Background: The simultaneous tracking and identification of multiple moving objects encountered in everyday life requires one to correctly bind identities to objects. In the present study, we investigated the role of spatial configuration made by multiple targets when observers are asked to track multiple moving objects with distinct identities.
Methodology/principal findings: The overall spatial configuration made by the targets was manipulated: In the constant condition, the configuration remained as a virtual convex polygon throughout the tracking, and in the collapsed condition, one of the moving targets (critical target) crossed over an edge of the virtual polygon during tracking, destroying it. Identification performance was higher when the configuration remained intact than when it collapsed (Experiments 1a, 1b, and 2). Moreover, destroying the configuration affected the allocation of dynamic attention: the critical target captured more attention than did the other targets. However, observers were worse at identifying the critical target and were more likely to confuse it with the targets that formed the virtual crossed edge (Experiments 3-5). Experiment 6 further showed that the visual system constructs an overall configuration only by using the targets (and not the distractors); identification performance was not affected by whether the distractor violated the spatial configuration.
Conclusions/significance: In sum, these results suggest that the visual system may integrate targets (but not distractors) into a spatial configuration during multiple identity tracking, which affects the distribution of dynamic attention and the updating of identity-location binding.