Abstract
Eye movements transform a spatial scene into luminance modulations on the retina. Recent work has shown that this transformation is highly structured: within the temporal range of human sensitivity, saccades deliver power that increases in proportion to spatial frequency up to a critical frequency and remains constant beyond that. Importantly, the critical frequency depends on the saccade, increasing with decreasing amplitude. Therefore, a larger saccade delivers a more effective input than a smaller one below the critical frequency. Here we tested the perceptual consequences of this input reformatting by examining how large and small saccades (6-deg and 1-deg amplitude) affect contrast sensitivity. To control for individual variability in the contrast sensitivity function, we measured relative sensitivity between two spatial frequencies: a reference, selected at the critical frequency for the smaller saccade (0.5 cycle/deg); and the probe, which, in separate experiments, either had a lower or higher frequency (0.1 or 2.5 cycles/deg). We predicted that, compared to the smaller saccade, the larger saccade will enhance probe visibility when the probe is at a lower spatial frequency than the reference, but not when the probe is at a higher spatial frequency. Subjects (N = 6) made an instructed saccade while presented with a plaid composed of overlapping orthogonal gratings (±45-deg orientation) at the two frequencies and reported which was more visible. We systematically varied the contrast of the probe to measure the point of subjective equality with the reference. Special care was taken to ensure that perceptual responses were primarily driven by saccade transients. Our results closely follow theoretical predictions. Specifically, psychometric functions following small and large saccades only differed when the probe was a lower frequency than the reference, in which case the larger saccade significantly enhanced the probe contrast. These results show that post-saccadic visibility critically depends on saccade amplitude.