Feed-forward inhibition as a buffer of the neuronal input-output relation

Area of Science:

• Computational neuroscience
• Neural circuit dynamics
• Synaptic plasticity

Background:

• Neuronal processing relies on the input-output (I/O) relationship between synaptic stimulation frequency and axonal firing rate.
• Steep I/O curves due to neuronal properties can limit information coding, causing abrupt shifts in output.
• Existing models struggle to explain robust signal integration in the face of neural noise.

Purpose of the Study:

• To investigate how feed-forward inhibition impacts neuronal input-output relationships.
• To determine if feed-forward inhibition can enhance the robustness of neuronal signal integration.
• To explore the modulatory capacity of synaptic parameters on circuit I/O properties.

Main Methods:

• Utilizing biophysically and anatomically realistic computational models of individual neurons.
• Simulating the effects of feed-forward inhibition on neuronal firing rate dynamics.
• Analyzing the modulation of I/O curves by altering synaptic weights and numbers.

Main Results:

• Feed-forward inhibition converts steep sigmoid I/O curves into a double-sigmoid, resembling buffer systems.
• A stabilized intermediate plateau in the I/O curve allows for robust integration of noisy synaptic inputs.
• The firing rate and dynamic range of this buffered response are independently tunable via synaptic parameters.

Conclusions:

• Feed-forward inhibition provides a mechanism for stabilizing neuronal firing rates and enhancing computational power.
• This circuit-level modulation offers a soft switch between digital and analog neural coding strategies.
• The findings suggest a significant increase in the computational capacity of neuronal integration through dynamic I/O control.