Abstract
We used combined monkey physiology with human fMRI to localize the neuronal correlates of visual awareness in humans and monkeys using visual masking illusions. In visual masking, visible targets are rendered invisible by modifying the context in which they are presented, but not by modifying the targets themselves. Visual masking illusions have the advantage that they can be presented in both monoptic and dichoptic format, allowing one to differentially test binocular and monocular levels of the visual system. We compared monoptic visual masking activation, which we found to be evident in all retinotopic visual areas, to dichoptic activation, which we found to be equal in strength to monoptic masking only in areas downstream of V2 in the human. Because monoptic and dichoptic masking are perceptually equal in strength, the results establish a lower bound in the visual hierarchy for maintenance of visual awareness of simple unattended targets. The results also establish for the first time that monoptic visual masking effects are explicitly processed in all retinotopic visual areas.