Action Observation and Execution Differentially Modulate Basal Ganglia-Cortical Activity in Humans

Abstract

INTRODUCTION: To understand an observed action, without overt execution, the observer is thought to run similar motor plans; indicating a strong coupling between the neural representation of action observation and execution. Imaging studies show activation of basal ganglia (BG) and motor cortex during movement yet comparative studies using direct and invasive human recordings are sparse. We aimed to characterize how movement execution and observation differentially modulate local and inter-regional activity across BG and sensorimotor cortex. METHOD(S): We recorded LFP from globus palidus internus (GPi) and electrocorticography from ipsilateral sensorimotor cortex in 9 PD subjects during deep brain stimulation surgery. Subjects performed block design tasks alternating between 30 s of rest and performing or observing finger tapping with random order.We assessed changes in spectral power, along with pallidocortical coherence and cortical PAC. RESULT(S): We observed suppression of alpha-beta (9-25 Hz) power in contralateral GPi and sensorimotor cortex during both activities. This power suppression was significantly weaker in the motor cortex during the action observation compared to the execution (P = .02). However pallidal spectral changes in alpha-beta frequencies were not different across tasks (P = .3). Uniquely during the action execution, there was a significant increase in the gamma power (80-200 Hz) at the motor cortex (P < .05). In addition, Pallidocortical beta coherence, and motor cortical beta-gamma PAC were significantly suppressed during action execution (P < .05) and not the observation. CONCLUSION(S): Our results support the functional dissociation within the BG-cortical network during action observation and execution. Although spectral power changes in a-s in the BG are largely similar across tasks, suppression of BG-cortical functional connectivity is a feature of movement execution. In addition, increase in the cortical gamma power and beta-gamma phase amplitude decoupling only happen during the movement execution, in line with the theory that during movement execution the gamma signal is released from the constraint of beta.