In this blog entry I’ll provide a brief description of recent studies on the measurement of psychopathy, which may help us conceptualize this disorder accurately, but please bear with some initial paragraphs to allow me to set things up.
In the history of psychiatry and psychology, there have been many great thinkers who had extensive clinical experience with psychopathic patients, and from their case studies we have the benefit of their published work that provides us with some information on the nature of psychopathic personality (e.g., Arieti, 1963; Cleckley, 1941). However, case studies, by definition, provide us with only a very weak scientific method, and one that is subject to huge biases (Hare & Neumann, 2008; Neumann, Hare, & Johansson, 2013). Moreover, the ‘measurement’ of psychopathy was all done in the heads of these thinkers, given they never developed scientifically rigorous measures to assess the precision of their conceptualization of psychopathy.
Measurement is not by itself sexy, unless of course we’re talking speed from 0 - 60 MPH in a classy Corvette, the volume of megapixels in the hottest smart phone screen, or perhaps the chest and/or buttocks specs of the latest trending public figure. Similarly, the intelligence quotient, or IQ, is a fascinating measurement topic for many individuals.
Often missing in ‘sexy’ measurements is understanding of the precision (validity) and consistency (reliability) of such specs. If a Corvette can obtain a time of ~3 seconds from 0 - 60 half of the time, or only under certain conditions, we’d seriously wonder about the reliability and validity of the claim that it is a fast car. The issue of IQ (mis-)measurement represents a serious historical as well as contemporary topic. In the 1950’s a large amount of the Army’s initial IQ assessment data was available for analyses and used to claim that African-Americans had inferior IQ scores compared to Whites. Thankfully, Horace Mann Bond was able to show that educational experience played a huge role in IQ test results, such that Whites from the southern States had lower IQ scores than African-Americans from the northern States where better educational opportunities were available (Schultz & Schultz, 2012).
The measurement of psychopathy, given the interest in this personality disorder itself, is often not treated with the seriousness it should. It is clear that (mis-)measuring someone and labeling them as a psychopath when they are not (false positive) has as serious of consequences as does claiming a person does not ‘measure up’ to that of a psychopath when indeed they are psychopathic (false negative). By analogy, if a measure of blood pressure was reliable or accurate only some of the time and wrong other times we would not use such a measure.
Contemporary psychopathy researchers often claim that their measure is reliable and valid by resorting to traditional classical test theory approaches (CTT) but this older approach has a fundamental flaw built in to it. The CTT approach assumes that a psychopathy measurement scale produces a score for an individual that represents both their ‘true’ score (e.g., level of psychopathy) plus error (e.g., deception, poor motivation). In other words, in a CTT approach to psychological measurement there is always some signal (true score variance) and some noise (error)! Thus, there is only a gross level of precision when traditional CTT approaches are used to measure some psychological attribute, much in the same way we might guess peoples heart health by looking at them in terms of how overweight they appear to be.
Thankfully, beginning in the 1990’s, Robert Hare and his colleagues developed a reliable and valid interview-based measure that has helped to provide a common metric for thousands of scientific papers, and help us better understand psychopathic offenders (Hare, 2003). Needless to say, Hare’s Psychopathy Checklist-Revised (PCL-R) has become the standard measurement for psychopathy. Based on the strength of the PCL-R’s performance, Hare and I along with our colleagues have developed self-report measures of psychopathy that are modeled after the PCL-R (Neumann, Hare, & Pardini, 2014). Research on the PCL-R scales has moved far beyond the traditional CTT approach.
We have employed modern sophisticated statistical techniques, such as structural equation modeling (SEM), which can be used to pull out the ‘signal’ from the ‘noise’ in all psychological measures such as the PCL-R. These modern approaches are able to provide precise error-free parameters, which help discriminate whether someone is psychopathic or not, much like modern blood pressure devices provide critical parameters on how hypertensive a person may be. In addition, techniques like SEM allow us to understand the number of dimensions that make up a phenomenon. Most things in the world, whether they be fast cars, sexy public figures, or psychopathic persons can be understood in terms of the underlying dimensions that make them up (e.g., number of cylinders, weight of car, chest/buttocks dimensions, facial symmetry, etc.).
