Bayesian Psychiatry and the Social Focus of Delusions

Abstract

Computational psychiatry has recently emerged as a leading theoretical approach for understanding psychiatric disorders, employing Bayesian modelling to decipher underlying dysfunctions. Whilst this approach has clear potential, we argue that it overlooks the uniquely human aspects of human psychopathology. Its emphasis on abstract, domain-general inferential processes often obscures the complex interplay of social and personal phenomena that characterise disorders. This chapter explores these limitations with reference to psychosis and delusions. It demonstrates that an over-reliance on abstract computational models masks the rich social phenomenology that characterise these phenomena. By correcting these blind spots, we pave the way for an integrated approach that combines the rigour of computational models with a richer framework for understanding many of the distinctive ways in which the human mind can break down and malfunction. This will ensure that the promise of computational psychiatry fully materialises, capturing the depth and breadth of psychiatric disorders.