Memristor-Based Neural Network Circuit of Associative Memory With Occasion Setting

Classical and operant conditioning are rarely studied together in the neural network based on memristor. But in real life, they often happen at the same time. This article proposes a neural network circuit based on the memristor of associative memory with occasion setting. The designed circuit includes four modules: 1) an operant synapse module; 2) a response module; 3) a classical synapse module; and 4) a perceptron module. The first two modules constitute operant conditioning, and the remaining two modules constitute classical conditioning. The designed circuit combines operant conditioning with classical conditioning to realize a special associative memory through an occasion setting. Besides, it implements the function of learning, forgetting, and long-term memory on the basis of associative memory with occasion setting. The neural network circuit of associative memory with occasion setting is closer to the real situation and provides a reference for classical and operant conditioning.