The perception of second-order motion

In this thesis the notion of an independent non-linear channel for the perception of second- order motion is investigated. An examination of speed discrimination thresholds for first- and second-order bars and edges showed no differences in the patterns of response over changes in the temporal and spatial parameters of the stimuli. The higher thresholds for second-order stimuli may be accounted for by appealing to the properties of their noise carriers. In a study of the direction of motion in reversed-phi stimuli, it was shown that luminance and contrast defined stimuli could elicit both forwards and reversed motion. The forwards motion in the contrast defined stimulus cannot be explained by the operation of a first- or second-order channel. Perception of motion direction in the contrast defined stimulus was dependent upon the characteristics of the noise carrier. Similar dependencies were observed when noise was added to the first-order stimulus. When the effect of carrier is taken into account, both types of stimulus show similar patterns of response over spatial and temporal frequency. Modulation depth tuning curves for the detection of motion direction in stimuli where motion was defined by contrast and luminance microstructure were also investigated. Luminance microstructure can affect perceived contrast in contrast defined motion and also in static noise pattems. This implies that some early non-linearity exists in the human visual system. The filtering and rectification approach to recovery of the second-order motion should be highly effecfive at recovering the modulant. However, an estimate of the size of the non-linearity shows that only a relatively small distortion was necessary to account for the modulation depth tuning curves. The results from this thesis indicate that the carrier is crucial to the perception of second- order motion. Differences in response to first- and second-order motion may depend upon properties of the stimuli rather than the operation of separate mechanisms. It is argued that the results cast some doubt over the notion of a second-order channel. A number of alternatives are discussed