Functional interaction between posterior cerebellar lobes and secondary somatosensory cortex during somatosensory mismatch detection.

According to the predictive coding theory, the brain predicts future sensory inputs based on previous experiences. When there is a mismatch between the expected and the actual stimulus, a mismatch response is transmitted from low to high cortical levels to adapt the predictive model. An important cortical area for somatosensory mismatch responses is the secondary somatosensory (S2) cortex, which has reciprocal connections with the cerebellum. This study aims to characterize the role of the cortico-cerebellar interactions in the modulation of S2 cortex responses according to the predictability of afferent tactile and proprioceptive somatosensory inputs. We enrolled 20 right-handed healthy adults who underwent three functional magnetic resonance imaging (fMRI) runs (6 min each, block-design) consisting of twelve 30-s alternating blocks (10 brain volumes/block, 120 brain volumes/session) of tactile oddball paradigms. The fMRI signals within the contralateral S1 (cS1), ipsilateral S2 (iS2), and contralateral S2 (cS2) cortices; within the cortical areas involved in multimodal sensory mismatch detection (i.e., the right anterior insula (AIns), temporoparietal junction (TPJ), middle frontal gyrus (MFG), and supplementary motor are/anterior cingular cortex (SMA/ACC)); and within the ipsilateral cerebellar lobule 8 (iCL8) and 6 (iCL6) were extracted using region-of-interest (ROI) analyses and compared using ANOVA. The modulation of cortico-cerebellar functional connectivity by afferent stimuli predictability was studied using psychophysiological interaction (PPI) analyses. Predictable tactile stimuli were associated with significantly lower fMRI signals within cS1 and bilateral S2 cortices, and the right AIns, TPJ, and MFG compared to random tactile stimuli. PPI analyses showed that predictable tactile stimuli were associated with a significant increase in the functional connectivity (negative correlation) between cS2 cortex and iCL8 BOLD levels, with no significant correlation during random tactile stimulation. This effect, identified by the PPI analyses, occurred solely between the cerebellum and cS2 cortex and not with the other cortical mismatch areas. This study provides evidence for a cerebello-cortical interplay between iCL8 and the cS2 cortex in tactile somatosensory mismatch responses. The lower BOLD response in the S2 cortex observed for predictable tactile stimuli is likely mediated by an inhibitory influence of the cerebellum on the somatosensory cortex when tactile inputs are predictable.