Abstract/Summary

We increase our probability of survival and wellbeing by minimizing our exposure to rare, extremely negative events. In this article, we examine the computations used to predict and avoid such events and to update our models of the world and action policies after their occurrence. We also consider how these computations might go wrong in anxiety disorders and Post Traumatic Stress Disorder (PTSD). We review evidence that anxiety is linked to increased simulations of the future occurrence of high cost negative events and to elevated estimates of the probability of occurrence of such events. We also review psychological theories of PTSD in the light of newer, computational models of updating through replay and simulation. We consider whether pathological levels of re-experiencing symptomatology might reflect problems reconciling the traumatic outcome with overly optimistic priors or difficulties terminating off-line simulation focused on negative events and over-generalization to states sharing features with those antecedent to the trauma.