Learning from the pupil: a spatial visual channel in the absence of V1 in monkey and human.

The pupil of the normal human subject constricts in response not only to average increases in light energy, but also selectively to the spatial structure of a visual stimulus even when there are no energy changes. This enables one to measure visual acuity and sensitivity as a function of spatial frequency. It is known that pupillometric measures of acuity correlate well with those determined psychophysically for normal human observers. The purpose of the present study was to measure pupillary changes with stimuli delivered to the 'blind' hemifields of monkeys with unilateral V1 removal, and also with stimulation of a human subject (G.Y.) with putative V1 destruction. The results show that there are small but reliable pupillary changes to flux-equated gratings in the blind fields both in monkeys and human. The response profile in both species is very similar: it is narrowly tuned, with a peak at approximately 1 cycle/degree and a cut-off acuity of approximately 7 or 8 cycles/degree, a significant reduction compared with the intact hemifield. The result also maps well onto the psychophysically determined spatiotemporal response profile to gratings in the blind field, as determined independently for G.Y. Thus, there is a narrowly tuned spatial visual channel that does not require the integrity of V1 in monkey or human. There was no evidence under these particular conditions in either monkeys or subject G.Y. of a pupillary response to an equiluminant change from white to green or red in the hemianopic field. The pupil holds promise as a non-verbal, indirect method for determining the spatial profile, sensitivity and other properties of residual visual capacity, both in animals and humans