Fractal dynamics of the healthy human brain: A subcortical-to-cortical gradient of temporal complexity

Loss of brain signal complexity can indicate signs of diseases, yet standard scans often miss these patterns. Establishing a high-resolution benchmark of healthy temporal complexity is essential for developing early cerebral biomarkers. To construct a fractal normative reference map of brain complexity, we analyzed high-resolution 7T fMRI scans from 173 healthy adults. We used fractal dimension analysis to map the brain spatiotemporal organization and test its relationship with human intelligence. We discovered a robust physiological gradient, from the globus pallidus (GP) (hub of maximal "healthy chaos") to fronto-parietal cognitive networks (which prioritize stability). This architecture predicts fluid intelligence. This study establishes the globus pallidus as the engine of neural flexibility while cortical network underlying cognitive abilities exhibit more regular dynamics, linked with fluid intelligence. These results offer insights to detect the dynamic rigidity that often precedes structural damage in neurological diseases, a hypothesis that requires validation in future clinical studies.