Curr Top Behav Neurosci. 2025 Aug 13. doi: 10.1007/7854_2025_596. Online ahead of print.
ABSTRACT
Like other medical illnesses, psychiatric conditions are defined by the presence of characteristic signs and symptoms that cause distress and dysfunction. Unlike most medical illnesses, however, the pathophysiological mechanisms underlying psychiatric symptoms remain largely unknown. Recent work in theory-driven computational neuroscience has proposed a unified account of how organisms derive estimates of both external and internal states by integrating prior beliefs with sensory input. This framework offers a powerful means of explaining typical perception and action, while also identifying how these processes may go awry to produce a panoply of psychiatric symptoms. Disruptions in this inferential process, which we term perceptual dysregulation, span exteroceptive, proprioceptive, and interoceptive systems and manifest across conditions ranging from psychosis and autism to anxiety and trauma-related disorders, functional neurological disorder, and dementia. This volume reviews emerging evidence for maladaptive sensory inference across psychiatric and neurological illness and explores how computational models based on Bayesian statistics can illuminate shared mechanisms beneath diverse symptom presentations. We argue that targeting perceptual dysregulation may not only enable new interventions but also catalyze the development of a revised psychiatric nosology grounded in testable models of brain-body interaction.
PMID:40794380 | DOI:10.1007/7854_2025_596
