Abstract
According to the Gestalt Principle of Common Fate, objects that move or change together tend to be perceived as a unified group. Here we examined the neural correlates of common fate in retinotopic areas in a standard block design (GE 1.5T, 25 slices, TR=2.5s, n=12; retinotopy in a separate session). Experiment 1: Four equiluminant green disks (equated individually for each subject) on a gray background, one in each quadrant, turned on and off at random, but with equal probability in each location. Flickering was either independent, or the disks flicked in unison. Local statistics within a quadrant were the same in the two conditions. The only difference was global correlation. Results: V1 demonstrated a negative BOLD response in the uncorrelated condition and a positive BOLD response in the correlated condition. Also, V2d and V2v displayed a greater BOLD response in the correlated condition than in the uncorrelated one. Thus global correlation appears to lead to greater neural activity than baseline, whereas uncorrelated flickering leads to inhibition of neural activity in V1 relative to baseline. Experiment 2: Same as experiment 1 except now the disks were always visible, and each randomly changed position within an imaginary circular zone within one quadrant, either independently or in unison. The former case looked like four separate objects in motion, and the latter case looked like a larger square jumping about jerkily. Results: V1 and V2d demonstrated a positive BOLD response in the uncorrelated condition and a negative BOLD response in the correlated condition. These results contrast with those of experiment 1, suggesting that V1 and V2 have different roles in determining grouping on the basis of spatial (motion-based) and temporal (flicker-based) correlation.