Abstract
Segmentation cues, such as differences in binocular disparity (relative disparity) and motion direction (relative motion), make important contributions to perceiving figure-ground segmentation. These two cues strongly covary at object boundaries, so visual areas that carry signals defined by both could potentially combine them to support a more robust representation of objects and surfaces. Here we use functional MRI to directly compare activations to the two cues in a within-subject design (n=9). We compared responses to the appearance and disappearance of a central 2° disc-shaped region. In the relative motion paradigm, blocks comprising uniform field motion alternated with blocks in which the central-disk motion was in antiphase with that of the background (segmentation by relative motion). In the second paradigm, red-blue anaglyphic dynamic random-dot stereograms were presented such that in one block, the central disk-region popped out from background when the center and surround motion disparities changed in anti-phase (segmentation by relative disparity). In the other block, the disparities changed uniformly across the whole field. We compared activations within 16 topographically organized regions-of-interest (ROIs) defined for each participant using a recently developed probabilistic atlas (Wang et al., 2014). We found that TO1 (corresponding to MT; Amano et al., 2009) responded to the relative motion cue exclusively. A comparison between V3A and V3B revealed that the latter responded strongly to both the relative motion and disparity cues, while V3A responded to neither cue. Besides V3B, several other ROIs including LO1, LO2 and IPS0 had equally strong responses to both motion and disparity cues. By contrast, visual field maps on the ventral surface (VO1 and VO2) were not activated by either cue. These results suggest that the joint encoding of relative motion and relative disparity cues may involve multiple, dorsally located topographically organized visual areas.
Meeting abstract presented at VSS 2016