Predictive coding in neuropsychiatric disorders: A systematic transdiagnostic review.

Area of Science:

• Cognitive Neuroscience
• Neuropsychology
• Psychiatry

Background:

• The predictive coding framework proposes the brain generates predictions to minimize prediction errors based on experience.
• Understanding predictive coding variations is crucial for diagnosing and treating neuropsychiatric disorders.

Purpose of the Study:

• To systematically review and compare predictive coding differences between individuals with neuropsychiatric disorders and healthy controls.
• To examine predictive coding transdiagnostically across various neuropsychiatric conditions.

Main Methods:

• A PRISMA-compliant systematic review of 72 articles.
• Inclusion of case-control studies focusing on predictive coding outcomes (behavioral, neuroimaging, electrophysiological).
• Analysis of studies across schizophrenia spectrum, neurodevelopmental, mood, neurocognitive, PTSD, and substance use disorders.

Main Results:

• Heterogeneous impairments in predictive coding were observed across neuropsychiatric disorders.
• Schizophrenia spectrum disorders showed consistent deficits in non-social predictive coding.
• Autism spectrum disorder exhibited selective deficits in social cue predictive coding; impairments correlated with symptom severity.

Conclusions:

• Predictive coding offers a valuable framework for understanding cognitive dysfunction in neuropsychiatric populations.
• Further research is needed on understudied conditions and potential confounders (medication, sex/gender).
• Predictive coding may inform personalized treatment interventions and predict treatment response.