Abstract
How fast are we to detect a particular make of car when shortly before we spotted a different-colored car of the same make or a same-colored car of a different make? Would another category of vehicle (a plane) affect detection of a car? Found and Müller (1996) investigated feature- and dimension-based inter-trial effects in “priming of pop-out” (Maljkovic & Nakayama, 1994) visual-search tasks for simple objects (red vertical or green tilted bar target among green vertical distractors). Analyzing for the cross-trial repetition versus change effects, they found a significant effect only for target dimension changes (e.g., from color to orientation), but not for feature changes (e.g., from red to blue). Their results suggest that inter-trial priming is primarily dimension-based. The present study was designed to investigate the pattern of inter-trial effects with pictures of real stimuli. Observers had to search for a target that was “not a fruit” (car, plane, or bird) among fruit distractors. Four different types of inter-trial relations between targets on successive target-present trials were examined. If the target on trial N was a car of a particular make, the target on trial N-1 could be the same car (identical object), another make (same category), a plane (same superordinate category: “vehicle”), or a bird (different superordinate category: “animal”). Change costs were calculated by comparing RTs for the identical object condition minus RTs for the three different change conditions (same category, same superordinate category, different superordinate category). The largest costs were observed if the superordinate category changed, smaller change costs were found for category changes within the superordinate category; by contrast, no costs occurred if targets changed within categories. Thus, category- and feature-changes in complex objects seem to produce similar inter-trial effects to dimension- and feature-changes in simple objects.
ER and JK were supported by the Swiss National Science Foundation grant project 100014-130252/1, MZ and HJM by the German Cluster of Excellence Cognition for Technical Systems grant project EC 142 and HJM by the DFG Research Group grant project FOR480.