Title : Psychic numbing and antimicrobial resistance: A coupled psycho-epidemiological agent-based model

Standard epidemiological models of antimicrobial resistance (AMR) often treat patient compliance as a fixed parameter, ignoring the nonlinear nature of human risk perception. We couple a four-compartment SISR model to a Weber-Fechner perceptual mechanism where risk perception scales logarithmically with infection prevalence. This cognitive driver generates two distinct behavioural responses: 1) treatment compliance and 2) symptomatic isolation. Structural sensitivity analysis reveals that previously assumed compliance-enhanced clearance mechanisms drive spurious bifurcations; our refined model relies strictly on transmission and mutation flux, producing a smooth threshold transition in equilibrium resistant burden. As the perceptual sensitivity parameter k increases, resistant burden falls from approximately 450 cases to 7 cases; a 60-fold reduction, with the steepest descent near  k = ~ 0.825. Channel-decomposition experiments reveal that symptomatic isolation accounts for nearly all of this reduction, while treatment compliance has no detectable effect at equilibrium, a result confirmed by cost-sensitivity heatmaps. Finally, a stochastic agent-based model (NetLogo 7.0.0) reproduces the ODE threshold curve, identifying apparent compliance effects at low N as finite-size artifacts. We conclude that symptomatic isolation, rather than antibiotic adherence, is the critical population-level lever eroded by "psychic numbing."

To be updated shortly..