Functional mapping of human sensorimotor cortex with electrocorticographic spectral analysis. I. Alpha and beta...

Human scalp EEG studies have shown that event-related desynchronization (ERD) in the alpha (8-13 Hz) and beta (15-25 Hz) bands may be used to detect functional activation of sensorimotor cortex. However, in most previous studies somatotopy has not been examined in detail and brief, self-paced movements, focusing on the planning of motor output, have been used. We recorded electrocorticographic (ECoG) signals in five clinical subjects during a visual-motor decision task that was designed to activate the representations of different body parts in sensorimotor cortex. To focus more on execution of motor output than on its planning, subjects were instructed to make sustained isometric muscle contractions in different body parts (tongue protrusion, fist-clenching or foot dorsiflexion) in response to randomized visual stimuli depicting each action. ECoG spectral analysis utilized a mixed-effects analysis of variance model in which within-trial temporal dependencies were taken into account, and the magnitude and statistical significance of alpha and beta ERDs were mapped onto a surface rendering of each subject's brain MRI. Cortical electrical stimulation was performed in all subjects for clinical purposes, and the resulting maps of sensorimotor function were compared with those generated by ECoG spectral analysis. During the early phases of the motor responses, alpha ERD commonly occurred in a diffuse spatial pattern that was not somatotopically specific. During the late phases, the spatial pattern of alpha ERD usually became more focused and somatotopically specific. Maps of alpha ERD were closer to cortical stimulation maps when alpha ERD was sustained throughout the late phases of the motor responses. Thus, the topography of alpha ERD more resembled traditional somatotopy when its temporal profile approximated that of the motor response. The topography of beta ERD was often more discrete and somatotopically specific than that of alpha ERD, but beta ERD was often transient and sometimes absent. Sometimes, unilateral limb movement produced sustained alpha and beta ERD over bilateral sensorimotor cortices, with overlapping patterns for different body parts. The topographical spread of alpha ERD beyond expected functional-anatomical boundaries during early (and sometimes late) phases of motor responses invites a re-examination of traditional assumptions about sensorimotor functional neuroanatomy, as well as the role of alpha ERD as an index of cortical activation. We agree with others that the somatotopic representations of different body parts overlap more than previously thought. Also, unilateral limb movements may be associated with both contralateral and ipsilateral activation of sensorimotor cortex. We conjecture that alpha ERD may reflect activity within a broad synaptic network with distributed cortical representations.