Eliminating bias in forensic psychology

Vivan Sharma    

When one talks about psychology in the context of aiding justice in the court, several names come to mind. Be it either through the highly sensationalized cases of infamous murderers and sex-offenders like Ted Bundy and Jeffrey Dahmer, or fictional felons in hit TV shows like ‘The Silence of the Lambs’ or ‘Criminal Minds’, forensic psychology has definitely made its mark.

Defined as a “specialty in professional psychology characterized by activities primarily intended to provide professional psychological expertise within the judicial and legal systems” (APA), forensic psychology aims to disseminate the subtleties of the human mind to provide a clearer picture of the case in question. Not only does it shed light on how criminals think and act, the behaviors of witnesses, victims and first responders also fall within its ambit.

However, while the field itself is very well known, lesser so are the tools it uses for psychological assessments. Neal et al. (2019), conducted a study on 364 such tools used by experienced forensic mental health practitioners across 22 surveys in the US. They found two disturbing results. The first was that tools which are either not generally accepted or do not have favorable reviews regarding their technical validity in peer-review journals like the Mental Measurements Yearbook are widely used. While 90% of the surveyed methods had been empirically tested, only 67% were regarded as ‘generally-accepted’, and only 40% received favorable feedback for the scientific basis they used for judging psychometric properties from fellow experts.

Clearly, this is highly problematic. Having this much variation in methods of assessment leaves too much room for individual interpretation. Moreover, if the foundation of these tools isn’t solid to begin with, how accurate could their diagnoses be? Therefore, in order to maximize precision, there should be a uniform set of tools for all forensic psychologists to adhere to.

One might say that there already do exist some standard procedural tools. Therein lies another latent problem. Lees-Haley et al. (1996), found that up to 32 different tests were used in forensic adult personal injury evaluations, from reports of 100 neuropsychological experts. The most frequently used was the Wechsler Adult Intelligence Scale-Revised (WAIS-R), while the Minnesota Multiphasic Personality Inventory (MMPI) /MMPI-2 came second. Neal and Grisso (2014) also collected data on structured assessment tools used to aid expert judgment. They asked forensic examiners to recall the tools they used in their last two cases (N = 434 experts, 868 cases). Not only did they find an extreme variation in the toolsets used (268), but also that clinical instruments were used very often. MMPI was used most frequently for personality assessment, Psychopathy Checklist-Revised (PCL-R) for violence risk assessments, and so on.

Using clinical psychological tools here may not be optimal. Even though they are psychometrically sound and are backed by research, they are not meant for forensic settings. One cannot use tools indiscriminately; they were made for specific contexts and have to be adapted accordingly. “Forensic psychiatrists applying the results of such measures in forensic settings are making a big assumption: that the scores mean the same thing that they mean in the clinic” (Archer and Wheeler, 2015). In order to use these tools, they must first be validated in the legal context.  

The second part of Neal et al.’s (2019) study found that evidence given by forensic psychologists is rarely challenged. Only 5.1% of the cases in the sample set had had some form of doubt cast on the same. Out of that, only one-third of the challenges succeeded. If psychological evidence is given so much weight, the greater the responsibility of the expert in making sure they do not err. In order to reduce the role of the psychologist in determining the result and to add an element of accountability to the mix, we must develop a set of psychological assessment tools specifically for the context of the courts, ensure its technical validity as well as test it empirically. Only through this process can a uniform procedure emerge, and justice be unencumbered by the bias of the psychologist. 

References:

1.     Neal, Tess, Slobogin, Christopher, Saks, Michael, Faigman, David, Geisinger, Kurt. (2019). Psychological Assessments in Legal Contexts: Are Courts Keeping “Junk Science” Out of the Courtroom? Psychological Science in the Public Interest. https://www.researchgate.net/publication/339294508_Psychological_Assessments_in_Legal_Contexts_Are_Courts_Keeping_Junk_Science_Out_of_the_Courtroom

2.     Lees-Haley, Paul R., Smith, Harold H., Williams, Christopher W., Dunn, John T., (1996) Forensic neuropsychological test usage: An empirical survey, Archives of Clinical Neuropsychology, Volume 11, Issue 1, Pages 45-51, ISSN 0887-6177, https://doi.org/10.1016/0887-6177(95)00011-9.

3.     Neal, M.S. Tess & Grisso, Thomas (2014), “Assessment Practices and Expert Judgment Methods in Forensic Psychology and Psychiatry: An International Snapshot”, Sage Journals, https://doi.org/10.1177/0093854814548449

4.    Archer, Robert R. & Wheeler, Elizabeth M.A. (2015), “Forensic Uses of Clinical Assessment Instruments”, The Journal of the American Academy of Psychiatry and the Law, Volume 43, Number 2. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.1017.4704&rep=rep1&type=pdf