Parcels and particles: Markov blankets in the brain

Area of Science:

• Neuroscience
• Computational Neuroscience
• Complex Systems

Background:

• Human brain mapping traditionally relies on functional segregation and integration principles.
• Functional integration is studied via mechanistic modeling of effective connectivity or phenomenological analysis of functional connectivity.
• Current approaches often treat these two aspects of connectivity separately.

Purpose of the Study:

• To present a unified framework for understanding effective connectivity.
• To demonstrate how effective connectivity can explain the emergence of intrinsic brain networks and critical dynamics.
• To bridge the gap between directed network dynamics and observed network phenomena.

Main Methods:

• Utilizing the concept of Markov blankets, crucial for self-organizing systems.
• Applying the renormalization group apparatus to analyze neuronal states.
• Modeling neuronal states and their partitioning across progressively coarser scales.

Main Results:

• Effective connectivity, when analyzed through Markov blankets and renormalization groups, naturally gives rise to intrinsic brain networks.
• The framework explains phenomena observed in network neuroscience, such as self-organized criticality and dynamical instability.
• A direct link is established between the dynamics on directed graphs and the phenomenology of intrinsic brain networks.

Conclusions:

• Markov blankets provide a powerful theoretical tool for understanding brain organization.
• The renormalization group offers a method to derive macroscopic network properties from microscopic neuronal dynamics.
• This work unifies mechanistic and phenomenological approaches to functional connectivity in the brain.