Studying relationships from the perspective of chaos theory

The study of relationship emotions, a set of emotional and psychological responses that arise in a relationship, can help develop more humanized artificial intelligence, improve human–computer interaction, and even create more immersive experiences in virtual and augmented reality. Due to the nonlinear and feedback-driven nature of relational affect, which aligns closely with chaos theory, and the ability of leaky integrate-and-fire (LIF) neuron models to simulate dopamine-related electrical activity in brain nuclei, this study innovatively integrates both approaches. By linking the membrane potential signals of LIF neurons to relational affect equations, it achieves a refined modeling of the mechanisms underlying relational affect generation. This paper adds the LIF neuron model to the relationship emotion model to construct a new LIF relationship emotion model (LRM). The effect of the parameters in the LRM on the relationship emotions generated by the model is investigated using numerical analysis. This includes the firing behavior produced by LIF neurons and a study of relationship emotions produced by different initial relationship emotion states under the same conditions. Finally, the feasibility of LRM is verified using a digital signal processing (DSP) platform. This process not only verifies the feasibility of LRM but also provides new ideas and methods for future research in affective computing and human–computer interaction.