Abstract
We found large differences in people's abilities to see motion in depth. Binocular motion in depth perception can theoretically be derived in two different ways: namely, based on changing disparity over time (CDOT), or based on inter-ocular velocity differences (IOVD). In the former, disparity is calculated first and then a change over time is monitored. In the latter, velocity is measured in each eye first and then the difference is taken. In three experiments we measured the abilities of 63 participants to see motion in depth from these two cues, both in isolation and in combination. Thirteen people could not see motion in depth in IOVD stimuli, but who were able to see motion in depth from CDOT stimuli; twelve participants were able to see motion in depth from IOVD stimuli, but not from CDOT. Thirteen people did not see motion in depth at all, and 21 people could see motion in depth from all stimuli. There were significant correlations in detection thresholds between CDOT stimuli and stimuli containing both cues, and between CDOT and IOVD, but not between IOVD and both cues. We performed a range of vision tests to control for possible effects of visual or motivational factors in the study. We also did a texture discrimination experiment with a similar design and stimulus as in the main experiments. Texture discrimination thresholds for those who could see motion in depth were not different from those who could not. The frequent double dissociation leads us to conclude that CDOT and IOVD mechanisms can exist independently of each other. Furthermore, motion in depth perception is achieved in fundamentally different ways in different people. The latter finding has widespread implications for our understanding of perception in general.
Funded by EPSRC and Wellcome Trust.