Abstract
Multiple object tracking (MOT) is widely used for studying sustained attention in dynamic environments. Previous research reported inter-target grouping based on feature similarity (e.g., targets sharing the same color or shape) can facilitate tracking. A recent study also found that symmetric relation among targets in MOT tasks can automatically improve tracking performance. And an additive grouping effect of symmetry and feature similarity was observed, but two feature-based groupings were not additive. This study used functional magnetic resonance imaging (fMRI) to identify brain areas involved in feature-based and symmetry-based groupings in MOT. The present study included five conditions of inter-target groupings based on different grouping cues or their combinations: no grouping, symmetry-based grouping, symmetry-color-based grouping, color-based grouping, and color-shape-based grouping. Comparing the symmetry-based grouping condition with the no grouping condition revealed significant differences in anterior cingulate, the middle temporal gyri, inferior temporal gyri, postcentral gyri, superior frontal gyri medial, right lingual gyrus, right calcarine, and cuneus. Comparing the color-based grouping with the no grouping revealed significant differences in temporoparietal junction (TPJ), inferior parietal sulcus, putamen, middle and superior frontal gyrus, and anterior cingulate. The results of our study showed that putamen and temporoparietal junction (pTPJ) were involved in feature-based groupings, and middle temporal gyri, inferior temporal gyri and the cuneus were associated with symmetry-based grouping. The putamen is activated in all feature-based groupings, which indicated that when targets and distractors process different feature, tracking is facilitated by keeping the targets feature and ignoring the distractors feature. Our findings indicate that in feature-based inter-target grouping conditions, attention is more easily oriented and maintained towards targets during tracking. And in symmetry-based grouping condition, attention is less divided, and symmetric dynamic scene made up of targets is less complex to process.
Meeting abstract presented at VSS 2017