Abstract
The genesis of this work lies in the subjective experience of some of us with strong myopia. Unsurprisingly, without our glasses the world looks blurry. Less intuitively, though, the dynamics also seem to undergo a change; the world appears to move more ‘smoothly’. Thus, a purely spatial optical transformation seemingly has temporal consequences. To formally test this anecdotal observation, we ask how convolution with a spatial Gaussian of individual images in a video-stack impacts the stack’s temporal structure. Our approach involves computing spatial and temporal Fourier spectra of several short natural videos in their original form, as well as after subjecting them to multiple levels of spatial Gaussian blur. The spatial spectrum corresponds to the mean radial-average of the 2DFFT of all images in the stack. The temporal spectrum corresponds to the mean of the 1DFFT of several single pixel cores extending through the entire depth of the stack. Comparing the spatial and temporal spectra reveals a very consistent result: For every natural video tested, spatial blurring leads to a progressive reduction in power in high spatial-, as well as in high temporal-frequencies. This straightforward and unequivocal result has several interesting implications. First, it suggests that uncorrected refractive errors lead not only to the visual system being deprived of high spatial frequency content, but also high temporal frequencies. Such deprivation would lead to deficits in high-frequency spatial as well as temporal visual processing. Indeed, the few studies that have investigated temporal aspects of amblyopia have found precisely this result. Second, it provides an explanatory account for why severe spatial degradation leads to the development of nystagmus; the uncontrolled eye-movements may serve to endogenously enhance temporal stimulation. Finally, it makes the surprising prediction that exposure to rapid temporal flicker may enhance spatial acuity. Recent reports corroborate this prediction.
Acknowledgement: NEI R01 EY020517