Purchase this article with an account.
Inna Tsirlin, Robert Allison, Laurie Wilcox; Decoding da Vinci: quantitative depth from monocular occlusions. Journal of Vision 2011;11(11):337. doi: https://doi.org/10.1167/11.11.337.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Nakayama and Shimojo (1990) demonstrated that quantitative depth percepts could be generated by monocular occlusions, a phenomenon they called da Vinci stereopsis. They used a configuration where a monocular bar was placed to one side of a binocular rectangle. When an occlusion interpretation was possible, the bar appeared behind the rectangle at a distance that increased as the lateral separation between the bar and the rectangle increased. Gillam, Cook and Blackburn (2003) argued that quantitative depth perception in da Vinci stereopsis was due to double-matching of the bar with the edge of the rectangle. They showed that when the monocular bar was replaced with a monocular dot only qualitative depth percepts remained. However, their stimulus differed from the original in ways that promoted double-matching and the range of separations of the monocular feature from the rectangle was different for the bar and the dot. To evaluate the contributions of monocular occlusions and double-matching to quantitative depth percepts in da Vinci arrangements, we have replicated and extended the Nakayama and Shimojo and Gillam et al. experiments. We reproduced the original stimuli precisely and used the same range of separations for the bar stimuli as for the dot stimuli. We also compared perceived depth from disparity in the bar and dot stimuli when they were presented binocularly. Three of six observers were able to see quantitative depth with the dot stimulus though less depth was perceived than when a monocular bar was used. Interestingly, we found a similar difference in perceived depth when the bar and the dot were presented binocularly. Taken together our results provide evidence that quantitative depth in da Vinci arrangements is based, at least in part, on monocular occlusions, and that this phenomenon depends on the properties of the monocular object and is subject to inter-observer differences.
This PDF is available to Subscribers Only