The brain uses differences between the images from the two eyes to generate stereoscopic 3D perception. By looking at a cluttered scene with each eye alternately, we can easily appreciate horizontal disparities, which are a primary source of information about the relative depth of surfaces. However, we tend not to notice vertical disparities, which are much smaller, and whose role in perception is less well understood.

Dr Phil Duke, in collaboration with researchers at University of Pennsylvania, examined how the visual system uses vertical disparity information, and uncovered some important properties of stereo vision. To perceive the location and orientation of a surface relative to the head, we need to know where our eyes are pointing. Eye position signals provide this information, and in principle so can vertical disparities.

Dr Duke and colleagues found that the visual system uses vertical disparities to modify our eye position signals. On their own, eye position signals are prone to error. Dr Duke's work indicates that vertical disparities are used to maintain accuracy of our eye position signals, and thereby maintain accuracy of 3D perception.

Publications

• Duke PA, Oruç I, Qi H and Backus BT. (2006). Depth aftereffects mediated by vertical disparities: Evidence for vertical disparity driven calibration of extraretinal signals during stereopsis. Vision Research, 46, 228-241.