A Review of Systems for Psychology and Psychiatry: Adaptive Systems, Personality Psychopathology Five (PSY–5), and the DSM–5

Abstract

We outline a crisis in clinical description, in which atheoretical categorical descriptors, as in the Diagnostic and Statistical Manual of Mental Disorders (DSM), has turned focus away from the obvious: evolved major adaptive systems. Adaptive systems, at the core of a medical review of systems (ROS), allow models of pathology to be layered over an understanding of systems as they normally function. We argue that clinical psychology and psychiatry would develop more programmatically by incorporating 5 systems evolved for adaptation to the external environment: reality modeling for action, short-term danger detection, long-term cost–benefit projection, resource acquisition, and agenda protection. These systems, although not exhaustive, coincide with great historical issues in psychology, psychopathology, and individual differences. Readers of this journal should be interested in this approach because personality is seen as a relatively stable property of these systems. Thus, an essential starting point in ROS-based clinical description involves personality assessment. But this approach also places demands on scientist-practitioners to integrate across sciences. An ROS promotes theories that are (a) compositional, answering the question: What elements comprise the system?; (b) dynamic, answering: How do the elements and other systems interact?; and (c) developmental: How do systems change over time? The proposed ROS corresponds well with the National Institute of Mental Health's recent research domain criteria (RDoC) approach. We urge that in the RDoC approach, measurement variables should be treated as falsifiable and theory-laden markers, not unfalsifiable criteria. We argue that our proposed ROS promotes integration across sciences, rather than fostering the isolation of sciences allowed by atheoretical observation terms, as in the DSM.

Acknowledgments

The authors deeply appreciate the work of Claire Harkness in providing the figures for this article, as well as astute comments on an earlier draft. Kyle Simmons of the Laureate Institute for Brain Research provided invaluable comments on an earlier draft of the article. Mackenzie Ayer, Jack Burgos, Jordan Heroux, Kara Kerr, Namic Kirlic, Kelsey Parker, Samantha Overstreet, Ashley Miller, Katherine Miller, and Catherine Morison, students in Allan R. Harkness's graduate course in emotion, also provided helpful comments on an earlier draft.

Notes

Because we focus on a few systems for which there is general agreement about function, we clarify two points. First, noting that a few systems enjoy general agreement over their function does not mean that we subscribe to the Wakefield (1992) “harmful dysfunction” formulation, which justified the mental disorders as damaging breakdowns in evolutionarily selected systems. Wakefield did not anchor his conception of psychopathology in a few psychobiological systems that enjoy widespread agreement on function. Instead, he began from the perspective of disorder, signs, and symptoms, where there might be far less agreement over function, compared to the major adaptive systems we discuss. In addition, whereas Wakefield's formulation addresses the question of which conditions are and are not mental disorders (a question that we suspect is not answerable scientifically), our analysis examines the question of how best to build theory. Second, although major adaptive systems can serve average functions relative to evolutionary history, new environments or demands can always reveal new, previously unconsidered functions. For example, Harvey considered the function of the heart to be to move blood through the circulatory system. When mechanical hearts were tried, it became apparent that a heart had to move blood without damaging red blood cells—the first mechanical hearts caused strokes. Although there might be broad consensus on function, function can never be a “fully settled question” because revealing new environments remain an open set.