Citation

Bradford JC, Lukos JR, Ferris DP. J. Neurophysiol. 2015; 115(2): 958-966.

Affiliation

University of Michigan.

Copyright

(Copyright © 2015, American Physiological Society)

DOI

10.1152/jn.00089.2015

PMID

26683062

Abstract

The objective of this study was to determine if electrocortical activity is different between walking on an incline compared to level surface. Subjects walked on a treadmill at 0% and 15% grades for 30 minutes while we recorded electroencephalography (EEG). We used independent component analysis to parse EEG signals into maximally independent components (IC) and then computed dipole estimations for each IC. We clustered cortical source ICs and analyzed event-related spectral perturbations synchronized to gait events. Theta power fluctuated across the gait cycle for both conditions, but was greater during incline walking in the anterior cingulate, sensorimotor and posterior parietal clusters. We found greater gamma power during level walking in the left sensorimotor and anterior cingulate clusters. We also found distinct alpha and beta fluctuations depending on the phase of the gait cycle for the left and right sensorimotor cortices, indicating cortical lateralization for both walking conditions. We validated the results by isolating movement artifact. We found that the frequency activation patterns of the artifact were different than the actual EEG data, providing evidence that the differences between walking conditions were cortically driven rather than a residual artifact of the experiment. These findings suggest that the locomotor pattern adjustments necessary to walk on an incline compared to level surface may require supraspinal input, especially from the left sensorimotor cortex, anterior cingulate, and posterior parietal areas. These results are a promising step towards the use of EEG as a feed forward control signal for ambulatory brain-computer interface technologies.

Language: en