Testing the role of belief updating in persecutory delusions

Persecutory delusions, or the strongly held belief that others intend me harm, are distressing and disabling transdiagnostic symptoms. They are the most common manifestation of delusions in psychotic disorders, present in over 70% of cases, and exist at the extreme end of the paranoia spectrum. Persecutory delusions are a leading cause of suicidal ideation and hospitalization amongst individuals with schizophrenia, yet effective, sustainable treatments remain limited. One potential route towards developing new treatments is identification of cognitive processes that directly contribute their maintenance. Predictive coding is a prominent mechanistic account of delusions based in the process of belief updating, which describes how we learn about the world and develop new beliefs. Development of adaptive, rational beliefs depends on an accurate understanding of the volatility of the environment, or how frequently the probabilities underlying the environment change. Over-estimation of volatility (i.e. inferring that the environment has changed when it has not) drives the formation of new beliefs that are based on faulty inference. There is growing support that aspects of volatility-related belief updating contribute to paranoia and persecutory delusions specifically. Longitudinal data from the PI’s K23 newly demonstrate that volatility-related belief updating is abnormal in delusional patients with schizophrenia, normalizes with symptom improvement, and is associated with activation in specific brain regions. Yet, these data remain limited by their correlational nature. Randomized controlled trials are the gold standard of causal inference testing, drawing more definitive conclusions about mechanism. Therefore, to push this research forward and directly test the belief updating model of delusions, we will recruit 120 individuals with a schizophrenia-spectrum disorder endorsing a persistent persecutory delusion and randomize them to receive either formulation-driven cognitive-behavioral therapy for persecutory delusions (CBTp), which has been previously shown to reduce delusion severity, or an active control therapy (befriending). We will then test whether modification of persecutory delusion severity using psychotherapy impacts belief updating at the level of behavior (Aim 1) and neurobiology (Aim 2), investigating specific brain circuits underlying distinct volatility parameters. Our ultimate goal is to leverage the insights gained into new, mechanistically informed treatments for persecutory delusions and paranoia, translating computational neuroscience to clinical applications.