Related Experiment Videos
Dissection and reduction of a modeled bursting neuron
R J Butera1, J W Clark, J H Byrne
1Dept. of Electrical and Computer Engineering, Rice University, Houston, TX 77251-1892, USA.rbutera@ece.rice.edu
Journal of Computational Neuroscience
|September 1, 1996
Summary
Developing reduced models of bursting neurons requires accounting for action potentials. A 4-variable model accurately predicted neural activity modes and transient responses, unlike simpler models.
Area of Science:
- Computational neuroscience
- Mathematical biology
- Systems neuroscience
Background:
- Hodgkin-Huxley models describe neuronal electrical activity.
- Bursting neurons exhibit complex firing patterns.
- Reduced models simplify complex biological systems for analysis.
Purpose of the Study:
- To develop and validate low-order models of bursting neurons.
- To investigate the impact of action potentials on reduced model accuracy.
- To compare the predictive power of different reduced models.
Main Methods:
- Numerical bifurcation analysis of an 11-variable Hodgkin-Huxley model.
- Development of 3-variable and 4-variable reduced models.
- Computer simulations to assess model predictions under various stimuli.
Main Results:
- A 3-variable model mimicked subthreshold oscillations but failed to capture action potential effects.
- A 4-variable model, incorporating action potential effects, accurately predicted activity modes (bursting, beating, silence).
- The 4-variable model demonstrated high fidelity in phase-response curves compared to the original model.
Conclusions:
- Low-order models neglecting action potential effects can lead to inaccurate predictions of bursting neuron activity.
- Accurate reduced models of bursting neurons must account for the influence of action potentials.
- It is feasible to create simplified models retaining key characteristics of complex neuronal dynamics.