×

zbMATH — the first resource for mathematics

Critical analysis of dimension reduction by a moment closure method in a population density approach to neural network modeling. (English) Zbl 1131.92016
Summary: Computational techniques within the population density function (PDF) framework have provided time-saving alternatives to classical Monte Carlo simulations of neural network activity. Efficiency of the PDF method is lost as the underlying neuron model is made more realistic and the number of state variables increases. In a detailed theoretical and computational study, we elucidate strengths and weaknesses of dimension reduction by a particular moment closure method [D. Cai et al., Commun. Math. Sci. 4, No. 1, 97–127 (2006; Zbl 1107.82037)] as applied to integrate-and-fire neurons that receive excitatory synaptic input only. When the unitary postsynaptic conductance event has a single-exponential time course, the evolution equation for the PDF is a partial differential integral equation in two state variables, voltage and excitatory conductance. In the moment closure method, one approximates the conditional \(k\) th centered moment of excitatory conductance given voltage by the corresponding unconditioned moment. The result is a system of \(k\) coupled partial differential equations with one state variable, voltage, and k coupled ordinary differential equations.
Moment closure at \(k=2\) works well, and at \(k=3\) works even better, in the regime of high dynamically varying synaptic input rates. Both closures break down at lower synaptic input rates. Phase-plane analysis of the \(k=2\) problem with typical parameters proves, and reveals why, no steady-state solutions exist below a synaptic input rate that gives a firing rate of 59 s\(^{-1}\) in the full 2D problem. Closure at \(k=3\) fails for similar reasons. Low firing-rate solutions can be obtained only with parameters for the amplitude or kinetics (or both) of the unitary postsynaptic conductance event that are on the edge of the physiological range. We conclude that this dimension-reduction method gives ill-posed problems for a wide range of physiological parameters, and we suggest future directions.

MSC:
92C20 Neural biology
35Q92 PDEs in connection with biology, chemistry and other natural sciences
37N25 Dynamical systems in biology
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] DOI: 10.1103/PhysRevE.48.1483
[2] DOI: 10.1093/cercor/7.3.237
[3] DOI: 10.1080/09548980601069787
[4] DOI: 10.1007/BF01026608
[5] DOI: 10.1002/cpa.3160460503 · Zbl 0817.76002
[6] Biella G., Journal of Neurophysiology 86 pp 164– (2001)
[7] DOI: 10.1162/089976604322860668 · Zbl 1054.92006
[8] DOI: 10.1023/A:1008925309027 · Zbl 1036.92008
[9] DOI: 10.1103/PhysRevLett.86.2186
[10] DOI: 10.1162/089976699300016179
[11] DOI: 10.4310/CMS.2006.v4.n1.a4 · Zbl 1107.82037
[12] DOI: 10.1073/pnas.0401906101
[13] DOI: 10.1162/089976602753633349 · Zbl 0995.92015
[14] DOI: 10.1103/PhysRevE.74.030903
[15] DOI: 10.1007/s00161-003-0171-z · Zbl 1100.82025
[16] Dreyer W., Institute of Physics Publishing 14 pp 881– (2001)
[17] DOI: 10.1146/annurev.ph.57.030195.002431
[18] DOI: 10.1162/089976602320264015 · Zbl 1009.92007
[19] Fourcard-Trocme N., Journal of Neuroscience 23 pp 11628– (2003)
[20] DOI: 10.1016/S0896-6273(00)00133-1
[21] DOI: 10.1103/PhysRevE.51.738
[22] DOI: 10.1162/089976600300015899
[23] DOI: 10.1162/089976698300017845
[24] DOI: 10.1088/0954-898X/12/2/304 · Zbl 1005.92007
[25] DOI: 10.1007/s10827-006-7753-2 · Zbl 1099.92010
[26] DOI: 10.1109/TNN.2004.832719
[27] DOI: 10.1152/jn.00190.2004
[28] DOI: 10.1007/s00161-002-0096-y · Zbl 1024.76044
[29] DOI: 10.1007/s00422-002-0390-6 · Zbl 1077.92009
[30] DOI: 10.1085/jgp.59.6.734
[31] DOI: 10.1162/089976600300015673
[32] DOI: 10.1162/089976600300015493
[33] DOI: 10.1038/86084
[34] DOI: 10.1016/0167-2789(91)90075-K · Zbl 0736.92001
[35] DOI: 10.1103/PhysRevE.66.051917
[36] DOI: 10.1016/j.jphysparis.2003.09.019
[37] DOI: 10.1073/pnas.110135097
[38] McNaughton B. L., Journal of Neurophysiology 46 pp 952– (1981)
[39] DOI: 10.1152/jn.00491.2005
[40] DOI: 10.1103/PhysRevLett.92.028102
[41] DOI: 10.1103/PhysRevLett.94.088103
[42] DOI: 10.1023/A:1008912914816 · Zbl 0999.92008
[43] DOI: 10.1162/089976601300014448 · Zbl 1052.92016
[44] DOI: 10.1023/A:1008964915724 · Zbl 1036.92010
[45] DOI: 10.1103/PhysRevLett.96.178101
[46] DOI: 10.1007/s10827-006-8526-7 · Zbl 05193544
[47] DOI: 10.1073/pnas.0509481102
[48] Shadlen M. N., Journal of Neuroscience 18 pp 3870– (1998)
[49] DOI: 10.1113/jphysiol.2004.060962
[50] DOI: 10.1088/0954-898X/14/2/305
[51] DOI: 10.1137/S0036139998344921 · Zbl 0991.92005
[52] Somers D. C., Journal of Neuroscience 15 pp 5448– (1995)
[53] DOI: 10.1007/s10827-005-5474-6 · Zbl 02224944
[54] Thompson A. M., Journal of Neurophysiology 60 pp 1896– (1988)
[55] DOI: 10.1088/0954-898X/4/3/002 · Zbl 0798.92009
[56] DOI: 10.1007/s10827-005-2491-4 · Zbl 05028083
[57] Wang X. J., Journal of Neuroscience 19 pp 9587– (1999)
[58] DOI: 10.1016/S0167-8760(00)00173-2
[59] DOI: 10.1016/S0022-5193(83)80013-7
[60] DOI: 10.1016/S0006-3495(72)86068-5
[61] Wilson H. R., Biological Cybernetics 13 pp 55– (1973)
[62] Yoshimura Y., Journal of Neuroscience 20 pp 1931– (2000)
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.