How to correctly quantify neuronal phase-response curves from noisy recordings

Area of Science:

• Computational Neuroscience
• Electrophysiology
• Neuronal Dynamics

Background:

• Neuronal coding properties are inferred from input-output relationships.
• Phase-response curves (PRCs) quantify the effect of small perturbations on spike timing.
• Experimental estimation of PRCs is often challenging and inconsistent with theoretical models.

Purpose of the Study:

• To investigate the impact of stimulus strength on experimental PRC estimation.
• To identify conditions under which experimental PRCs accurately reflect theoretical predictions, specifically the spike-onset bifurcation.
• To provide guidelines for selecting appropriate stimulus strengths for reliable PRC estimation.

Main Methods:

• Simulations of conductance-based model neurons.
• Analysis of the input-output relationship under varying perturbation strengths.
• Comparison of experimentally derived PRCs with theoretical PRCs.

Main Results:

• Large stimuli used for PRC estimation distort the measured curve, masking the underlying spike-onset bifurcation.
• Distorted PRCs exhibit a generic shape dependent on the measurement protocol.
• Theoretical and experimental PRCs only correspond when small, appropriate stimulus strengths are used.

Conclusions:

• The accuracy of experimental PRC estimation is highly dependent on stimulus amplitude.
• Careful selection of stimulus strength is crucial for capturing the spike-onset bifurcation.
• Methods for estimating PRCs must account for neuronal noise and select stimuli that avoid distortion.