Thresholds and Plateaus in the Hodgkin-Huxley Nerve Equations

Abstract

Phase space methods and an analog computer are used to analyze the Hodgkin-Huxley non-linear differential equations for the squid giant axon membrane. V is the membrane potential, m the Na+ activation, h the Na+ inactivation, and n the K+ activation. V and m change rapidly, relative to h and n. The (V, m) phase plane of a reduced system of equations, with h and n held constant at their resting values, has three singular points: a stable resting point, a threshold saddle point, and a stable excited point. When h and n are allowed to vary, recovery and refractoriness result from the movement with subsequent disappearance of the threshold and excited points. Multiplying the time constant of n by 100 or more, and that of h by one-third, reproduces the experimental plateau action potentials obtained with tetraethylammonium by Tasaki and Hagiwara, including the phenomena of abolition and of refractoriness of the plateau duration. The equations have, transiently, two stable states, as found in the real axon by these authors. Since the theoretical membrane conductance curves differ significantly from the experimental ones, further experimental analysis of ionic currents with tetraethylammonium is needed to decide whether the Hodgkin-Huxley model can be generalized to explain these experiments completely.

A two-current model for the dynamics of cardiac membrane

“…The two-current model is similar in spirit to the FitzHugh-Nagumo model (FitzHugh, 1960(FitzHugh, , 1961), but we believe that the two-current model is more closely related to the physiology of the heart for two reasons. A minor issue: the twocurrent model does not exhibit voltage overshoot, a phenomenon observed in neural but not cardiac tissue.…”

Solitary wave solution of nonlinear multi-dimensional wave equation by bilinear transformation method

“…; a n j Þ 2 R n , (j = 1, 2, 3), are three distinct constant vectors, and ðU À a 1 ; U À a 2 Þ P n i¼1 ðU i À a i 1 ÞðU i À a i 2 Þ means the Euclidean scalar product of vectors (U À a 1 ) and (U À a 2 ) and $ 2 = D = o 2 /ox 2 + o 2 /oy 2 + o 2 /oz 2 is the Laplace operator, a, b are real constants. In the scalar case, when n = 1, a = 0, and in the one space dimension, for different choices of parameters a 1 , a 2 , a 3 the model reduces to the well-known nonlinear diffusion equations appearing in a different fields sometimes with different names: the Fitzhugh-Nagumo equation (a 1 = 0, a 2 = 1, a 3 = a) arising in population genetics [9] and models the transmission of nerve impulse [10], autocatalytic chemical reaction model introduced by Schlögl [11,12], generalized Fisher equation [2], Newell-Whitehead equation [13] or Kolmogorov-Petrovsky-Piscounov equation [14] (a 1 = 0, a 2 = 1, a 3 = À 1), Huxley equation (a 1 = a 2 = 0, a 3 = 1).…”

Generation of computational meshes from MRI and CT-scan data

“…A classical Fitzhugh-Nagumo PDE model [4,10] has been implemented here as a first attempt to perform numerical simulations on the computational meshes. The figures show what result can be obtained with a conventional computer equipment in a few hours.…”