Developmental Interactive Framework for Psychotic Disorders

Our training in psychiatry during the early 1960s was at a time of intense controversy regarding the origins of schizophrenia. Rather comprehensive explanatory theory for psychopathology was postulated at the level of genetics/biology or psychology or social theory. These perspectives competed for dominance more often than seeking integration although the biopsychosocial medical model, espoused by Engel,1 integrated these perspectives in a general systems framework. At that time the Danish adoption studies confirmed that inherited genes conveyed familial risk for schizophrenia, but were often interpreted as not only establishing schizophrenia as a genetic disease but also as falsifying psychological and social theories of etiology.2 At the time of our early research concerning schizophrenia, there was the view that a careful diagnosis based on stringent criteria would identify persons with a disease entity that sharply reduced the heterogeneity observed in development and course of illness in cohorts based on a broad concept of schizophrenia. Schneider viewed symptoms of first rank as distinguished nuclear schizophrenia and Langfeldt separated true from pseudo schizophrenia with criteria including first rank symptoms. The view that a broad concept of schizophrenia without stringent criteria in the United States led to over-inclusion, and hence heterogeneity, suggested a solution partly based on first rank symptoms. Our understanding of the data including our analyses in the context of the International Pilot Study of Schizophrenia (IPSS)3 led to a very different interpretation. We had found that criteria specified by Schneider and Langfeldt were not associated with course of illness and did not reduce heterogeneity when applied to a broadly defined cohort.4 We also documented that these special symptom criteria were observed in other disorders with psychotic symptoms.5,6 We concluded that schizophrenia was a clinical syndrome, a poor target for discovery, and called attention to separable psychopathological targets within the syndrome as meaningful targets for investigation.7 We also found that components of developmental history predicted future course of the same component independent of positive psychotic symptoms including those of first rank.8 As a footnote on the role of ideology over science, note that of the 9 IPSS centers only the Washington center collected development based prognostic data and follow-up data related to heterogeneity. Also note that DSM-III allowed a single first rank symptom to be sufficient to meet the A criteria (corrected in DSM-5), omitted negative symptoms from the A criteria, and imposed a duration of illness criteria to assure chronicity. In the 1981 book, Schizophrenia,9 we introduced the developmental interactive concept of schizophrenia. Now, with 36 years of research in the field, how so we view this conceptual framework? The essential elements in the developmental interactive model of schizophrenia are: Etiology may be genetic, environmental, or interactive between risk elements. At that time we assumed most cases would be interactive. Now, we would think probably all cases are interactive across time even when the genetic association is very strong as in 22q11 syndrome. The model assumed pathways from gestation to manifest schizophrenia and future course. We did not think this model unique for schizophrenia but would now be explicit and includes other psychotic disorders and also disorders that do not reach a psychotic stage. In 1981, the model could illustrate environmental risk factors at every stage of development and presume interaction with genetic vulnerability. At present, the environmental risk factors could be expanded and the gene/environmental interaction placed more firmly in the epigenetic and neuroplasticity framework. The 1981 model visualized progressive platforms of vulnerability culminating in schizophrenia. We would now attempt to define a vulnerability platform preceding development of psychopathology, a second platform representing the clinical high risk syndrome at an early stage capturing a wide range of emerging pathologies and at a later stage representing the attenuated psychosis syndrome. This concept would be action oriented as proposed in the staging concept by McGorry.10 The 1981 model included factors relating to resiliency, characteristics of competence, and positive social factors, and the hope for primary and secondary prevention of schizophrenia and pathways to a more favorable course and outcome. We would now place substantial emphasis on enhancing resiliency at each stage considering the limited advances in therapeutics and the absence of primary prevention. Since complex interactions among nonlinear variables are involved, innovative methods for defining mechanisms will be required. In addition to current statistical approaches such as machine learning and computational analyses, it is possible that understanding these interactions in the context of the life story may result in narratives that might be especially useful to generate hypotheses regarding causal processes and/or resiliency. What is needed now is the fleshing out of relevant models with an understanding at the causal and mechanism levels that enables detection and intervention in support of primary and secondary prevention. This issue of the Bulletin reviews the current knowledge base,11 provides an outline of how science can advance in human studies12 and illustrates how preclinical models can inform regarding putative mechanisms.13 Commentary on the current role of genetic and environmental factors in neurodevelopmental disorders are provided by Weinberger and Murray.14,15 Barbara Fish was a pioneer in this area and is remembered in a special feature along with a short history of the neurodevelopmental hypothesis of schizophrenia by Morgan et al in this issue.16 What may be missing in current work is sufficient knowledge from a phenomenological perspective that facilitates connecting behavioral constructs to developing psychopathology and the substantial individual variation that may relate to social factors and to a person’s own efforts and competence with attention to the person’s life story. Although these issues present challenges for implementation at the cohort level for many traditional scientific designs they are of major clinical importance for the individual.