Spiking neural P systems with mute rules

Abstract

Spiking neural P (SNP) systems are a class of neural network models that draw inspiration from the functioning of biological neurons. It was experimentally found that there exist autapses from neurons onto themselves in the brain, i.e., a neuron can transmit a signal back to itself through an autapse. In this work, inspired by the characteristics of autapses, a new variant of the SNP system, termed SNP systems with mute rules (SNPMR systems), is considered. Specifically, mute rules have no communication functioning, namely the execution of a mute rule only applies the change on the number of spikes within its residing neuron, rather than affecting other postsynaptic neurons. The computational power of SNPMR systems is examined by demonstrating that SNPMR systems achieve Turing universality with four or ten neurons. In addition, a simulator for SNPMR systems is developed to provide an experimental validation of the systems designed theoretically.