Chromatic and achromatic monocular deprivation produce separable changes of eye dominance in adults

Abstract

Temporarily depriving one eye of its input, in whole or in part, results in a transient shift in eye dominance in human adults, with the patched eye becoming stronger and the unpatched eye weaker. However, little is known about the role of colour contrast in these behavioural changes. Here, we first show that the changes in eye dominance and contrast sensitivity induced by monocular eye patching affect colour and achromatic contrast sensitivity equally. We next use dichoptic movies, customized and filtered to stimulate the two eyes differentially. We show that a strong imbalance in achromatic contrast between the eyes, with no colour content, also produces similar, unselective shifts in eye dominance for both colour and achromatic contrast sensitivity. Interestingly, if this achromatic imbalance is paired with similar colour contrast in both eyes, the shift in eye dominance is selective, affecting achromatic but not chromatic contrast sensitivity and revealing a dissociation in eye dominance for colour and achromatic image content. On the other hand, a strong imbalance in chromatic contrast between the eyes, with no achromatic content, produces small, unselective changes in eye dominance, but if paired with similar achromatic contrast in both eyes, no changes occur. We conclude that perceptual changes in eye dominance are strongly driven by interocular imbalances in achromatic contrast, with colour contrast having a significant counter balancing effect. In the short term, eyes can have different dominances for achromatic and chromatic contrast, suggesting separate pathways at the site of these neuroplastic changes.

Keywords: colour vision; sensory eye dominance; short-term deprivation.

Figure 1.

Design and results for experiment 1. (a) The binocular phase combination task. Patching effect on sensory eye dominance was quantified by the change of binocularly perceived phase. The stimuli were two horizontal sine-wave gratings with equal and opposite phase-shift of 22.5° relative to the horizontal centre of the screen, which were dichoptically presented to the two eyes. The two gratings, either both chromatic or both achromatic, had a size of 90 pixels×180 pixels (1.045 degrees×2.09 degrees), and a spatial frequency of 1.0 cycle deg⁻¹. (Details of the stimuli and colour space are provided in the electronic supplementary material). Note that the stimuli we show here are only for illustration only, and the contrasts used for the study were not the same as the ones illustrated. (b) Binocular perceived phase was measured before and after the 2.5 h patching stage, started at multiple time points (0–60 min) after translucent patch removal. (c) The effect of 2.5 h translucent patching on binocular combination was measured with isoluminant red–green chromatic (RG-chromatic) gratings (circles) and achromatic gratings (squares). Error bars indicate the standard errors across five subjects.

Figure 2.

Design and results for experiment 2. (a) Monocular contrast threshold was measured before and after the 2.5 h patching stage that was started at several time points (in minutes) for three observers. (b) Contrast thresholds for isoluminant RG-chromatic (circles) and achromatic (squares) stimuli were normalized to the pre-patching baseline and plotted as a function of time sessions for three observers. Different symbols represent different conditions. Error bars indicate standard errors that were derived from a 500-time bootstrap procedure. (Online version in colour.)

Figure 3.

Design and results for experiment 3. (a) Binocular perceived phase was measured before and at several time points (in minutes) after watching the 2.5 h achromatic-only monocular deprivation (devoid of any matching chromatic content) at several time points (in minutes). (b) The effect of 2.5 h colour-only deprived movie watching on binocular combination was measured with RG-chromatic (squares) and achromatic (circles) stimuli. Error bars indicate the standard errors across five subjects. (Online version in colour.)

Figure 4.

Design and results for experiment 4. (a) Binocular perceived phase was measured before and at several time points (in minutes) after watching the 2.5 h luminance-deprived movie (but with balanced chromatic content) watching stage at several time points (in minutes). (b) The effect of 2.5 h unilateral achromatic (luminance) contrast-deprived movie watching on binocular combination was measured with RG-chromatic (circles) and achromatic (squares) stimuli. Error bars indicate the standard errors across five subjects.

Figure 5.

Design and results for experiment 5. (a) Binocular perceived phase was measured before and at several time points (in minutes) after watching the 2.5 h colour-only deprived movie (devoid of achromatic content). (b) The effect of 2.5 h colour-only deprived movie watching on binocular combination was measured with chromatic (circles) and achromatic (squares) stimuli. Error bars indicate the standard errors across five subjects.

Figure 6.

Design and results for experiment 6. (a) Binocular perceived phase was measured before and at several time points (in minutes) after watching the 2.5 h colour-deprived movie (but with balanced achromatic content) watching stage at several time points (in minutes). (b) The effect of 2.5 h monocular colour-deprived movie watching on binocular combination was measured with chromatic (circles) and achromatic (squares) stimuli. Error bars indicate the standard errors across five subjects.

Figure 7.

Summary of all the experiments in which changes in dominance over time were reported in experiments 1, 3, 4, 5 and 6. The eye dominance change is computed in terms of how magnitude of dominance change (quantified by perceived spatial phase) over time (phase × time) from 0 to 30 min after the completion of the period of deprivation. Dichoptic movies were used to provide the 2.5 h of monocular deprivation and the different types of deprivation are depicted above each panel in this figure. Error bars indicate the standard errors across subjects. *p < 0.05, two-tailed paired-samples t-test.