Memristor-Based Temporal Memory Neural Network Circuit Influenced by Emotional Arousal and Memory Interaction

Temporal memory is an important component of the memory system, which has the ability to remember a series of events in chronological order. Although simple temporal memory and recall processes are implemented by some memristor based neural networks. However, previous work has not considered the interaction between temporal stimuli and the effect of emotions on the memory and recall processes of temporal stimuli. A memristor based temporal memory neural network circuit influenced by emotional arousal and memory interaction is designed in this paper. The circuit is composed of hippocampus module, memory interaction module, memory threshold module, emotional arousal module and temporal recall module. Firstly, when there is correlation between temporal stimuli, the interaction effect function between memories is realized through hippocampus module and memory interaction module. Secondly, the memory and recall processes of emotional arousal to temporal stimuli in different emotional states are realized through the emotional arousal module, hippocampus module and temporal recall module. Finally, the functions of memory fatigue, memory saturation and time-distance effect are realized. The simulation results of PSPICE verify the above functions. This circuit provides more reference for the further development of brain-like system.