Abstract
Decades of research has demonstrated that increased target knowledge improves search performance. However, most work has used pictorial or text cues that reference target features (i.e., positive target cues). More recently, work has examined the role of negative cues (Beck, Hollingsworth, & Luck, 2018), in which the distractor features you should avoid are cued, but the target can be determined from the given information. Previous work on negative cues have used simple stimuli and pictorial cues or specific text cues. We build on that work by comparing positive and negative cues (cue type) for pictorial cues and categorical text cues (cue specificity), using real-world objects. We predict that pictorial negative cues will prime the visual system and result in a cost to search performance relative to negative categorical cues or positive cue conditions. Search displays contained a target, a lure from the target category (which was cued on negative trials and uncued on positive trials), and four categorically unrelated distractors. Consistent with prior literature, we found two main effects indicating positive cues relative to negative cues, and pictorial cues relative to categorical cues, produce superior search performance across multiple eye movement and manual response measures (% trials target or lure is fixated first; target verification time; reaction time; and accuracy; all p< .001). We also observed an interaction of cue type and cue specificity (all measures p< .001), in which the benefit of pictorial cues was present, but greatly reduced on negative cue trials. Contrary to our hypothesis, we observed a pictorial benefit even on negative cue trials. This suggests that the target matching details of a pictorial negative cue are better at overcoming the capture of attention by a lure than a categorical negative cue.