Neural Action Representations in the Human Frontoparietal Network

Motor cognition relies on the brain's ability to form dynamic, context-sensitive representations of actions. While the mirror neuron theory has provided foundational insights into action observation, it does not fully account for the cognitive flexibility required in motor tasks. Recent theories highlight the role of internal models that allow the brain to combine simpler components (compositionality) and adapt learned actions to new contexts (generalizability). The posterior parietal and dorsolateral prefrontal cortex are crucial for these functions, particularly in action intention and observation. This proposal aims to explore the encoding mechanisms of action representations in these regions using microelectrode arrays in tetraplegic individuals and neurosurgical epilepsy patients. Participants will observe and either intend or execute a variety of actions with different effectors to identify the key dimensions driving neural activity, using both videos and VR to enhance realism. In addition, they will be presented with multisensory stimuli to examine semantic encoding, modality interactions, and conflict processing. This research will map the dynamics of the action observation system and its compositional nature, offering insights into motor cognition, contributing to brain-computer interface development, and advancing theories on how the brain generalizes complex representations.