Normative Modeling in Psychiatric Disorders

Normative modeling provides a principled way to characterise how individuals with psychiatric conditions diverge from population‐based expectations. Analogous to growth charts in paediatrics, these models use large healthy cohorts to map trajectories of brain structure and function across the lifespan. Individual neuroimaging measures—such as regional grey matter volume or cortical thickness—are benchmarked against reference curves to yield personalised deviation scores. This approach transcends traditional case–control designs by capturing the full spectrum of biological heterogeneity within and across diagnostic categories. By quantifying idiosyncratic patterns of atypicality and revealing common functional circuits, normative modeling lays the groundwork for precision psychiatry, enabling more refined stratification of patients and informing targeted interventions on a global scale.

Research from Nature Portfolio

Recent studies have applied multiscale normative models to large neuroimaging cohorts, revealing that individual patients with varied psychiatric diagnoses diverge significantly from population‐derived expectations of regional grey matter volume. Despite low spatial overlap of extreme deviations within diagnostic groups, common functional circuits such as the salience–ventral attention network emerge transdiagnostically, offering insight into shared pathophysiological mechanisms. Complementing these findings, normative reference charts for brain morphology across the lifespan have been developed by aggregating tens of thousands of MRI scans. These charts assign centile scores to individuals, benchmark structural trajectories from prenatal stages to older age, and highlight previously unreported neurodevelopmental milestones. By providing stable measures of atypical morphology, they enable standardised comparisons across studies and foster a quantitative framework for identifying deviations linked to psychiatric disorders.

Normative Modeling in Psychiatric Disorders publication trend

The graph below shows the total number of articles in normative modeling in psychiatric disorders across all publications each year (not limited to Nature Index journals).

Technical terms

Normative modelling: A statistical framework that establishes expected distributions of biological measures in healthy populations to quantify individual deviations.

Centile score: A ranking of an individual’s measure relative to a normative distribution, indicating the percentage of the population below that measure.

Transdiagnostic: An approach or finding that spans multiple diagnostic categories, highlighting shared mechanisms.

Heterogeneity: Variability within a clinical population in symptoms, biology or outcomes, challenging uniform case–control comparisons.

Salience–ventral attention network: A functional brain circuit implicated in detecting and orienting to salient stimuli, often altered across psychiatric conditions.

References

1. Regional, circuit and network heterogeneity of brain abnormalities in psychiatric disorders. Nature Neuroscience (2023).
2. Brain charts for the human lifespan. Nature (2022).
3. Understanding Heterogeneity in Clinical Cohorts Using Normative Models: Beyond Case-Control Studies. Biological Psychiatry (2016).
4. Conceptualizing mental disorders as deviations from normative functioning. Molecular Psychiatry (2019).
5. Charting brain growth and aging at high spatial precision. eLife (2022).
6. Individual differences v. the average patient: mapping the heterogeneity in ADHD using normative models. Psychological Medicine (2019).