Abstract
Purpose:
To determine if the size of fixated images in near 3D displays affects the range of clear and single binocular vision (CSBV).
Methods:
20 binocular adults ages 18 to 28 years underwent a screening session to ensure they qualified for the experiment. Subjects were dilated with 2.5% phenylephrine to reduce depth of focus complications within and between conditions. The fixation target was a vertically-elongated maltese cross, viewed against a textured 3D background within an electronic mirror haploscope, with displays 40cm from the eyes. Target disparity was smoothly increased for convergence or decreased for divergence until the observer reported blur or diplopia. In three size conditions the target's size was fixed at 1.5mm x 4.5mm ("small"), 10mm x 30mm ("medium"), or 25mm x 74mm ("large"). In a fourth condition target size was modulated congruently with disparity ("loom") as is characteristic of 3D display content.
Results:
Separate repeated-measures ANOVAs of CSBV limit on target size were computed for convergence and divergence. In convergence, the mean limits of CSBV were 17.1Δ (small), 27.5Δ (medium), 28.9Δ (large), and 29.4Δ (loom). The overall effect of target size was significant in convergence (F=22.9, p<.000001), with most of the significance arising from the reduced CSBV limit for the small target. The CSBV limit for looming did not differ from the large target limit (t=.583, p=.57) or from the medium target limit. In divergence, the mean limits of CSBV were 10.4Δ (small), 14.5Δ (medium), 15.3Δ (large), and 12.3Δ (loom). The overall effect of target size was significant in divergence (F=8.2, p=.00012), with most of the significance related to small target size.
Conclusions:
The fixed target size conditions showed that larger size increases the range of CSBV in both convergence and divergence. The similarity of the looming convergence CSBV limit to the large target limit and of the looming divergence CSBV limit to the small target limit suggests that a major factor influencing the limits of CSBV in 3D displays is the size of the fixated target. Because the size of images in 3D displays depends entirely on the content creator, it is clear that a simple disparity limit standard for 3D content may not adequately protect viewers from blurred and/or double vision in 3D displays.
Keywords: 522 eye movements •
602 motion-3D •
434 binocular vision/stereopsis