Beliefs and the brain: understanding hallucinations and delusions through predictive coding

Phil Corlett

Assistant Professor in Psychiatry, Yale School of Medicine

The problem of whether and how information is integrated across hierarchical brain networks embodies a fundamental tension in contemporary cognitive neuroscience, and by extension, cognitive neuropsychiatry. Indeed, the penetrability of perceptual processes in a ‘top-down’ manner by higher-level cognition—a natural extension of hierarchical models of perception—may contradict a strictly modular view of mental organization. Furthermore, some in the cognitive science community have challenged cognitive penetration as an unlikely, if not impossible, process. I will present behavioral and functional imaging evidence that perception is penetrated by top-down expectation and that hallucinations, the percepts without stimulus that characterize serious mental illnesses like schizophrenia are associated with excessively strong top-down expectation. This is the case in patients with a psychotic disorder as well as healthy voice hearers. However, patients without voices and matched controls do not show the effect. Indeed, people with delusions may be protected from the effect. I will discuss these findings in the context of predictive coding accounts of symptoms, and data linking delusions to weak conceptual priors. I will conclude that our data defy a simple explanation in terms of weaker priors in schizophrenia. Rather, the hierarchical level of a deficit may be crucial and the relationships between lower level sensory prediction errors and top down priors (mediated by their precision) may well govern whether a particular network disruption manifests as a hallucination or delusion and will contribute to the contents of the symptom.