Computational neuroscience of social behavior and psychopathology.

People playing economic games in a brain scanner will reveal how social interactions go wrong in mental illness. This research addresses a fundamental gap in psychiatry: most diagnoses rely on subjective symptom checklists rather than objective biological measures. The team uses interpersonal games—like bargaining or trust exercises—as probes to watch the brain compute social decisions in real time. By combining neuroimaging with computational models, they aim to identify distinct neural signatures—called endophenotypes—for conditions that involve impaired social exchange, such as autism or schizophrenia. If successful, these brain-based assays could transform how psychopathologies are diagnosed and treated. Clinicians might one day use a patient’s neural response during a social game to differentiate between overlapping diagnoses, predict which therapy will work, or track whether a drug is repairing the underlying circuitry. The work is fundamental science: it seeks to map the computational steps the brain uses to navigate social interactions, then link breakdowns in those steps to specific disorders. That mechanistic understanding could eventually lead to more precise, biologically grounded treatments—moving psychiatry beyond symptom checklists toward a neuroscience of social behaviour.