My colleagues and I have used SEM to describe psychopathic traits in thousands of offenders, psychiatric patients, individuals from the general community, and even those in corporate settings. Moreover, we’ve used this same approach to study over 30,000 individuals from all parts of the globe (Neumann et al., 2013, 2014). We find that, irrespective of sample type or the specific PCL-based measure used, psychopathy can be understood in terms of four inter-related dimensions that together make up the syndrome of psychopathic personality. The dimensions reflect disturbances in Interpersonal (e.g., glib, grandiose, conning, pathological lying) and emotional or Affective functioning (e.g., callousness, shallow affect, lacking remorse, failure to accept responsibility), as well as behavioral Lifestyle (e.g., parasitic, lacking goals, irresponsible, impulsive, stimulation seeking) and overt Antisocial tendencies (e.g., early behavior problems, poor behavioral controls, engaging in a diversity of immoral or unlawful acts). Thus, psychopathic personality is multi-dimensional and within each dimension are trait parameters that discriminate psychopathic from non-psychopathic individuals. Only individuals who manifest elevations on all four dimensions would be considered diagnostically psychopathic.
No other research to-date has been able to provide such an exhaustive set of studies using sophisticated statistical approaches to uncover how psychopathic personality manifests across a wide range of individuals. This is not to say that other measures of psychopathy are of little value, only that none have matched up to the rigor that’s been used with the PCL-R and related scales. Most other studies on various psychopathy measures have simply relied on error-prone CTT and thus we have no idea of their statistical parameters or their dimensionality. By analogy, it is as if one would try to understand the rocks from the moon through visual inspection versus conducting sophisticated chemical analysis on them. Given the serious nature of psychopathy, it seems worthwhile to conduct the sophisticated analyses on all psychopathy measures, as has been done with the PCL-based scales.
Having a scientifically sound measure represents only a beginning. Once we have such a measure in place, then a host of studies can be conducted with greater confidence in the results; such as, are psychopathic traits like hypertension that varies from low to high, what are the genetic and environmental contributions to these traits, how does psychopathy develop, etc?
Cheers,
c
References
Arieti, S. (1963). Psychopathic personality: Some views on its psychopathology and psychodynamics. Comprehensive Psychiatry, 4, 301-312. doi: 10.1016/S0010-440X(63)80056-5.
Cleckley, H. (1941). The mask of sanity. St. Louis, MO: Mosby.
Hare, R. D. (2003). The Hare Psychopathy Checklist-Revised, 2nd edition. Toronto, ON: Multi-Health Systems.
Hare, R. D., & Neumann, C. S. (2008). Psychopathy as a clinical and empirical construct. Annual Review of Clinical Psychology, 4, 217-46.
Neumann, C. S., Hare, R. D., & Pardini, D. A., (2014). Antisociality and the Construct of Psychopathy: Data from Across the Globe. Journal of Personality, DOI: 10.1111/jopy.12127.
Neumann, C. S., Johansson, P., & Hare, R. D., (2013). The Psychopathy Checklist-Revised (PCL-R), Low Anxiety, and Fearlessness: A Structural Equation Modeling Analysis. Personality Disorders: Theory, Research, and Treatment, 4, 129-37. doi: 10.1037/a0027886.
Neumann, C. S., Schmitt, D. S., Carter, R., Embley, I., & Hare, R. D. (2012). Psychopathic traits in females and males across the globe. Behavioral Sciences and the Law, 30, 557–574. DOI: 10.1002/bsl.2038.
Schultz, DP & Schultz, SE. (2012 10th ed). A History of Modern Psychology. Cengage Learning, ISBN: 9781111344979.