Tuning Criticality through Modularity in Biological Neural Networks

• Published in: The Journal of neuroscience Vol. 43; no. 33; pp. 5881 - 5882
• Main Authors: Irani, Martín, Alderson, Thomas H
• Format: Journal Article
• Language: English
• Published: United States Society for Neuroscience 16.08.2023
• Subjects: Brain; Brain architecture; Critical point; Journal Club; Modularity; Neural networks; Neural Networks, Computer; Optimization
• ISSN: 0270-6474; 1529-2401
• DOI: 10.1523/JNEUROSCI.0865-23.2023

The brain operates at a critical point between runaway excitation and rapid extinguishment, where activity propagates optimally. The underlying organizational principles that drive the brain towards criticality are not fully understood. Recent studies have shown that local circuits in the cortex operate away from the critical point in the supercritical regime. However, a revised version of the criticality framework suggests that neural networks operate in a broad range of configurations around the critical point, known as the Griffith phase. Modularity, or the modular organization of brain networks, plays a crucial role in stretching the critical point towards extended Griffiths regions. In a recent study, researchers manipulated the modularity of cortical cell cultures and measured neuronal avalanches to assess the criticality of networks with different degrees of modularity. They found that networks with intermediate degrees of modularity displayed optimal propagation of activity. The results suggest that modularity is an important feature of critical dynamics in brain networks. The exact mechanisms by which neuronal dynamics guide the development of modularity remain unknown, but activity-dependent factors and synchronized activity between modules may play a role. Understanding the relationship between modularity and dynamics across different spatial scales is crucial for a better understanding of brain function.