Peripheral prism glasses: effects of moving and stationary backgrounds

Abstract

Purpose: Unilateral peripheral prisms for homonymous hemianopia (HH) expand the visual field through peripheral binocular visual confusion, a stimulus for binocular rivalry that could lead to reduced predominance and partial suppression of the prism image, thereby limiting device functionality. Using natural-scene images and motion videos, we evaluated whether detection was reduced in binocular compared with monocular viewing.

Methods: Detection rates of nine participants with HH or quadranopia and normal binocularity wearing peripheral prisms were determined for static checkerboard perimetry targets briefly presented in the prism expansion area and the seeing hemifield. Perimetry was conducted under monocular and binocular viewing with targets presented over videos of real-world driving scenes and still frame images derived from those videos.

Results: With unilateral prisms, detection rates in the prism expansion area were significantly lower in binocular than in monocular (prism eye) viewing on the motion background (medians, 13 and 58%, respectively, p = 0.008) but not the still frame background (medians, 63 and 68%, p = 0.123). When the stimulus for binocular rivalry was reduced by fitting prisms bilaterally in one HH and one normally sighted subject with simulated HH, prism-area detection rates on the motion background were not significantly different (p > 0.6) in binocular and monocular viewing.

Conclusions: Conflicting binocular motion appears to be a stimulus for reduced predominance of the prism image in binocular viewing when using unilateral peripheral prisms. However, the effect was only found for relatively small targets. Further testing is needed to determine the extent to which this phenomenon might affect the functionality of unilateral peripheral prisms in more real-world situations.

Figure 1

Press-on 40 Fresnel peripheral prisms in the oblique configuration on the left lens as fitted for the study to a patient with left hemianopia. The upper prism is base-out and -down, and the lower prism base-out and -up, shown with 10mm inter-prism separation, providing expansion in the central area of the visual field used when driving (Figure 3). The effects of the prisms are notable by the apparent shift of the upper and lower lid margins and iris (imaged by the prism segments). The wearer has an uninterrupted binocular view through the central prism-free area of the lens. A color version of this figure is available online at www.optvissci.com.

Figure 2

A still frame image presented to participants during testing with bipolar (black and white) central fixation cross and a peripheral bipolar (black and white) 1.4° checkerboard target (shown here at high 95% contrast). Only the central portion of the screen is shown. Still images and videos were displayed in color in the study.

Figure 3

Binocular central visual field plot and static detection test zones for a patient with left homonymous hemianopia fitted with unilateral oblique 40Δ peripheral prisms. Thick black lines mark the kinetic isopter. The outer boundaries of the seeing hemifield are limited by the display screen to 34° vertical by 40° horizontal eccentricity. Black-filled rectangles represent test zones in the lower prism expansion area and the seeing hemifield (20 targets per zone). Open diamonds represent arbitrary positions of additional targets (16) included to prevent anticipation of a target’s location. Black-filled circles outside the seeing hemifield represent catch trial targets (8). Light-gray shaded areas within the seeing hemifield illustrate the optical apical scotomas, which are compensated for by the non-prism eye in binocular viewing.

Figure 4

Detection rates for each participant under binocular and monocular viewing on the still frame background: (A) prism expansion area test zone and (B) seeing hemifield test zone. Detection rates were not significantly different in binocular and monocular viewing in the prism expansion area, but were slightly higher in binocular viewing in the seeing hemifield (p = 0.046). Legend includes target diameter and contrast for each subject. S9 wore 57Δ prisms; all other subjects wore 40Δ prisms.

Figure 5

Detection rates for each participant under binocular and monocular viewing on the motion video background: (A) prism expansion area test zone and (B) seeing hemifield test zone. Detection rates were significantly lower (p = 0.008) in binocular than monocular viewing in the prism expansion area (points all below the diagonal) but not the seeing hemifield. Legend includes target diameter and contrast for each subject. S9 wore 57 Δ prisms; all other subjects wore 40Δ prisms.

Figure 6

Detection rates in the prism-expansion area test zone over the video background for bilateral and unilateral prism fittings. (A) Participant with left hemianopia (S6 in Figures 4 and 5) and (B) Normally-sighted participant with simulated left hemianopia. For both subjects, detection rates were as good in binocular viewing as in monocular viewing when binocular rivalry was reduced with the bilateral fit, but were lower in the condition with a stimulus for rivalry (binocular unilateral fit). For each condition, there were 40 presentations in the prism expansion zone. Error bars are 95% confidence limits.