Enhanced Sensitivity to Rapid Input Fluctuations by Nonlinear Threshold Dynamics in Neocortical Pyramidal Neurons

Area of Science:

• Computational neuroscience
• Biophysics
• Neural coding

Background:

• Standard Integrate-and-Fire models assume reduced sensitivity to rapid fluctuations with increased input strength.
• Understanding how single neurons process input into output spike trains is crucial for network function.

Purpose of the Study:

• To investigate the adaptive mechanisms in L5 pyramidal neurons that preserve sensitivity to rapid input signals.
• To develop a more accurate neuron model that captures adaptive firing threshold dynamics.

Main Methods:

• Mathematical modeling combined with in vitro experiments on L5 pyramidal neurons.
• Introduction of a Generalized Integrate-and-Fire model with nonlinear firing threshold dynamics and conductance-based adaptation.
• Comparison of the new model's predictive power against state-of-the-art neuron models using in vivo-like fluctuating currents.

Main Results:

• Demonstrated that firing threshold dynamics adapt somatic integration timescales to preserve sensitivity to rapid signals.
• Developed a Generalized Integrate-and-Fire model that accurately predicts spiking activity across diverse input statistics.
• Showed that the model can be mapped to a Generalized Linear Model with dynamically adapting input and spike-history filters.

Conclusions:

• Nonlinear firing threshold dynamics, driven by Na+-channel inactivation, are key to regulating sensitivity to rapid input fluctuations.
• The findings offer insights into the computational roles of biophysical processes in adaptive coding.
• The developed model provides a framework for understanding and predicting neural responses to complex inputs.