Computational Psychiatry Needs Time and Context

Abstract

Even the most ardent proponents of computational psychiatry admit that the field is far from influencing routine clinical practice. We propose one reason for this is that the field has had difficulty recognizing the variability among mental health problems—and the resulting need to model context and temporal dynamics for many problems. We develop three heuristics for estimating whether time and context are important to a mental health problem. Is it characterized by a core neurobiological mechanism? Does it follow a straightforward natural trajectory? And is intentional mental content peripheral to the problem? For many problems the answers are no, suggesting modeling time and context is critical. We review computational psychiatry advances toward this end, including modeling state variation, using domain-specific stimuli, and interpreting differences in context. We discuss complementary network and complex systems approaches. Novel methods and unification with adjacent fields may inspire a new generation of computational psychiatry

Why Is Psychiatry so Intimately Linked to the Brain?

“…This centralization of disciplines that are adjacent to psychiatry allows their didactic integration into clinical practice. The relevance of brain sciences for psychiatry is provided by the capacity to study a single organ, which is functionally described at different levels ( 6 ).…”

Studying mental health problems as systems, not syndromes

“…Because mental health categories and thresholds are manufactured rather than discovered, the question remains whether such designs are optimally positioned to inform research on psychopathology (Cai et al, 2020;Hitchcock et al, 2021). Of note, this position should not be mistaken view the view that mental disorders do not exist, do not involve suffering, or should not be treated.…”

“…NIMH director Insel concluded, after "13 years at NIMH really pushing on the neuroscience and genetics of mental disorders […] I succeeded at getting lots of really cool papers published by cool scientists at fairly large costs-I think $20 billion-I don't think we moved the needle in reducing suicide, reducing hospitalizations, improving recovery for the tens of millions of people who have mental illness" (Rogers, 2017). This lack of progress is not because biology is not crucially involved in predisposing or mediating disease trajectory (Cai et al, 2020;Hitchcock et al, 2021)-it arises because we have focused on studying the biology of particular DSM labels that are likely the wrong targets, and because we have studied biology (at least largely) in isolation. Diagnostic literalism and reductionism have formed a vicious cycle of reification, and continue to do so whenever we talk about "risk factors for schizophrenia", "genes for major depression", and "symptoms of PTSD".…”

The Hitchhiker’s Guide to Computational Linguistics in Suicide Prevention

“…Since the appearance of the biopsychosocial model (Engel, 1977), the scientific consensus supports the position that mental illnesses, in general, may result from complex interactions between a wide range of biological, psychological, interpersonal, and environmental risk factors (Lehman et al, 2017). As stated in the DSM-5, "in the absence of clear biological markers or clinically useful measurements of severity for many mental disorders" (American Psychiatric Association, 2013, p. 21), an accurate psychological assessment is an extremely difficult task both for humans and for computers ( Hitchcock et al, 2021). Not only can different causes (e.g., a specific trauma vs. a genetic predisposition or general poverty) lead to similar disorders, but also, in many cases, the manifestations of a given disorder can be quite heterogeneous.…”