Structured networks support sparse traveling waves in rodent somatosensory cortex

Area of Science:

• Neuroscience
• Computational Neuroscience
• Sensory Processing

Background:

• Neurons in rodent barrel cortex layer 2/3 (L2/3) are intermixed and respond to different whiskers.
• Whisker deflection activates sparse, distributed neuronal populations across cortical columns.
• The organization of L2/3 for distributed sensory representations remains unclear.

Purpose of the Study:

• To investigate the hypothesis that sensory representations in L2/3 are formed by horizontal activity propagation.
• To model the mechanisms underlying sparse, distributed neural activity in the barrel cortex.

Main Methods:

• Development of a computer model simulating neural activity in L2/3.
• Testing the hypothesis of horizontal activity propagation within L2/3.
• Analysis of model parameters to explain observed neuronal response properties.

Main Results:

• The model successfully replicated key properties of L2/3 neurons, including low response probability and a subset of reliable responders.
• Sparsely propagating traveling waves, mirroring experimental observations, were generated.
• These waves occurred specifically when a strongly connected subnetwork was embedded within a larger, weakly connected network.

Conclusions:

• Horizontal activity propagation within L2/3 is a plausible mechanism for forming distributed sensory representations.
• The network structure, with embedded strongly connected subnetworks, is crucial for generating traveling waves.
• This model provides insights into the computational principles of sensory processing in the barrel cortex.