Cortical Adaptation Dynamics in Human-Exoskeleton Interaction Using Multi-Model AMICA.

The human-machine interface is a crucial component of exoskeleton design, and understanding how the human nervous system adapts to and learns to coordinate with wearable robotic systems is essential for optimizing assistive device functionality. Research has shown that brain activity reflects movement-related effort and adaptation, with studies using mobile brain-body imaging approaches and EEG analysis revealing changes in cortical activity during locomotion and exoskeleton-assisted walking. This study aimed to investigate cortical dynamics during human-exoskeleton interactions using multi-model Adaptive Mixture ICA (AMICA), hypothesizing that the approach would separate EEG data into distinct phases corresponding to adaptation levels and reveal changes in brain area engagement over time. This study found a significant difference in brain activity between the early and late adaptation phases of exoskeleton-assisted walking, with higher dipole density in the frontal and premotor cortices during the early phase and higher dipole density in the primary motor and somatosensory cortices during the late phase. The early phase of adaptation was characterized by increased frontal cortex involvement, suggesting that walking with the exoskeleton initially poses a significant challenge for participants, while the late phase showed a reduction in frontal cortical areas and an increase in motor-related areas. The study's findings provide valuable insights into the neural mechanisms underlying human adaptation to lower-limb exoskeletons, highlighting the effectiveness of advanced EEG analysis techniques in capturing dynamic cortical states and offering a deeper understanding of the brain's adaptation process.