A small world of neuronal synchrony

Area of Science:

• Neuroscience
• Computational Neuroscience
• Network Science

Background:

• Small-world networks are hypothesized to balance modular and global processing in the brain.
• Understanding functional connectivity in cortical networks is crucial for deciphering brain computations.

Purpose of the Study:

• To investigate the functional network properties of cortical neurons.
• To determine if cortical networks exhibit small-world characteristics.
• To assess the optimization of neural networks for local and global computations.

Main Methods:

• Functional connectivity was identified using correlation analysis of neuronal activity.
• The Ising model, based on maximum entropy, was applied to reconstruct interaction networks.
• Pairwise correlations and synchrony were measured from simultaneous recordings of up to 24 neurons in the visual cortex of anesthetized cats.

Main Results:

• Pairwise correlations reliably captured population activity patterns, enabling network reconstruction.
• The reconstructed cortical networks demonstrated small-world properties with short average path lengths and high clustering coefficients.
• Significant variation in neuronal interaction strength and number suggested the presence of network hubs, but no scale-free properties were observed.

Conclusions:

• Cortical networks possess small-world properties, optimizing the balance between local and global processing.
• This network architecture supports efficient feature detection (local) and feature integration or binding (global) computations.
• The findings provide insights into the principles underlying neural computation and network organization.