Abstract
The pupil constricts in response to visual stimuli that keep net luminance unchanged but that do introduce local luminance increments and decrements-a reaction here called "isoluminant constriction." This response can form a pupillometric index of visual processing, but it is unclear what kind of processing it reflects; some authors have suggested that the constriction arises from subcortical, luminance-based neural signals, whereas others have argued for an origin at cortical, feature-based processing stages. We tested the involvement of cortical neural activity in isoluminant constrictions. To this end, we measured constrictions to stimuli presented after contrast adaptation, an adaptation procedure thought to lessen cortical stimulus responses. If cortical processing is involved in the isoluminant constriction, then such adaptation should lead to reduced isoluminant constriction amplitudes. We tested this prediction in the course of three experiments. We found no evidence for the prediction in any of the experiments, and did find Bayesian evidence against the prediction. These results suggest that, at least in the conditions of our experiments, isoluminant constrictions may not reflect visual cortical processing.