Consider these contrasts identified in Krueger et al's paper1: authoritative vs. empirical, ex cathedra (dogmatic) vs. evidence‐based, and tradition vs. empiricism. It is powerful verbiage, suggesting that the members of the Hierarchical Taxonomy of Psychopathology (HiTOP) consortium are arguing for their generation's Copernican turn in distinction to the system that preceded it.
Freud famously described his achievement as an intellectual revolution2, the neo‐kraepelinians used revolutionary idioms against the psychoanalysts that preceded them3, and now they are being used against the neo‐kraepelinians.
Doubtlessly, readers will have a range of reactions to these contrasts. If the reaction “this is a coup by clinical researchers to replace the DSM and ICD with the factor analytic, dimensional models used in psychological testing” formed one end of a bipolar continuum, the other end would be “this is a heroic scientific revolution”. I doubt many readers would assent to either pole wholeheartedly, but they may lean more toward one side or the other. I will argue that “coup” is too antagonistic an attribution and the proposed transition would be more appealing to psychiatrists if diplomatic alternatives to the “revolution” metaphor were used.
With respect to the coup, importing research traditions from scientific psychology into psychiatry not only has historical precedent; it has been historically important. To illustrate, consider E. Kraepelin, a groundbreaking architect of psychiatric classification, and R. Spitzer, who was the driving force behind the DSM‐III and DSM‐III‐R.
Kraepelin's career plan was inspired by his contact with the founder of scientific psychology, W. Wundt. From his earliest days in the field, Kraepelin wanted to orient psychiatry away from speculative anatomical hypotheses and reductionism, and replace them with the experimental methods and concepts used in scientific psychology4. His descriptive psychopathology owed much to Wundt's strategy of decomposing complex psychological states into components that are more measurable.
Spitzer majored in psychology at Cornell University. In her biography of him, H. Decker5 reports that Spitzer was trained as a psychoanalyst, but his interests lay in developing structured interviews and rating scales. He began his academic career at the New York State Department of Mental Hygiene in the Biometrics Research Unit, under the psychologist J. Zubin. The unit's purpose was to advance the quantitative study of psychopathology6. The psychological nature of Spitzer's early work is further documented by his collaborations with J. Endicott – a psychologist who had training in psychometrics.
In current terms, Kraepelin and Spitzer each had an interdisciplinary focus. With respect to classification, it has not been such a bad thing for psychiatry to occasionally take note of what the scientific psychologists are doing and rethink current practices – and it does not require a coup.
Turning to the revolution, many psychiatrists, including Spitzer7, would assert that they are aware that psychiatric distress occurs with degrees of severity and that the distinction between normal and abnormal can be fuzzy. Indeed, one could argue that a manifest dimensionality is fundamental to descriptive psychopathology. Understanding it is a prerequisite for the competent use of a categorical classification system. If so, rather than a revolution, the HiTOP model is better seen as an attempt to translate common background knowledge of psychopathology into something more precise and substantive. One disadvantage of revolutionary talk is that it emphasizes the discontinuity between past and present, often drawing attention away from the many continuities8.
Illuminated by the light of dimensionality, our understanding of psychopathology can be expanded in useful and interesting ways. Krueger et al's paper emphasizes an expansion in the scope of research questions asked. Here I would like to discuss another area of expansion. In doing so I will explain what is meant by my title “Quantitative classification as (re‐)descriptive psychopathology”.
I begin by giving an example of descriptive psychopathology: the depiction of panic disorder. After imipramine was introduced in the late 1950s, working at Hillside Hospital on Long Island, D. Klein and M. Fink began prescribing the drug to patients to learn about its mode of action9. In a historical retrospective based on interviews with Klein, F. Callard10 recounts Klein and Fink's treatment of the man who would become the ur‐patient for panic disorder.
