Memristive biophysical neuron models forming an excitatory–inhibitory neural network for modeling PING rhythm generation

Abstract

SPICE models are constructed for memristive devices to form associated biophysical neuron circuit models such as the Hodgkin–Huxley (HH) type II excitability neuron circuit model, the HH type III excitability neuron circuit model, the simplified HH neuron circuit model, the Morris–Lecar neuron circuit model, and the memristive based direct-current (DC) circuit model. Rigorous nonlinear circuit-theoretic principles are also applied to analyze the different behaviors of the generic memristor Na $$^{+}$$ -ion, K $$^{+}$$ -ion, and Ca $$^{++}$$ -ion channels forming these biophysical neuron circuit models. Detailed explanations and clarifications are presented on the memristive HH type II and HH type III axonal excitabilities based on mathematical analysis as well as the circuit models. This is done from the perspective of the spike patterns generated by both of these biophysical neuron circuit models. Moreover, various experimental studies have revealed a synchronous brain state known as gamma rhythms that are responsible for sensory, memory, and motor processes. This suggests that understanding how the gamma oscillation (30–100 Hz) is generated in the brain will be extremely important to unravel the link between the activity of an individual neuron and the cognitive processing achieved by a population of networked neurons. We thus also study the dynamics of an interconnected excitatory–inhibitory (E–I) network population, which is ubiquitous in the brain. Utilizing biophysical models of the E–I network, we investigate the generation of pyramidal-interneuronal network gamma (PING) rhythms caused by the external input to the network and the connectivity heterogeneities. The results reveal that synchronous strong PING and sparsely firing weak PING rhythms are generated based on the network connectivities and external input heterogeneities in simulations of 100 memristive HH type II excitability neurons forming an E–I network.