Abstract:The intrinsic organizational structure of the brain is reflected in spontaneous brain oscillations. Its functional integration and segregation hierarchy have been discovered in space by leveraging gradient approaches to low-frequency functional connectivity. This hierarchy of brain oscillations has not yet been fully understood, since previous studies have mainly concentrated on the brain oscillations from a single limited frequency range (similar to 0.01-0.1 Hz). In this work, we extended the frequency range and performed gradient analysis across multiple frequency bands of fast resting-state fMRI signals from the Human Connectome Project and condensed a frequency-rank cortical map of the highest gradient. We found that the coarse skeletons of the functional organization hierarchy are generalizable across the multiple frequency bands. Beyond that, the highest integration levels of connectivity vary in the frequency domain across different large-scale brain networks. These findings are replicated in another independent dataset and demonstrated that different brain networks can integrate information at varying rates, indicating the significance of examining the intrinsic architecture of spontaneous brain activity from the perspective of multiple frequency bands.

Connectivity Gradients in Spontaneous Brain Activity at Multiple Frequency Bands