Spatial Connectivity Pattern of The Human Brain's Action Network.

This study proposes a spatiotemporal connectome-based framework to characterize the human brain's action network. Unlike temporal connectivity (TPC), this approach leverages full functional connectivity profiles derived from large-scale neural wave dynamics, termed spatial connectivity (SPC). We applied SPC to map the action network for the first time in a non-Western young adult cohort. Our results delineate the network's detailed functional architecture across the cerebral cortex, cerebellum, and subcortical nuclei. Compared to TPC, SPC is more robust to global signal regression when characterizing anticorrelation between the action and default networks. Critically, this intrinsic antagonism reflects a fundamental energy-saving balance in the brain's dynamical system. Moreover, SPC reveals that the salience/parietal memory network resides at the interface of this antagonism, exhibiting weak but highly variable connectivity, a position that enables rapid state switching between external action and internal contemplation. These findings offer a mechanistic view of the brain's resting "dark energy" and suggest a blueprint for brain-inspired AI. Embedding a dual-system opponent architecture, balanced by a flexible switching hub, may foster adaptive, energy-efficient intelligence. All derived high-resolution SPC patterns and computational code are publicly shared to promote open science ( https://ccndc.scidb.cn/en ).