English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Poster

Seeing in 3D: human psychophysics, modelling and brain imaging

MPS-Authors
/persons/resource/persons84307

Welchman,  A
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

/persons/resource/persons214722

Fleming,  R
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;
Max Planck Institute for Biological Cybernetics, Max Planck Society;

Fulltext (public)
There are no public fulltexts available
Supplementary Material (public)
There is no public supplementary material available
Citation

Welchman, A., Muryy, A., Ban, H., & Fleming, R. (2011). Seeing in 3D: human psychophysics, modelling and brain imaging. Poster presented at AVA/BMVA Spring Meeting 2011 (AGM), Cardiff, UK.


Cite as: http://hdl.handle.net/21.11116/0000-0002-4C20-5
Abstract
Successful behaviour relies on reliable estimates of the three-dimensional structure of the environment, facilitating recognition and interaction. Here we review recent work that seeks to understand the neural mechanisms that achieve 3D vision. We focus on the perceptual use of binocular disparity – considering both the computational principles that guide perception and the neural implementation that achieves it. First, we cover work that examines the disparities produced when viewing reflective (specular) objects binocularly. Typically, this produces a considerable discrepancy between the distance specified by disparity (the adanaclastic surface) and the physical surface of the object. We measure perceptual judgements of 3D shape when viewing these shapes, and find that the human visual system tempers its use of disparity signals depending on the structure of the adanaclastic surface. Second, we review neuroimaging work that identifies the neural circuits that process disparity to support depth perception. This work uses high-resolution fMRI combined with machine-learning analysis. In particular, we make parametric stimulus manipulations and test the degree to which different visual areas contain information about the viewed stimulus. This is quantified by the accuracy of a support vector machine in predicting the viewed stimulus from patterns of brain activity. We show that higher portions of both the dorsal and ventral visual pathways process perceptually-relevant disparity signals. However, the type of representation differs between pathways – dorsal responses relate to the metric depth structure, while ventral areas represent depth configurations (sign of depth rather than magnitude).