A master equation formalism for macroscopic modeling of asynchronous irregular activity states

Area of Science:

• Computational Neuroscience
• Theoretical Neuroscience
• Systems Neuroscience

Background:

• Asynchronous irregular (AI) activity states are crucial for brain function, mimicking cerebral cortex activity in awake animals.
• Existing models often rely on balanced networks of spiking neurons with sparse connectivity, requiring complex simulations.

Purpose of the Study:

• To develop a novel mesoscopic description for modeling asynchronous irregular (AI) activity states in neural networks.
• To provide a framework applicable to various neuron models and network configurations.

Main Methods:

• Utilized master equation formalism to derive a second-order mean-field set of ordinary differential equations.
• Developed a mesoscopic approach to describe the temporal evolution of randomly connected balanced networks.
• Incorporated finite size effects into the modeling framework.

Main Results:

• The derived mean-field equations accurately predict the behavior of balanced neural networks.
• The mesoscopic approach demonstrates applicability across different network configurations and parameter spaces.
• Validated the model's predictions against numerical simulations.

Conclusions:

• The proposed mesoscopic description offers an efficient way to model AI states in balanced neural networks.
• This formalism accounts for finite size effects and neuron model variability.
• The approach facilitates the construction of large-scale networks for modeling macroscopic brain activity.