Modelling a multiplex brain network by local transfer entropy.

Area of Science:

• Neuroscience
• Network Science
• Information Theory

Background:

• Understanding causal relationships between brain regions is crucial for neuroscience.
• Resting-state functional magnetic resonance imaging (fMRI) provides insights into brain connectivity.
• Information transfer mechanisms underlying these connections remain an active area of research.

Purpose of the Study:

• To investigate information transfer mechanisms and causal relations between brain regions during resting state.
• To model the functional architecture of the brain based on observed causal rules.

Main Methods:

• Utilized fMRI data from healthy individuals (1000 Functional Connectomes Beijing Zang dataset).
• Applied Transfer Entropy concepts to quantify causal information transfer between brain regions.
• Developed a directional multilayer network model based on identified interaction rules.

Main Results:

• Identified four distinct time-dependent causal interaction rules (e.g., deactivation, turn-off).
• These rules exhibit characteristics of both positive and negative functional connectivity.
• Modeled the emergent functional architecture using a multilayer network with four interaction matrices.

Conclusions:

• The developed model provides a framework for understanding brain functional architecture based on information transfer.
• The approach demonstrates statistical significance within the homogeneous subject database.
• This methodology holds promise for detecting differences in brain function across clinical conditions and cognitive states.