slowed search. This cost for “ignore” cues compared to neutral cues was consistent across several experiments using several different types of “ignore” cues. We conclude that observers are unable to explicitly avoid selecting items appearing in to-be-ignored colors. Instead, we propose observers use a strategy of immediately selecting the irrelevant item in order to subsequently inhibit it, a strategy we term “Search and Destroy”. In a follow-up experiment, we used a probe dot detection paradigm (cf. Kim & Cave, 1999) to determine whether observers were initially selecting the to-be-ignored item as predicted by the “Search and Destroy” hypothesis. Results confirmed this prediction; when probes were presented early in the search process, observers responded more quickly to probes appearing at the location of the to-be-ignored item compared to probes appearing at the remaining locations (453 ms vs. 485 ms, p <.05). This suggests that observers were selecting the to-be-ignored item prior to inhibiting it. We suggest that “Search and Destroy” may be a useful framework for interpreting other instances of inhibition in search shown in the previous literature, and that “Search and Destroy” may be an adaptive strategy under special conditions, such as when there are abundant items matching the to-be-ignored feature.