The relevance of network micro-structure for neural dynamics

Area of Science:

• Computational Neuroscience
• Network Science
• Systems Neuroscience

Background:

• Cortical neuron activity depends on presynaptic input.
• Previous studies often used simplified random network models.
• Understanding network structure's impact on neuronal activity is crucial.

Purpose of the Study:

• To investigate the relationship between complex network structures and neuronal activity.
• To explore how quasi-fractal network models reveal more about recurrent networks.
• To determine if network structure can be inferred from activity data.

Main Methods:

• Constructed quasi-fractal networks with higher variability than typical models.
• Simulated integrate-and-fire neuron networks in an asynchronous, irregular state.
• Analyzed statistical relationships between network topology and spiking activity.

Main Results:

• Network structure significantly influences single neuron and population activity dynamics.
• Consistent correlations found between activity (e.g., spike train irregularity) and network properties (e.g., connection distributions, clustering).
• Network structure dictates higher-order spiking activity correlations.

Conclusions:

• Detailed network structure profoundly impacts neural activity, even with fixed global parameters.
• It is possible to estimate network structural characteristics from observed neuronal activity.
• Findings offer new interpretations for spike train recordings and guide future theoretical network studies.