Visual Attention in 

Multiple Object Tracking (MOT)

 

Doran, M. M., &Hoffman, J.E (under revision). The Role of Visual Attention in Multiple Object Tracking: Evidence from ERPs.

    We examined the role of visual attention in MOT by measuring ERPs elicited by probes presented on targets, distractors, or areas of the background. The objects and background were composed of random dot textures in order to equate the physical change associated with object and background probes. In Experiment 1, observers tracked 2 out of 4 objects while maintaining central fixation and all probes were irrelevant to the observers since they did not require any behavioral response. In this case we found evidence of visual attentional selection in the N1 component. Note that the following are demonstration versions of these displays. The displays in the actual experiment were higher resolution, about twice as long, and the motion was smooth.

Experiment 1 Example Trial 1

Experiment 1 Example Trial 2

    In Experiment 2, participants performed the same MOT task as in Experiment 1. However, the probes were relevant since a subset of probe presentations required a behavioral response. These “response probes” were identical to the probes that did not require a response except that they had a small gray square in their centers. In this case, no evidence of selection was observed in the N1 component as if visual attention was diffusely allocated throughout the display in order to detect the probes. This suggests that visual attention may not be necessary for MOT –at least when tracking is fairly easy.

Experiment 2 Example Trial 1

Experiment 2 Example Trial 2

    In Experiment 3, we increased tracking difficulty by adding a target and distractor to the displays. Participants performed the same MOT and probe detection tasks as in Experiment 2 except that they tracked 3 out of six objects. In this case, selection was once again observed in the N1 component even though probes were still relevant. This suggests that early visual attention may aid tracking when MOT is difficult.

Experiment 3 Example Trial 1

Experiment 3 Example Trial 2