Energy efficiency in a channel model for the spiking axon

Abstract

We examine the spiking axon as a communication channel. We develop a first principles channel model that encompasses the noise in the axon, which manifests itself as spike jitter, and the power consumption, which arises from the activity of the Na⁺–K⁺ pump. This model enables us to examine the trade-off between the information rate and power consumption. Using parameters from the frog myelinated axon, we determine the spike rate that corresponds to the maximum energy efficiency. This spike rate is consistent with experimental observations, which suggests that neural communication may have developed to maximize energy efficiency rather than information rate alone.