Abstract
Purpose: The response advantage for shifts of attention within compared to between objects is thought to reflect either an advantage for shifts within objects or a disadvantage for shifts between objects. We examined the object advantage in terms of engage and disengage operations of the object- and location-based attention systems by comparing conditions where attention shifted within, between, out of, and into objects, as well as between locations (loc), using one and two object displays. Responses were expected to be slowest for shifts between and out of objects because only they would require object-based attention to disengage from an object before shifting. Method: Objects were bars (Exp 1: vertical; Exps 2 & 3: diagonal). Cues and targets appeared inside (at a bar end) and outside objects (opposite the bar in one object displays) at one of four positions equidistant from fixation. On 20% of the trials no target appeared. When a target appeared, cues were valid on 60% of the trials and invalid on the rest. Of the six invalid cue conditions (names indicate where attention shifted from cue-to-target), two had two objects (within-obj & between-obj) and four had one object (within-obj, obj-to-loc, loc-to-obj, & loc-to-loc). Observers responded to target onset. Results: RTs were always faster on valid than invalid trials. RTs were the same for invalid conditions where object-based attention did not have to disengage from an object to shift (within-obj, loc-to-obj, & loc-to-loc). RTs were slowest, and no different from each other, for the invalid conditions where object-based attention had to disengage from an object before shifting (between-obj & obj-to-loc). Conclusions: Disengage operations associated with object-based attention are the primary cause for the object advantage in cuing studies. The object advantage is more accurately described as a disadvantage associated with object-based attention having to shift out of, or away from, an object.