Crossmodal visual predictions elicit spatially specific early visual cortex activity but later than real visual stimuli

Previous studies have indicated that crossmodal visual predictions are instrumental in controlling early visual cortex activity. The exact time course and spatial precision of such crossmodal top-down influences on visual cortex have been unknown. In the present study, participants were exposed to audio-visual combinations comprising one of two sounds and a Gabor patch either in the top left or in the bottom right visual field. Event related potentials (ERP) were recorded to these frequent crossmodal combinations (Standards) as well as to trials in which the visual stimulus was omitted (Omissions) or the visual and auditory stimuli were recombined (Deviants). Standards and Deviants elicited an ERP between 50 and 100 ms of opposite polarity known as a C1 effect commonly associated with retinotopic processing in early visual cortex. In contrast, a C1 effect was not observed in Omission trials. Spatially specific Omission and Mismatch effects (Deviants minus Standards) started only later with a latency of 230 ms and 170 ms, respectively. These results suggest that crossmodal visual predictions control visual cortex activity in a spatially specific manner. However, visual predictions do not elicit neural responses that mimic stimulus-driven activity but rather seem to affect early visual cortex via distinct neural mechanisms.