Millimeter-scale selective amplification in the developing visual cortex

Area of Science:

• Neuroscience
• Computational Neuroscience
• Developmental Neuroscience

Background:

• Cortical processing is thought to involve selective amplification by recurrent networks.
• It is unclear if this selective amplification occurs in millimeter-scale networks relevant to perception.

Purpose of the Study:

• To investigate whether cortical networks preferentially amplify inputs aligned with endogenous subnetworks.
• To understand how this amplification impacts sensory representation stability and specificity.

Main Methods:

• Utilized patterned optogenetic stimulation guided by computational models.
• Employed calcium imaging in immature ferret visual cortex.
• Analyzed spontaneous activity to infer dominant subnetworks.

Main Results:

• Cortical networks showed preferential activation for inputs aligned with endogenous recurrent subnetworks.
• Response reliability and specificity depended on input overlap with dominant modes.
• Well-aligned inputs were selectively amplified, yielding more stable and specific responses.

Conclusions:

• Early cortical networks leverage endogenous dynamics for selective amplification.
• This process stabilizes and refines sensory representations during development.