Abstract
The benefits of binocular vision have been debated throughout the history of vision science yet few studies have considered its contribution beyond a viewing distance of a few metres. What benefit, if any, does binocular vision confer for distance vision? Elementary texts commonly assert that stereopsis is ineffective beyond modest distances despite theoretical analysis suggesting a much larger effective range. We compared monocular and binocular performance on depth interval estimation and discrimination tasks at and beyond 4.5m.
Stimuli consisted of a combination of: 1) the reference stimulus, a smoothly finished wooden architectural panel, mounted upright, and facing the subject, 2) the test stimulus, a thin rod that could be precisely moved in depth and 3) a homogeneous background. An aperture prevented view of the top and bottom of the stimuli. Subjects made verbal, signed estimates in cm of the depth between the test and reference stimuli. On each trial, the depth was set between ±100cm. Observers viewed the displays either monocularly or binocularly from 4.5, 9.0 or 18.0m. Depth discrimination at 9.0 m was also evaluated using adaptive staircase procedures.
Regression analysis provided measures of the scaling between perceived depth and actual depth (the ‘gain’) and the precision. Under monocular conditions, perceived depth was significantly compressed. Binocular depth estimates were much nearer to veridical although also compressed. Both raw precision measures and those normalized by the gain were much smaller for binocular compared to monocular conditions (ratios between 2.1 and 48).
We confirm that stereopsis supports reliable depth discriminations beyond typical laboratory distances. Furthermore, binocular vision can significantly improve both the accuracy and precision of depth estimation to at least 18m. We will discuss additional experiments to extend these results to larger viewing distances and to evaluate the contribution of stereopsis under rich cue conditions.
The support of an ARC International Fellowship (LX0667316), an ARC Discovery Grant (RM00362), and an NSERC Discovery Grant are greatly appreciated.