Abstract
The prevailing view of object-based attention holds that once any part of an object is attended, attention automatically “spreads” throughout the object. This view, which accounts for the same-object advantage in the two-rectangle spatial cueing paradigm (Egly et al., 1994), has been challenged recently (Shomstein & Yantis, 2002, 2004). We examined whether the effects of contingent attentional capture (Folk et al., 1992) are modulated by ignored distracters appearing on either a cued or uncued object. Displays contained a multicolored RSVP stream located at each of four rectangle ends. A central (Experiment-1) or peripheral (Experiment-2) 100% valid cue indicated the stream that would contain a red (green for half the subjects) target letter, which subjects identified. Two frames prior to target onset, a target-colored or nontarget-colored distracter could appear in one of two uncued locations (same or different rectangle). Robust contingent capture was observed in both experiments, but its magnitude did not differ for distracters on the same versus different objects. A control experiment ensured that our paradigm is capable of producing an object-based attention effect. Thus, when attention is directed with 100% certainty to a cued location, other locations on that object do not enjoy an attentional advantage. These data are consistent with priority-based theories of object-based attention (Shomstein & Yantis, 2002, 2004) in which the same-object advantage is a result of a predisposition to explore unattended parts of a currently-attended object when target location is uncertain. We conclude that object-based attention does not automatically spread throughout the object.