The mechanism of short-term monocular deprivation is not simple: separate effects on parallel and cross-oriented dichoptic masking

Abstract

Short-term deprivation of the input to one eye increases the strength of its influence on visual perception. This effect was first demonstrated using a binocular rivalry task. Incompatible stimuli are shown to the two eyes, and their competition for perceptual dominance is then measured. Further studies used a combination task, which measures the contribution of each eye to a fused percept. Both tasks show an effect of deprivation, but there have been inconsistencies between them. This suggests that the deprivation causes multiple effects. We used dichoptic masking to explore this possibility. We measured the contrast threshold for detecting a grating stimulus presented to the target eye. Thresholds were elevated when a parallel or cross-oriented grating mask was presented to the other eye. This masking effect was reduced by depriving the target eye for 150 minutes. We tested fourteen subjects with normal vision, and found individual differences in the magnitude of this reduction. Comparing the reduction found in each subject between the two masks (parallel vs. cross-oriented), we found no correlation. This indicates that there is not a single underlying effect of short-term monocular deprivation. Instead there are separate effects which can have different dependencies, and be probed by different tasks.

Figure 1

Panel A shows baseline detection thresholds averaged over 14 subjects. Coloured dots show data from individual subjects. Thresholds are given for the non-dominant (NDE) and dominant eyes (DE) without masking. Thresholds are also given for the dominant eye with parallel and cross-oriented 4% contrast masks in the other eye. Error bars show standard error of the mean. Panel B shows the relationship between the parallel and cross-oriented masking measured for each subject. The symbol for Subject 5 is obscured by that for Subject 14.

Figure 2

The effect of patching on the strength of dichoptic masking. Panel A shows the average threshold elevation from parallel and cross-oriented masks before and after patching. Coloured dots show data from individual subjects. In panel B the effects are plotted against each other for each subject, showing there to be no significant correlation.

Figure 3

Test-retest reliability of the masked thresholds following patching. Panel A shows the correlation between the threshold elevation (relative to the baseline threshold) introduced by the parallel mask in the first and second testing sessions. Panel B shows the correlation between the measurements from the two sessions for the cross-oriented mask. Subjects 3 and 8 are omitted from this figure as they each repeated the experiment three times rather than twice.

Figure 4

Panel A shows an example dichoptic pair of the stimuli used in this study. The two stimuli were presented separately to the two eyes using frame-interleaving. In this example, the target was a +45° oblique grating in the right eye. The dichoptic mask was a −45° oblique grating in the left eye. Panel B shows the sequence of stimuli shown in a single trial. In this example, the target was shown to the left eye in the second interval.

Figure 5

Sequence of conditions tested for the baseline and patching sessions. In the baseline sessions, the subjects were tested once on each condition in a random order. For the patching sessions, the subjects were tested on one of two fixed sequences of conditions.