Unique scales preserve self-similar integrate-and-fire functionality of neuronal clusters

Area of Science:

• Neuroscience
• Computational Neuroscience
• Artificial Intelligence

Background:

• Brain networks exhibit hierarchical clustering across diverse spatial scales.
• Understanding neuronal cluster size is crucial for both neuroscience and artificial intelligence.
• Existing research supports various forms and sizes of neural clustering, from dendrites to thousands of neurons.

Purpose of the Study:

• To investigate the structural and functional self-similarity of brain networks across different scales.
• To explore the preservation of neuron-like signal integration functionality at specific clustering scales.
• To propose a coarse-graining method for neuronal networks in computational modeling.

Main Methods:

• Utilized computational simulations based on brain-derived fMRI networks.
• Analyzed structural self-similarity across multiple spatial scales.
• Examined the functional "integrate and fire" property at various clustering scales.

Main Results:

• Demonstrated that brain networks maintain structural self-similarity across scales.
• Confirmed the preservation of neuron-like signal integration functionality at specific scales.
• Identified a fractal-like spatiotemporal property in both brain network structure and function.

Conclusions:

• Proposed a coarse-graining approach for neuronal networks into ensemble-nodes and ensemble-spikes.
• Highlighted the utility of fractal-like properties for bridging experimental scales in computational modeling.
• Suggested these properties may impose constraints on brain size evolution, aligning with punctuated equilibrium theories.