Circuit-motivated generalized affine models characterize stimulus-dependent visual cortical shared variability.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Correlated neural variability in the visual cortex is influenced by stimulus characteristics, impacting neural coding and revealing circuit organization.
• An existing affine model explains orientation-dependent variability in primary visual cortex (V1), but its applicability to other variables and inter-areal communication is unclear.

Purpose of the Study:

• To extend the affine model to account for contrast-dependent neural variability within visual areas (V1, V2) and orientation-dependent variability between V1 and V2.
• To bridge neural circuit mechanisms with statistical modeling for a comprehensive understanding of stimulus-modulated neural variability.

Main Methods:

• Modification of the established affine model based on a simple neural circuit mechanism.
• Application of the modified model to analyze neural variability data within and between visual cortical areas (V1 and V2).

Main Results:

• The modified affine model successfully explains the contrast dependence of shared neural variability within V1 and V2.
• The model also accurately describes the orientation dependence of shared neural variability between V1 and V2.
• The findings demonstrate the model's efficacy in unifying explanations for stimulus-dependent variability across different contexts.

Conclusions:

• The study presents a parsimonious and unified statistical model for stimulus-dependent correlated variability in the visual cortex.
• The findings link neural circuit properties to observable neural variability patterns within and between visual areas.
• This work offers a framework for understanding how neural circuits encode stimulus information through correlated variability.