Algorithmic Circuit Psychiatry: A Roadmap for Precision Treatment of Serious Mental Illness

Treating serious mental illness is a formidable challenge, in part because current diagnostic categories may not capture the biological processes determining treatment response. Progress in genetics, systems neuroscience and neuroimaging has produced important mechanistic insights, yet these advances have not yielded a practical framework for precision psychiatry. We propose Algorithmic Circuit Psychiatry (ACP) as a roadmap for precision treatment of serious mental illness, where stratification is realized at the level of algorithms: computational processes identified via clinically relevant tasks and mappable onto neural substrates. Using psychotic spectrum disorders as a primary case study, we argue that algorithmic description may connect circuit mechanisms to clinical symptoms more effectively than isolated neural measures. We review evidence showing how frontal thalamocortical interactions support reasoning under uncertainty, how disruption of the relevant networks produces schizophrenia-relevant behavioral signatures, and how corresponding behavioral and neural readouts may serve as algorithmic circuit biomarkers. We then outline how these biomarkers, combined with pharmacological perturbation, support trial design logic that can yield mechanistic insights regardless of whether a trial achieves its primary endpoint.