Evaluating bridges across hemispheres: DCM reveals an interhemispheric interplay between premotor cortices during grasping

Despite the major contribution of the contralateral hemisphere, complex movements rely on a fine regulation of the motor output orchestrated by interhemispheric dynamics between motor and premotor cortices. In a recent paper [1], by using Dynamic Causal Modelling (DCM) and Parametric Empirical Bayes (PEB) analyses applied to fMRI data we provided evidence of a serial involvement of left parietofrontal areas during right-hand grasping planning and execution. Concurrently, we found that during the imagination of the same movement a similar, though less complex, motor program is planned; furthermore, additional processes occur to prevent motor execution. Here, we reanalyzed the same dataset to assess a) whether right parietofrontal areas (ipsilateral to the moving hand) showed connectivity couplings similar to those of their left homologues, and b) the interhemispheric dynamics between regions of interest across the two hemispheres. To this aim, we conducted two DCM analyses. First, we estimated a DCM modelling only the areas in the right hemisphere. We detected a network architecture comparable to the left hemisphere, with a similar involvement of associative and premotor areas in both conditions, and of the primary motor cortex (M1) only during motor execution. This result supports the hypothesis that the ipsilateral M1 provides additional resources to perform complex movements; overall, the similarity of the left and the right hemisphere DCMs confirms the wide involvement of a bilateral parietofrontal network during grasping. Second, we built an interhemispheric model including the suprathreshold intrahemispheric connections as resulting by the two single-hemisphere DCMs; we used their posterior estimates to inform the priors of a third, interhemispheric DCM [2]. We also modelled the interhemispheric connections between parietofrontal areas. Results showed no additional interhemispheric couplings involving parietal areas. Instead, during executed grasping an inhibitory influence from the right dorsal premotor cortex (PMd) toward the left premotor and motor areas emerged, as well as couplings between homologous ventral premotor regions. During imagery, an inhibition from the right PMd to the left M1 emerged, presumably concurring to prevent the motor output. Overall, our results support the view of a complex interplay both within and between hemispheres during complex movements such as grasping.