The referring therapist believed that this patient had schizophrenia, but Klein disagreed, describing him as anxious, dependent and demanding. After four weeks of treatment with imipramine, neither the patient, his resident therapist, nor the supervising psychiatrist believed that the medication had made any difference. The ward staff did not concur, but they were not sure why. Eventually one nurse noticed that the patient no longer ran to the nurse's station several times a day asking for help because he feared he was dying.
For much of the 20th century, the symptoms of panic were a commonly manifested feature in the population of psychiatric phenotypes, but they were seen as parts of a coherent anxiety neurosis. Klein and Fink re‐described these symptoms by putting a boundary around them, thus separating what they called episodic anxiety from anticipatory anxiety. With this re‐description, even though panic had long been a background feature of the psychiatric landscape, it came into the foreground.
Descriptive psychopathology has been derided as a shallow emphasis on surface features. A successful re‐description, however, is also a conceptual achievement of a synthetic nature in Kant's sense – it guides the way to the acquisition of information that is not contained in the description itself. For example, once Klein and Fink saw panic as distinct from worry and avoidance, they learned that the primary problem in agoraphobia is not fear of open spaces, but fear of having another panic attack. Their discovery that the same patients also avoided crowded theaters would have been a puzzling feature of agoraphobia, but not of panic disorder.
The Research Domain Criteria (RDoC) initiative, with its focus on causality, might represent the abandonment of descriptive psychopathology, but it is equally consistent with RDoC's anti‐reductionist aspiration that mechanisms will be maps for locating new descriptions in the psychiatric landscape. The same is true for HiTOP. Proposing a meta‐structure for how things fit together affords some options for recognizing new patterns. HiTOP has an immediate advantage over RDoC because it does not have to translate biological findings into psychological descriptions; it is already psychological.
Using a taxonomy, however, is only a part of understanding psychopathology, including descriptive psychopathology. It is unrealistic, therefore, to expect that statistical correlations can do all the descriptive work. With respect to panic disorder, Klein claimed that the ward nurse who reported that the ur‐patient no longer ran to the nurse's station was a good observer. This was their first clue to describing what they called a psychiatric reaction pattern. It was followed by prolonged observations of what the patient did and said, how he reacted to others, and how others reacted to him.
Hopefully, good observers will notice some of the clues that a comprehensive dimensional hierarchy presents, recognize patterns, and subject them to validation studies. Concepts like borderline and narcissistic personality disorder are so entrenched that they assert themselves when certain features are present. HiTOP offers a way to take a second look. Ideally, clinicians and scientists could learn to see anew something that has been there before them all along – and let it guide them to other things that they did not recognize before.
The author would like to thank S. Lilienfeld for thoughtful comments on an earlier draft of this commentary.
References
- 1. Krueger RF, Kotov R, Watson D et al. World Psychiatry 2018;17:282‐93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Freud S. A general introduction to psychoanalysis. New York: Liveright, 1920. [Google Scholar]
- 3. Wender PH, Klein DF. Mind, mood, and medicine. New York: Farrar Straus Giroux, 1981. [Google Scholar]
- 4. Engstrom EJ, Kendler KS. Am J Psychiatry 2015;172:1190‐6. [DOI] [PubMed] [Google Scholar]
- 5. Decker HS. The making of DSM‐III: a diagnostic manual's conquest of American psychiatry. New York: Oxford University Press, 2013. [Google Scholar]
- 6. Zubin J, Salzinger K, Fleiss JL et al. Annu Rev Psychol 1975;26:621‐71. [DOI] [PubMed] [Google Scholar]
- 7. Spitzer RL. Schizophr Bull 1982;8:592. [Google Scholar]
- 8. Zachar P. In: Kendler KS, Parnas J. (eds). Philosophical issues in psychiatry II: Nosology – definition of illness, history, validity, and prospects. New York: Oxford University Press, 2012:21‐34. [Google Scholar]
- 9. Klein DF, Fink M. Am J Psychiatry 1962;119:432‐8. [DOI] [PubMed] [Google Scholar]
- 10. Callard F. Osiris 2016;31:203‐26. [DOI] [PMC free article] [PubMed] [Google Scholar]