Somatosensory processing during movement observation in humans

OBJECTIVES: A neural system matching action observation and execution seems to operate in the human brain, but its possible role in processing sensory inputs reaching the cortex during movement observation is unknown. METHODS: We investigated somatosensory evoked potentials (SEPs), somatosensory evoked fields (SEFs) and the temporal spectral evolution of the brain rhythms (approximately 10 and approximately 20 Hz) following electrical stimulation of the right median nerve in 15 healthy subjects, during the following randomly intermingled conditions: a pure cognitive/attentive task (mental calculation); the observation of a motoric act (repetitive grasping) with low cognitive content ('Obs-grasp'); and the observation of a complex motoric act (finger movement sequence), that the subject had to recognize later on, therefore reflecting an adjunctive cognitive task ('Obs-seq'). These conditions were compared with an absence of tasks ('Relax') and actual motor performance. RESULTS: The post-stimulus rebound of the approximately 20 Hz beta magnetoencephalographic rhythm was reduced during movement observation, in spite of little changes in the approximately 10 Hz rhythm. Novel findings were: selective amplitude increase of the pre-central N(30) SEP component during both 'Obs-grasp' and 'Obs-seq', as opposed to the 'gating effect' (i.e. amplitude decrease of the N(30)) occurring during movement execution. The strength increase of the 30 ms SEF cortical source significantly correlated with the decrease of the approximately 20 Hz post-stimulus rebound, suggesting a similar pre-central origin. CONCLUSIONS: Changes took place regardless of either the complexity or the cognitive content of the observed movement, being related exclusively with the motoric content of the action. It is hypothesized that the frontal 'mirror neurons' system, known to directly facilitate motor output during observation of actions, may also modulate those somatosensory inputs which are directed to pre-central areas. These changes are evident even in the very first phases (i.e. few tens of milliseconds) of the sensory processing.