Lower Bound Estimates for Electrophysiological Power Dissipation in Human Gray Matter

The human brain is popularly described as remarkably energy efficient, with its metabolic power consumption estimated at approximately 20 watts (W). Biophysical models have partitioned this power budget across distinct cellular processes, but these theoretical estimates have yet to be empirically constrained using whole-brain electrophysiological data. Here, we used magnetoencephalography (MEG) source imaging of resting-state human brain activity to derive empirical lower-bound estimates of electrical power dissipation in cortical gray matter. We found that the total power dissipated by currents, primarily associated with post-synaptic potentials, ranges between 10-9 and 10-10 W, several orders of magnitude lower than prevailing metabolic estimates. Using finite element modeling (FEM), we observed that electrophysiological power dissipation is predominantly confined to gray matter. Additionally, spatial variations in MEG-derived power dissipation partially correlated with regional oxygen metabolism measured by PET, yet notable discrepancies emerged across large-scale functional brain networks. These results underscore a critical divergence between electrophysiological and metabolic indices of brain energy use, and highlight the need for more integrated biophysical models to bridge this gap and better characterize the physiological underpinnings of regional brain energetics.