Characterizing ISI and sub-threshold membrane potential distributions: Ensemble of IF neurons with random squared-noise intensity. (April 2018)

• Record Type: Journal Article
• Authors: Kumar, Sanjeev; Karmeshu,
• Is Part Of: Bio systems. Volume 166(2018)
• Page Start: 43
• Page End: 49
• Publication Date: 2018-04
• Subjects: Ensemble of IF neurons -- Random squared-noise intensity -- ISI distribution -- Generalized K-distribution -- Monte-Carlo simulation -- Empirical validation; Biological systems -- Periodicals; Biology -- Periodicals; Biology -- Periodicals; Evolution -- Periodicals; Biologie -- Périodiques; Évolution -- Périodiques; 570
• DOI: 10.1016/j.biosystems.2018.02.005
• Languages: English
• ISSNs: 0303-2647

Abstract

Abstract: A theoretical investigation is presented that characterizes the emerging sub-threshold membrane potential and inter-spike interval (ISI) distributions of an ensemble of IF neurons that group together and fire together. The squared-noise intensity σ 2 of the ensemble of neurons is treated as a random variable to account for the electrophysiological variations across population of nearly identical neurons. Employing superstatistical framework, both ISI distribution and sub-threshold membrane potential distribution of neuronal ensemble are obtained in terms of generalized K-distribution. The resulting distributions exhibit asymptotic behavior akin to stretched exponential family. Extensive simulations of the underlying SDE with random σ 2 are carried out. The results are found to be in excellent agreement with the analytical results. The analysis has been extended to cover the case corresponding to independent random fluctuations in drift in addition to random squared-noise intensity. The novelty of the proposed analytical investigation for the ensemble of IF neurons is that it yields closed form expressions of probability distributions in terms of generalized K-distribution. Based on a record of spiking activity of thousands of neurons, the findings of the proposed model are validated. The squared-noise intensity σ 2 of identified neurons from the data is found to follow gamma distribution. The proposed generalized K-distribution is found to be in excellent agreement with that of empirically obtained ISI distribution of neuronal ensemble.