A comparison of the effects of active and passive image-displacements on visual thresholds

Abstract

The effects on visual thresholds of active displacement of the peripheral retinal image (produced by saccadic eye movement) and passive displacement of the peripheral retinal image (produced by field rotation) were compared. The purpose of this comparison was to investigate whether it is necessary to invoke a saccade-locked central inhibition of visual sensitivity in order to account for the elevation of threshold associated with saccadic eye movement.
It was found that active image displacements of six degrees extent were associated with a mean threshold elevation of 0.5 log units, whereas closely similar passive displacements of the same retinal image were associated with a mean threshold elevation of 0.3 log units.
It is possible that some of the difference between the threshold elevation associated with active and passive displacement of the retinal image may have been produced by a small difference between the image displacement in the extreme periphery under the two conditions. Whether this is so or not, the maximum threshold elevation associated with a saccade rather than with the retinal image displacement caused by the saccade, was shown to be 0.2 log units. This is considerably smaller than the 0.5 log units threshold elevation previously believed to be associated with saccadic eye movement. It is possible that even this slight threshold elevation is not a corollary of the programming of a saccade but rather reflects a disturbance of visual processing by the shearing of the retina believed to occur during a saccade.
In other experiments the elevation of threshold associated with small passive displacements of the peripheral and parafoveal retinal image (similar in extent and timecourse to those caused by microsaccadic eye movements) was studied. It was shown that there was no significant elevation of threshold associated with passive pseudo- microsaccadic displacements of the retinal image tinder conditions in which threshold elevation has been reported to occur in association with microsaccades. This discrepancy may, however, reflect a deficiency of the experiments with microsaccadic eye movements rather than a true difference between the effects of active and passive microdisplacements on visual thresholds. It was shown that small displacements of large high-contrast peripheral and parafoveal patterns can produce elevation of threshold for foveal flashes, and the form of this effect was investigated as a function of displacement magnitude.
The variation of the threshold elevation associated with three degree passive displacements of the retinal image as a function of the eccentricity of the displaced contour was studied and it was shown that even contours more than 40 degrees eccentric to the test flash produced clear threshold elevation.
The possible relation of the psychophysical threshold elevation associated with displacement of peripheral contours and the physiological periphery effect is discussed. Although there are similarities between the two effects, it is by no means certain that the two are correlates.
Subsidiary experiments showed that elevation of threshold could be produced by peripheral stimuli involving no displacement of contour, but only tachistoscopically presented pattern appearance or pattern change, and that flicker thresholds as well as flash thresholds were raised by peripheral stimulation. Scalp-evoked-potentials were recorded to investigate whether an electrophysiological correlate of the psychophysical threshold elevation associated with peripheral image displacement existed. No such effect was found.
Substantially independent experiments concerned with the control of human rapid eye movements are reported in appendices six and seven.