Functional brain network dynamics based on the Hindmarsh–Rose model

Abstract

In order to reveal the dynamics of brain network, we proposed a new research method based on the Hindmarsh–Rose model. In the method, a neural network model was developed by constructing a functional brain network topology based on functional magnetic resonance imaging resting-state data and using Hindmarsh–Rose neurons as nodes in place of the brain regions belonging to the functional brain network. The dynamics of the functional brain network were investigated using the dynamics model. The simulation results showed that the dynamic behaviors of the brain regions in the functional brain network could be divided into three types: stable, chaotic, and periodical bursts. A state space was introduced to analyze the dynamic behavior of the brain regions in the network. We find that increasing excitation and mutual connection strength among brain regions enhanced network communication capabilities in the state space. Both the periodic and stable modes exhibited stronger communication capabilities than the chaotic mode. Despite individual differences in the dynamics of brain regions among subjects, brain regions in the periodic mode were highly consistent and corresponded to key regions of the default mode network in the resting state.