The burdens of mental illness on both individuals and society are painful and immense. For instance, in the U.S., the mortality from suicide is twice that of homicide and HIV; a conservative estimate of the financial costs of mental illness exceeds $300 billion annually [1]. Yet psychiatry is in a crisis, as the development of new treatments has stalled. A large part of the problem is that the diagnostic definitions of mental illnesses are based on phenomenological factors, rather than biological underpinnings. In other areas of medicine – whether cancer or diabetes – a biological understanding of disease allows doctors and scientists to distinguish dysfunctions associated with specific body components, and to treat them appropriately. When such biological mechanistic understanding is lacking, as with schizophrenia (which appears to reflect several biologically distinct dysfunctions) and depression and anxiety (which may both arise from a common vulnerability) effective treatment development and decisions are hobbled.

Connecting psychiatric disorders to the function of the biological organ that supports them, the brain, has been a great challenge. This in part reflects the challenge of neuroscience generally: the great complexity of the brain and the difficulty bridging neural machinery to complex human behaviors of the sort affected by mental illness. However, recent advances in cognitive and computational neuroscience provide a new, often quantitative and precise, understanding of how multiple aspects of brain function support complex processes such as perception, memory, decision-making and executive control. Because these mechanisms contribute heavily to evaluation, anxiety, mood, control and compulsion – precisely the sorts of functions affected by mental illnesses such as schizophrenia, post-traumatic stress disorder, depression and drug abuse – they provide a promising new foundation for a neuroscientific understanding of these illnesses. Indeed, alterations in such functions have been identified as ‘endophenotypes’ that are common to many disorders. Computational psychiatry is well-suited for the identification and characterization of endophenotyopes, and promises to provide a new level of biological understanding of these disease-related dysfunctions.

The Rutgers-Princeton Center for Cognitive Computational Neuropsychiatry has been formed to pursue this exciting opportunity. The goal is to leverage the expertise in Princeton’s department of Psychology and Neuroscience Institute, and in Rutgers’ departments of Psychology, Psychiatry and Computer Science, Rutgers University Behavioral Health Care, Robert Wood Johnson Hospital, and the Rutgers Brain Health Institute, in a major collaborative initiative that has the potential to be much greater than the sum of its individual parts.