Firing propagation in empirical cognitive networks of human brain

Understanding the physical mechanisms of brain functions has always been a challenging problem in the fields of nonlinear dynamics and network science. A promising approach to address this problem is by studying signal propagation on brain cognitive networks. So far, in the context of signal propagation, some progress has been achieved on complex networks, especially on the Caenorhabditis elegans network, but little attention has been paid to the empirical cognitive networks of the human brain, which are the networks responsible for cognitive tasks. Here we study how neural firings are propagated in the empirical cognitive networks of human brain. We find that the firing propagation can be seriously influenced by both the global topology of the network and the local topology of the source node. There is an optimal range of coupling strength related to synchronization for each source node, and multiple source nodes favor firing propagation. Further, we show that peripheral nodes of a network may have stronger ability of firing propagation than hub nodes. Interestingly, a remote firing propagation is observed, where firings are not propagated in a sequential rule, but propagated to farther distant nodes without the firings of intermediate nodes. A detailed theoretical analysis is provided to explain both the firing propagation and remote firing propagation.