Abstract
Visual attention can be allocated to an empty region of space (space-based attention) or to an individual object (object-based attention), even when that object overlaps with other, irrelevant, objects. What is the relationship between these two modes of attention? Previous research has suggested that both types of selection reflect a single underlying spatial attention mechanism that can either be allocated to a region of space, analogous to a spotlight, or can be “shrink-wrapped” to conform to object contours or surfaces. We evaluated this single-mechanism account by measuring the N1 component of the event-related brain potential (ERP), which is larger for flashes that appear at attended versus unattended locations. If the spatial attention system can conform to object boundaries, then object-based attention should produce a larger N1 component when a flash appears on an attended object compared to an unattended, overlapping object because the latter would be outside of the shrink-wrapped “attentional spotlight”. In the current task, participants were simultaneously presented with a pair of overlapping objects located to the left and to the right of a central fixation point. Individual objects were briefly flashed in a rapid, random, sequence while participants continuously attended to a designated object in order to respond to occasional target events. As expected, we found larger N1 amplitudes when the flash appeared on the attended object compared to a flashed unattended object located in the opposite visual field. Importantly, however, we also found larger N1 amplitudes for flashes of attended objects compared to unattended, overlapping objects. These results suggest that the spotlight of attention can be shrink-wrapped to the contours of attended objects, largely excluding irrelevant, overlapping objects.