Brain network connectivity and topological analysis during voluntary arm movements

Functional connectivity (FC) estimates the temporal synchrony amongfunctionally homogeneous brain regions based on the assessment of thedynamics of topologically localized neurophysiological responses. The aimof this study was to investigate task-related changes in brain activity andFC by applying different methods namely event-related desynchronization(ERD), coherence and graph-theoretical analysis toelectroencephalographic (EEG) recordings, for comparing their respectivedescriptive power and complementarity. As it is well known, ERD providesan estimate of differences in power spectral densities between active (ortask) and rest conditions, FC allows assessing the level of synchronizationbetween the signals recorded at different scalp locations and graphanalysis enables the estimation of the functional network features andtopology.EEG activity was recorded on 10 subjects during left/right arm movements(LAM/RAM). The theta, alpha and beta bands were considered.Conventional analysis showed a significant ERD in both alpha and betabands over the sensorimotor cortex during the LAM and in beta bandduring the RAM, besides identifying the regions involved in the task, as itwas expected. On the other hand, connectivity assessment highlighted thatstronger connections are those that involved the motor regions for whichgraph analysis revealed reduced accessibility and an increased centralityduring the movement. Jointly, the last two methods allow identifying thecortical areas that are functionally related in the active condition as well asthe topological organization of the functional network. Results support thehypothesis that network analysis brings complementary knowledge withrespect to established approaches for modeling motor-induced FC andcould be profitably exploited in clinical contexts.