Abstract
The conventional view of binocular vision is that the visual input from the two eyes is fused to produce the percept of a single scene. In doing so, the stereoscopic system also provides extremely high-resolution information about the relative depth of objects in space (fine stereopsis). However, reliable stereoscopic depth is also obtained from images with very large disparities that cannot be fused into a single image (coarse stereopsis). While there is some evidence that stereoacuity improves with age, very little is known about the development of coarse stereopsis.
We compared performance in children (5–6, 11–12 years) and adults on computerized tests of fine and coarse stereopsis. Stereoscopic stimuli were presented using liquid crystal shutter glasses. On each trial, cartoon characters were presented simultaneously above and below a fixation marker. The stimuli were displaced in depth by equal amounts in opposite directions and the observer's task was to indicate which character appeared to be closer in depth. We assessed perceived depth for a set of fine (0.02, 0.08, 0.17, 0.33, 0.67 degrees) and a set of coarse (1, 2, 2.5, 3, 3.5 degrees) disparities.
The younger children were above 80% accuracy at all disparities. Performance was poorest at the smallest disparity, and they showed a shallow decline in accuracy with increasing disparity in the coarse range. Surprisingly, for both the adults and older children, accuracy was consistently lower (58% to 71%) at all test disparities. In control experiments we have determined that the decline in performance with age is not due to motivation, changes in interpupillary distance, visual hemispheric asymmetries or the slant of the horoptor. We will discuss these results in terms of a narrowing of Panum's area (Dowd et al., 1980) and the disruptive effect of vergence eye movements with age.
Funded by the Natural Science and Engineering Research Council of Canada.