The “multisensory” visual cortex: crossmodal shaping of visual cortical ​responses and ​​perception​

The convergence and integration of information from the different sensory channels represents a fundamental ability of the human brain: in our daily life, multisensory cues impact our sensory system, shaping our perception of sensory events. Classical models of multisensory perception defer integration until sensory-specific information has been extensively processed. Strikingly, several anatomical studies now suggest that this view may be over-simplistic and that the substrate for multisensory integration is not constrained to the classical area of multisensory convergence; rather, it also occurs in the early stages of sensory processing, rising the intriguing hypothesis that most of the brain, including the primary sensory cortices, is essentially multisensory. The present dissertation inquires the causal involvement of classical ‘sensory-specific’ visual areas in multisensory processing. Using a combination of behavioral, neuromodulatory, and neuropsychological evidence, I seek the behavioral and brain signatures of a causal link between visual cortical excitability and multisensory perception. In parallel, I provide a characterization of the impact of distinct crossmodal stimuli on subjective visual experience. By directly measuring visual cortical excitability via Transcranial Magnetic Stimulation (TMS), the first study shows facilitatory effects by spatially-specific bimodal and trimodal stimuli on visual cortical responses, which, in turn, improve visual perception. Moreover, by using Transcranial Direct Current Stimulation (tDCS), I demonstrate the role played by higher order multisensory cortices in mediating such spatially-specific crossmodal influences on visual perception, reporting the presence of regional preferences for auditory or somatosensory influences on visual responses. In the second experiment, I further extend these findings by showing that multisensory influences on vision not only express themselves through an enhancement of visual perception, but they can also provoke phenomenological changes in conscious visual perception, namely a crossmodal illusion, when incongruent auditory cues are provided. Crossmodal illusory effects show a specific time-course, compatible with the occurrence of early visual-auditory interaction in the primary visual cortex. At complement with evidence in the healthy brain, in the third study I investigate how a well-known crossmodal illusion, the Sound-Induced Flash Illusion (SIFI), is processed by brain-damaged patients with visual field defect (with damage to the primary visual cortex, the sensory visual pathways, or both), and unilateral spatial neglect (with damage to the posterior parietal and fronto-temporal regions). Perception of the SIFI is defective in patients with visual field defects, but not in those with unilateral spatial neglect, further supporting the role of low-level visual areas in integrating multisensory cues. Overall, this set of experiments shows a causal link between the crossmodal modulation of visual perception and the activity of the primary visual areas, which represents a key site for multisensory integration.