Synchronization performance of memristive photosensitive thermosensitive neuron model in multi-architecture neural networks

At the forefront of spiking neural network (SNN) optimization and brain-computer interface (BCI) technology research, synchronization studies of FitzHugh-Nagumo (FHN) neural networks have consistently demonstrated broad application prospects. The signal transmission between neurons within neural networks affects the dynamic characteristics of ion channels on neuronal membranes, thereby altering their firing patterns. Light, temperature, and the magnetization and polarization of the medium under electromagnetic fields can significantly influence neuronal activity. By incorporating photoelectric tubes, thermistors and memristors, we can effectively evaluate how these external factors affect neuronal discharge patterns and neural network synchronization efficiency. This study integrates photoelectric tubes, thermistors, electric field-controlled memristors (EFM) and magnetic field-controlled memristors (MFM) into FitzHugh-Nagumo (FHN) neural circuits to construct a novel multimodal neural circuit - the memristive photosensitive thermosensitive neuron model (MPTN) - designed to investigate neural network synchronization under complex environmental conditions. Experimental results reveal the synchronization regulation mechanisms of memristors in the MPTN model and demonstrate the synchronization behavior of MPTN networks under chaotic current interference. Furthermore, to better simulate biological neural networks, various network architectures are employed to explore collective behaviors and synchronization performance under complex environmental conditions.