Noise correlation length effects on a Morris-Lecar neural network.

Area of Science:

• Computational Neuroscience
• Neural Network Dynamics
• Stochastic Processes in Biology

Background:

• Neural networks process information through complex electrical signaling.
• Stochastic perturbations (noise) can significantly influence neural activity and information processing.
• Understanding noise effects is crucial for deciphering neural computation.

Purpose of the Study:

• To investigate the impact of spatially correlated stochastic perturbations on a Morris-Lecar neural network model.
• To analyze how these perturbations affect signal-to-noise ratio (SNR) and network behavior under subthreshold stimulation.

Main Methods:

• Analysis of a Morris-Lecar neural network model.
• Introduction of spatially correlated Gaussian forcing and a subthreshold Poisson process.
• Evaluation of signal-to-noise ratio, periodic excitation, collective behavior, and information transfer.

Main Results:

• Optimal signal-to-noise ratios are achieved at two critical noise intensities, resulting from the interaction between the subthreshold signal and correlated noise.
• Enhanced information transfer and development of collective network behavior (measured by averaged activity) are observed, particularly for the second peak of excitation.
• Maximum SNR increases with correlation length, saturating under global coupling, and specific mean frequencies of the subthreshold signal further boost SNR.

Conclusions:

• Spatially correlated noise plays a crucial role in enhancing information processing in neural networks.
• The findings highlight a non-trivial interplay between noise characteristics and network dynamics for optimal signal transmission.
• This study provides insights into how noise can facilitate collective behavior and improve signal detection in biological neural systems.