Abstract
Attention operates not only over spatial locations but also over perceptual objects. One of the hallmarks of object-based attention is that following an invalid peripheral cue, perceptual judgments are more efficient when the target appears in the same object as the cue versus in a different object. However, the manner in which attention is oriented to and within objects is not well understood. We investigated a core issue in object-based selection: the spatial distribution of attention within an object. We manipulated the within-object distance between the cue and target. Evidence of spatial distance effects within objects would support grouped-array theories of object-based attention (Vecera, 1994), in which object perceptual structure serves to constrain the spatial distribution of attention.
We modified the object-based task developed by Egly et al. (1994), using more realistic, three-dimensional stimuli with cues and targets intrinsic to the objects. Two curved, “tube” objects were used so that all points on each object were equally distant from fixation. In the first experiment, we replicated the Egly et al. same-object detection advantage. In the second experiment, we manipulated the distance between cue and target within an object. On each trial, the target appeared at one of three possible locations: at the cued location, at a near location within the cued object, or at a far location within the cued object. Targets were detected more efficiently at the cued location than at either of the uncued locations. In addition, targets were detected more efficiently at the near location within the object than at the far location within the object. These results demonstrate that object-based selection is inherently spatial and that attention has a spatial profile across the extent of an attended object. These results support grouped array models of object-based attention.