Neural correlates of motion perception in the human visual brain

Introduction One of the most fascinating problems in visual neuroscience is finding a direct relationship between brain activity and perception. When a visual stimulus is presented to the eyes, it elicits a series of responses in many and different parts of the visual system, from the retina to the ’higher’ cortical areas, leading to a conscious visual percept. Dissociating which part of the visual brain activity is reflecting our perception is thus hard, since at the same time this activity is directly related to the processing of the visual stimulus itself. To try and answer this question, binocular rivalry has been used in the past, where the stimulus (which is different for each eye) remains constant but the perception alternates between the two rivalring monocular inputs. In this way one can dissociate the stimulus from the percept and, by studying the alternations in brain activation under such conditions, get an insight into which brain areas correlate their activity with what the subject actually perceives. Methods Binocular rivalry was used in fMRI experiments that were performed on a Siemens Trio 3T system. A different random dot kinematogram was shown to each eye, one consisting of red and the other of green dots. In one of the kinematograms 50 of the dots moved in the same direction producing a coherent motion signal whereas in the other all dots moved in random directions thus producing pure motion noise. As binocular rivalry developed between red dots in one eye and green dots in the other, subjects inside the scanner used two different buttons to report whether they perceived one the another color. In this way we could relate the BOLD signal we recorded in the magnet to the subjects’ percept and investigate how motion perception is reflected in the cortical activation of the various visual areas. Results Averaging the event-related time-courses across all subjects showed a range of different responses, with no significant effect in areas V1, V2 and V4, only a slight difference in area V3, and a much more clear difference in areas V3a, V5 and LOC (Fig. 1). Discussion In this study we were able to show that a number of visual areas are involved in motion perception. In general, the more involved an area is in motion processing, the more it is modulated by motion perception, supporting the idea that processing and perceptual areas are not distinct and separable, but rather the same areas are involved in both processing and perception