Principles of cortical interactions in modular recurrent networks

Area of Science:

• Neuroscience
• Computational Neuroscience
• Visual System Research

Background:

• Cortical processing adapts to sensory input strength, integrating weak signals and refining strong ones.
• Recurrent cortical circuits, involving amplification and competition, are hypothesized to mediate this adaptation.
• Previous research in mouse V1 suggested novel circuit motifs with strong excitatory-inhibitory cell connections.

Purpose of the Study:

• To investigate the behavior of cortical columns within functionally topographic circuits.
• To determine if cortical columns act as amplifiers or engage in feature competition.
• To test the prediction that cortical circuits switch between amplification and suppression based on sensory drive strength.

Main Methods:

• Utilized single-cell perturbations in ferret primary visual cortex (V1).
• Employed two-photon optogenetic stimulation combined with a generalized linear model (GLM) for perturbation quantification.
• Developed and analyzed a recurrent network model with specific connectivity patterns.

Main Results:

• Demonstrated that cortical networks switch between amplification and suppression depending on stimulus strength (contrast).
• Observed amplification between functionally coupled cells at low contrast and suppression at high contrast.
• Confirmed stronger suppressive influences from inhibitory onto excitatory cells in recurrent networks.

Conclusions:

• Cortical recurrence with functional topography dynamically toggles between amplification and suppression.
• Provides direct experimental evidence supporting predictions from normative and mechanistic circuit models of the visual cortex.
• Highlights the role of stimulus contrast in modulating network interactions within the visual cortex.