The effect of self- vs. externally generated actions on timing, duration and amplitude of BOLD response for visual feedback processing

It has been widely assumed that internal forward models use efference copies to create predictions about the sensory consequences of our own actions. While these predictions had been frequently associated with reduced neural processing in sensory cortices, the timing and duration of the hemodynamic response of self-generated as opposed to externally generated movements is poorly investigated. In the present study we tested the hypothesis that predictive mechanisms for self-generated actions lead to early and shorter neural processing compared with externally generated movements. Using a first and second-order Taylor approximation in terms of the temporal (TD) and dispersion (DD) derivatives of a canonical hemodynamic response function, we investigated the timing and duration of activation for self-generated and externally generated movements using a custom-made fMRI-compatible movement device. Visual video feedback of the active and passive hand movements were presented in real time or with variable delays (0 - 417 ms). Participants had to judge, whether the feedback was delayed. We found earlier feedback processing for self-generated compared to externally generated movements in several regions including the supplementary motor area, cerebellum, subcortical structures such as the putamen and visual cortices. Shorter processing was found in areas, which show also lower blood oxygen level dependent (BOLD) amplitudes, such as the SMA, occipital and parietal cortex. Specifically, earlier activation in the putamen, of self-generated movements was associated with worse performance in detecting delays. These findings support our hypothesis, that efference copy based predictive mechanisms enable earlier processing of action feedback, as potential source for behavioral effects.