Large-Scale Intrinsic Functional Brain Organization Emerges from Three Canonical Spatiotemporal Patterns

Abstract The past decade of functional neuroimaging research has seen the application of increasingly sophisticated advanced methods to characterize intrinsic functional brain organization. Accompanying these techniques are a patchwork of empirical findings highlighting novel properties of intrinsic functional brain organization. To date, there has been little attempt to understand whether there is an underlying unity across this patchwork of empirical findings. Our study conducted a systematic survey of popular analytic techniques and their output on a large sample of resting-state fMRI data. We found that the apparent complexity of intrinsic functional brain organization can be seamlessly reduced to three fundamental low-frequency spatiotemporal patterns. Our study demonstrates that a long list of previously observed phenomena, including functional connectivity gradients, the task-positive/task-negative pattern, the global signal, time-lag propagation patterns, the quasiperiodic pattern and the network structure of the functional connectome are simply manifestations of these three spatiotemporal patterns. An in-depth characterization of these three spatiotemporal patterns using a novel time-varying complex pattern analysis revealed that these three patterns may arise from a single hemodynamic mechanism.