Dynamics and synchronization of a Hopfield neural network with a locally active memristor

Locally active memristors play a vital role in neurodynamics by supporting signal generation, transmission, and regulation between neurons. This paper introduces a memristor-based coupled Hopfield neural network (HNN) to emulate brain-like information processing. A locally active memristor model is designed and integrated into an HNN with heterogeneous synaptic weights. Analysis reveals that the system exhibits multimodal chaotic attractors and transient chaos, and enables inter-subnetwork communication in cephalic regions through memristor-mediated energy transfer. Furthermore, increasing memristor coupling strength enhances neuronal synchronization, demonstrating effective dynamic regulation. This study advances neuromorphic dynamics and offers key insights into underlying mechanisms.