Characteristics and dynamical signatures of recurrent cortical circuits during context-dependent processing.

Area of Science:

• Computational Neuroscience
• Systems Neuroscience
• Neurobiology of Vision

Background:

• Context significantly influences neural processing and behavior.
• Recurrent connections are crucial for integrating sensory inputs and feedback.
• Understanding cell-type-specific contributions to context-dependent processing remains a challenge.

Purpose of the Study:

• To investigate how different cell types interact via recurrent connections to achieve context-dependent processing and circuit stability.
• To identify dynamical signatures that reveal the roles of individual cell types.
• To model neural responses in the mouse primary visual cortex during context-dependent processing.

Main Methods:

• Development of data-driven, spatially extended, stabilized supralinear network models.
• Analysis of models capturing diverse cell type responses in the primary visual cortex.
• Utilized patterned perturbations to assess cell-type-specific circuit stabilization.

Main Results:

• The dominant inhibitory cell type influencing excitatory neurons varies dynamically with stimulus and space.
• Both Parvalbumin (PV)-mediated and Somatostatin (SST)-mediated stabilization are essential for circuit stability, with SST-dependent stabilization being stimulus-dependent.
• Patterned, not uniform, perturbations are necessary to reveal paradoxical effects in cell-type-specific circuit stabilization.

Conclusions:

• Recurrent connections and nonlinearities are vital for integrating feedforward and feedback inputs, reproducing spatial response profiles.
• Recurrent excitatory connections are critical for processing weak external inputs, especially for small stimuli.
• Ubiquitous biological components play essential roles in context-dependent processing, with specific dynamical signatures characterizing cell-type functions.