FitzHugh-Nagumo model

The FitzHugh-Nagumo model is a two-dimensional reduction of the four-dimensional Hodgkin-Huxley model, developed independently by Richard FitzHugh in 1961 and Jinichi Nagumo in 1962. It captures the essential dynamics of neural excitability — threshold behavior, rapid depolarization, and slow recovery — using only two variables: a fast activator (voltage) and a slow inhibitor (recovery). The model belongs to the class of relaxation oscillators and has become the canonical example of an excitable system in dynamical systems theory.

The Equations and Phase Plane

The FitzHugh-Nagumo equations are:

dv/dt = v − v³/3 − w + I dw/dt = ε(v + a − bw)

where v represents the membrane potential (the fast variable), w represents a recovery variable (the slow variable), I is an external stimulus current, and ε << 1 enforces the timescale separation. The cubic nullcline (dv/dt = 0) is N-shaped, while the linear nullcline (dw/dt = 0) is a straight line. Their intersection determines the fixed points of the system.

The geometry of the phase plane reveals everything. When the nullclines intersect on the left branch of the cubic, the fixed point is stable and the system is excitable: a small perturbation decays, but a perturbation exceeding threshold triggers a large excursion — the action potential — followed by a refractory period as the slow variable recovers. When the intersection moves to the middle branch (via increased stimulus I), the fixed point loses stability through a Hopf bifurcation and the system oscillates spontaneously. The transition from excitable to oscillatory is a canard explosion: over an exponentially small parameter interval, small oscillations inflate to full relaxation oscillations.

This phase plane geometry is not merely a visualization. It is the explanatory core. The Hodgkin-Huxley model produces the same qualitative behavior, but buried under four dimensions of ionic detail. FitzHugh's insight was that the four-dimensional dynamics collapse onto a two-dimensional slow manifold, and that the slow manifold's shape — not the specific biophysical mechanisms — determines the system's behavior.

From Neuron to Universal Parable

The FitzHugh-Nagumo model was invented to understand the squid axon, but its domain quickly expanded. In cardiac tissue, modified FitzHugh-Nagumo equations describe the propagation of electrical excitation across the myocardium — and the spiral waves that cause ventricular fibrillation. In chemical systems, the Belousov-Zhabotinsky reaction produces traveling pulses and target patterns that map onto FitzHugh-Nagumo dynamics. In ecology, predator-prey models with timescale separation exhibit the same excitable structure: a slow resource recovery followed by a rapid predator outbreak.

What unifies these applications is not the details but the geometry. Each system has a fast positive feedback (voltage-gated sodium channels, autocatalytic chemical reaction, predator reproduction) coupled to a slow negative feedback (potassium channel closure, reactant depletion, resource exhaustion). The FitzHugh-Nagumo model is the minimal mathematical skeleton of this architecture — the simplest system that can be excitable, oscillatory, or bistable depending on parameter values.

The model also plays a central role in the theory of phase response curves — the map describing how a periodic oscillator responds to perturbations at different phases. Because the FitzHugh-Nagumo model has a well-defined limit cycle with explicit slow and fast segments, it provides an analytically tractable case study for understanding synchronization in coupled oscillator networks.

The Reduction Controversy

Not everyone accepts the FitzHugh-Nagumo reduction as legitimate neuroscience. Critics argue that by collapsing four ionic variables into one recovery variable, the model discards biologically real phenomena: the distinct timescales of sodium inactivation and potassium activation, the effects of pharmacological blockers, the temperature dependence of channel kinetics. The FitzHugh-Nagumo model cannot distinguish between tetrodotoxin (which blocks sodium) and tetraethylammonium (which blocks potassium) — a distinction that the Hodgkin-Huxley model handles naturally.

This criticism is correct but misses the point. The FitzHugh-Nagumo model is not a competitor to Hodgkin-Huxley; it is a complement. Hodgkin-Huxley answers the question how