Abstract
Purpose :
To evaluate typical adult coherence thresholds for global stereopsis using random dot stereograms of corrugated surfaces containing noise.
Methods :
Stimuli were random dot stereograms that carried disparity-defined sinusoidal gratings (0.77 cycles/ degree) presented at each of 5 disparity levels (1.2, 2.5, 3.7, 5.0 or 7.4 minutes of arc). Noise dots were positioned with random disparities within the depth defined by the corrugated surface of the stereograms to vary the coherence of the stimulus. A VPixx® 3D stimulus system equipped with Nvidia 3D goggles was used for dichoptic presentation. Coherence thresholds for coarse grating orientation discrimination (vertical vs. horizontal vs. oblique) were obtained using a psychometric staircase procedure. Ten adults with normal binocular vision participated. For the 3.7 disparity stimulus, thresholds were re-tested with neutral density filters (0.6, 1.2 and 1.8) in front of the non-dominant eye to simulate abnormal binocular function.
Results :
The mean coherence required for disparity defined orientation discrimination was 43.0% (±11.6%). Coherence thresholds varied with stimulus disparity (ANOVA, p<0.001). Specifically, thresholds were significantly elevated for the smallest and largest disparity levels (p<0.001). Disrupting binocular vision using ND filters affected coherence thresholds (ANOVA, p<0.001). Thresholds increased with increasing filter density (35±10%, 44±18% and 61±14%, respectively).
Conclusions :
The coherence threshold for global form detection in random-dot stereograms with noise dots is a measure of stereopsis that is complementary to the conventional disparity threshold measure. Coherence stereo-thresholds are systematically affected by simulated disruption of binocularity and appear to reflect the quality of binocular vision.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.