Efficient information transfer by Poisson neurons

Recently, it has been suggested that certain neurons with Poissonianspiking statistics may communicate by discontinuously switchingbetween two levels of firing intensity. Such a situation resembles inmany ways the optimal information transmission protocol for thecontinuous-time Poisson channel known...

Full description

Saved in:
Bibliographic Details
Main Authors: Lubomir Kostal, Shigeru Shinomoto
Format: Article
Language:English
Published: AIMS Press 2015-12-01
Series:Mathematical Biosciences and Engineering
Subjects:
Online Access:https://www.aimspress.com/article/doi/10.3934/mbe.2016004
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recently, it has been suggested that certain neurons with Poissonianspiking statistics may communicate by discontinuously switchingbetween two levels of firing intensity. Such a situation resembles inmany ways the optimal information transmission protocol for thecontinuous-time Poisson channel known from information theory. In thiscontribution we employ the classical information-theoretic results toanalyze the efficiency of such a transmission from differentperspectives, emphasising the neurobiological viewpoint. We addressboth the ultimate limits, in terms of the information capacity undermetabolic cost constraints, and the achievable bounds on performanceat rates below capacity with fixed decoding error probability. Indoing so we discuss optimal values of experimentally measurablequantities that can be compared with the actual neuronal recordings ina future effort.
ISSN:1551-0018