Abstract
Stereo vision has a well-known anisotropy: at low frequencies, horizontally-oriented sinusoidal depth corrugations are easier to detect than vertically-oriented corrugations (where both are defined by horizontal disparity). Serrano-Pedraza & Read (2010, Journal of Vision, 10(12)) suggested that this stereo anisotropy may arise because the stereo system uses multiple spatial-frequency disparity channels for detecting horizontally-oriented modulations, but only one for vertically-oriented modulations. In this work we tested this hypothesis using the power-spectrum model of visual masking (imported from audition and luminance studies) in order to examine whether there are one or several channels, and to measure the bandwidth of channels. In the first experiment (4 subjects), we used Bayesian adaptive staircases to measure disparity thresholds for horizontal and vertical sinusoids near the peak of the disparity sensitivity function (0.4c/deg), both in the presence of white noise with five different power levels, and in the presence of notched noise with 6 different bandwidths. The use of notched noise avoids off-frequency looking. The masking linear model fit our results assuming a channel centered on 0.4c/deg with bandwidths of 2.95 octaves for horizontal corrugations and 2.62 octaves for vertical corrugations. In our second experiment (8 subjects), we measured disparity thresholds for horizontal and vertical sinusoids of 0.1c/deg in the presence of band-pass noise centered on 0.4c/deg with a bandwidth of 0.5 octaves. This mask had no effect on the disparity threshold at 0.1c/deg, for either horizontal or vertical corrugations. We ran the power-spectrum model assuming two types of detection, single channel detection and multiple channel detection. The multiple-channel model fitted better for both horizontal and vertical corrugations. We conclude that, contradicting our earlier hypothesis, the human stereo system must contain (at least) two channels for detecting horizontally-oriented and vertically-oriented depth modulations and that the channels at 0.1 and 0.4c/deg must be operating almost totally independently.
Meeting abstract presented at VSS 2013