Mutual Information and Information Gating in Synfire Chains.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Information Theory

Background:

• Coherent neuronal activity, including synchronous firing and oscillations, is crucial for brain information processing.
• Synfire chains are theoretical models for coherent spiking and information transfer in feedforward networks.
• Synfire-gated synfire chains (SGSCs) extend classical models to support graded firing rate transfer.

Purpose of the Study:

• To investigate information propagation in synfire-gated synfire chains (SGSCs).
• To analyze the impact of gating and noise on information transfer within these networks.
• To identify conditions supporting different modes of information propagation (binary vs. graded).

Main Methods:

• Utilized mutual information analysis to quantify information transfer as a function of network layer depth.
• Simulated feedforward networks with synfire chain architecture and added gating currents.
• Varied network parameters to explore the effects of gating and noise on signal propagation.

Main Results:

• Identified two primary parameter space regions for information propagation in SGSCs.
• Demonstrated that pulse-gating controls the transfer of information.
• Showcased a large region supporting binary code propagation and a smaller region supporting graded propagation across multiple layers.

Conclusions:

• Synfire-gated synfire chains can effectively propagate information through feedforward networks.
• Pulse-gating is a key mechanism for controlling information transfer modes (binary or graded).
• The findings provide insights into neural coding and information processing mechanisms in the brain.