Modularity and robustness of frontal cortical networks

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• Short-term memory (STM) relies on persistent neural activity within interconnected brain networks.
• Mechanisms for maintaining STM robustness against corrupted information remain poorly understood.
• Interactions between brain regions in maintaining STM are a key area of investigation.

Purpose of the Study:

• To investigate how brain regions interact to maintain persistent activity for short-term memory.
• To understand the mechanisms underlying robustness to corrupted information in neural networks.
• To explore individual variability in frontal cortical network organization for memory.

Main Methods:

• Simultaneous measurement of large neuronal populations across mouse frontal hemispheres.
• Analysis of coordinated activity patterns between hemispheres during memory tasks.
• Perturbation experiments to assess network robustness and information gating.

Main Results:

• Hemispheric coordination maintains coherent short-term memory.
• Individual variability exists in frontal cortical network organization.
• A modular network organization confers robustness to perturbations, preventing spread.
• A dynamic gating mechanism facilitates coherent information transfer while excluding corrupt data.

Conclusions:

• Robust short-term memory is supported by redundant modular representations across brain regions.
• These redundant modular representations naturally emerge in neural network models trained for robustness.
• Dynamic gating and modularity are critical for resilient cognitive functions like short-term memory.