What activates the human mirror neuron system during observation of artificial movements: bottom-up visual features or top-down intentions?

In a recent study we could show that during observation of artificial object movements a similar cortical network, including the areas of the so-called human mirror neuron system (hMNS), was activated as during the observation of hand movements. The present study investigated whether activation of the hMNS during the observation of artificial object movements depends more on the visual features of the movements (buttom-up), or, by manipulating the task instructions, on the intentional goal of an observer (top-down). Using a factorial design we recorded the hemodynamic responses in 20 healthy participants while they watched arbitrary artificial object movements following two types of movement trajectories (smooth vs. discontinuous). In one part of the experiment participants had to detect color changes of two objects (color task) and in another part they had to judge whether the movement pattern of two objects could be performed with human hands (simulation task). We found stronger activation in the hMNS during the simulation than during the color task for both types of movement trajectories. In contrast, the color task activated the left ventral-occipital area (human V4). A direct comparison of smooth vs. discontinuous movement trajectories revealed significant effects neither in the structures of the hMNS nor in human V4. The present findings suggest that it is not a specific visual feature, such as a smooth biological movement trajectory, that activates the hMNS. Rather, the hMNS seems to respond when an observed movement is matched to a motor representation triggered by the intentional goal of the observer.