Glucose metabolism echoes long-range temporal correlations in the human brain

Intrinsic brain activity is characterized by pervasive long-range temporal correlations. While these scale-invariant dynamics are a fundamental hallmark of brain function, their implications for individual-level metabolic regulation remain poorly understood. Here, we address this gap by integrating resting-state functional Magnetic Resonance Imaging (fMRI) and dynamic [¹⁸F]FDG Positron Emission Tomography (PET) data acquired from the same cohort of participants. We uncover a systematic relationship between long-range temporal correlations, quantified via the Hurst exponent, and glucose metabolism. Our findings reveal that persistent temporal dependencies are associated with a measurable metabolic cost, with brains exhibiting higher long-range temporal correlations incurring greater energetic demands. Full kinetic modeling of the [¹⁸F]FDG PET data traces this association specifically to intracellular glucose phosphorylation, pointing to a direct link with neuronal energy metabolism. Beyond glucose metabolism, we also show that these dynamics are likely supported by continuous biosynthetic processes, such as protein synthesis, which are critical for neural circuit maintenance and remodeling. Overall, our results suggest that a significant fraction of the brain's so-called "Dark Energy" may be linked to spontaneous long-range temporal correlations.