An Agent Based Model of Effects of Sleep Deprivation on Suicidality

The use of computational models to understand psychological phenomena, including suicidal behavior and outcomes, has become more common. Here we extend a formal computational model of suicide with social and state-dependent dynamics enabled by ABM architecture to evaluate their effects on suicidal thought and aversive internal state. The proposed model implements the General Escape Theory of Suicide [34], combined with aspects of the Interpersonal Theory of Suicide [33], particularly burdensomeness and connectedness. It is based on a system of differential equations proposed by Wang et al. [34] and extended by Engels [11]. The prevalence of suicidal thoughts and aversive internal states was compared in three conditions: a baseline model, a model that included sleep deprivation, and a model that included both state-dependent and social dynamics. We found that sleep deprivation significantly increased both outcomes, while the addition of social dynamics reduced them and increased resistance to suicidal thought. These dynamics qualitatively align with contemporary suicide research showing the protective effects of sleep and socializing. The results demonstrate the utility of ABMs in representing the interactions of internal, social, and state-dependent processes. The model may support simulation of group-level interventions, although future work is needed for quantitative validation and refinement.