2021/03/27 by Saeed Aljaberi, Adriano Bellotti, Aljaberi, Saeed +5
Engineering · Neuroscience · #Adaptation and Self-Organizing Systems (nlin.AO) #Advanced Memory and Neural Computing #FOS: Electrical engineering #FOS: Physical sciences #Neural dynamics and brain function #Neuroscience and Neuropharmacology Research #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2103.15001
openalex publication_date 2021/03/27 · openalex created_date 2021/04/13 · openalex updated_date 2026/07/28
Neurons regulate the distribution of signaling components across an extended tree-like cellular structure using both local and global feedback control. This is hypothesized to allow homeostatic control of the electrical activity of a neuron and at the same time enable normalization of distribution of inputs received from other cells. The performance and robustness of these mechanisms are poorly understood, and are subject to nonlinearities, making their analysis difficult. Firstly, we formally show that global homeostasis of electrical activity and local activity-dependent degradation can coexist under sufficient timescale separation. The interplay of the two feedback mechanisms is also analyzed through simulations, which reveal a bidirectional effect (stabilizing and destabilizing) of activity-dependent degradation on the overall neuron performance.