Is criminal profiling a pseudoscience?

By , December 18, 2009 2:47 pm

I’ve blogged about pseudoscience before — techniques or methodologies that sound like they have firm scientific support but that do not. Although the general public knows that polygraph tests are not admissible in court as evidence of deception or lying, there are many other investigative techniques that are described in the media and employed by investigators that also do not meet rigorous scientific standards.

Criminal profiling may be another one of those techniques. I just came across an article published last year in Criminal Justice and Behavior (Vol. 35 No. 10, October 2008, 1257-1276) that addresses this issue. The article is “The Criminal Profiling Illusion: What’s Behind the Smoke and Mirrors?” and was authored by Brent Snook, Richard M. Cullen, Craig Bennell, Paul J. Taylor, and Paul Gendreau. Here are some snippets from the article, which the author was kind enough to send me:

Criminal profiling …. is being utilized by police agencies around the world despite no compelling scientific evidence that it is reliable, valid, or useful (Snook, Eastwood, Gendreau, Goggin, & Cullen, 2007).

[…]

Current profiling practices typically involve some type of classification system, or typology, which is used to categorize both crime scene behaviors and offender background characteristics. By far the most routinely used typology is the FBI’s organized–disorganized dichotomy, which was developed from interviews with offenders (Ressler, Burgess, Douglas, Hartman, & D’Agostino, 1986). The assumptions underlying this typology are that (a) offenses can be categorized as organized (e.g., well planned) or disorganized (e.g., unplanned) based on the behaviors present at a crime scene, (b) offenders can be categorized as organized (e.g., high functioning) or disorganized (e.g., low functioning) based on the background characteristics of the offender, and (c) there is a correspondence between offenses and offenders (i.e., organized offenders commit organized crimes and disorganized offenders commit disorganized crimes)……. Preliminary evaluations of this typology suggest that it does not match the variations in offender behavior. For example, Canter, Alison, Alison, and Wentink (2004) examined whether 100 murders committed by serial killers in the United States could be categorized as organized or disorganized. Using a multidimensional scaling procedure, they analyzed 39 crime scene behaviors that had been categorized as organized or disorganized in the FBI’s Crime Classification Manual (Douglas, Burgess, Burgess, & Ressler, 1992). In contrast to what would be predicted from the typology, the analysis did not reveal any distinct subsets of organized and disorganized behaviors, thus making it difficult to see how serial homicides may be categorized easily in this way. Similar findings have been reported for other popular classification systems (Canter & Wentink, 2004; Melnyk, Bloomfield, & Bennell, 2007). For example, Melnyk et al.’s (2007) examination of Keppel and Walter’s (1999) serial sexual homicide typology, which was borrowed from research on rape and refined on the basis of Keppel’s experiences as a homicide investigator, failed to find any evidence supporting the hypothesis that offenses or offenders could be categorized as power–assertive, power–reassurance, anger–retaliatory, and anger–excitation. As with the organized–disorganized dichotomy, the credibility of Keppel and Walter’s typology collapses under empirical scrutiny.

[…]

In a similar way to the classic trait theory that was popular in personality psychology up until the late 1960s (Mischel, 1968), the majority of CP approaches (e.g., the organized–disorganized dichotomy) assume that criminal behavior is determined by underlying dispositions within offenders that make them behave in a particular way (e.g., Åsgard, 1998; Badcock, 1997; Boon, 1997; Canter, 1995; Douglas et al., 1992). The assumptions that emerge from this theory are fundamental to CP (Alison, Bennell, Mokros, & Ormerod, 2002; Woodhams & Toye, 2007). For example, the theory leads to the assumption that offenders will exhibit similar behaviors across their offenses as traits, versus situational factors, are the determinants of behavior. Perhaps more important for the practice of CP, the theory also suggests that offenders will display similar behaviors in their crimes and in other aspects of their lives (e.g., interpersonal relationships)……. It is our contention that trait-based models of profiling are fundamentally flawed. Criminal profilers do not seem to have recognized that a consensus began to emerge in the psychological literature some 40 years ago that to rely on traits or personality dispositions as the primary explanation for behavior was a serious mistake. Situational factors contribute as much as personality dispositions to the prediction of behavior (Bowers, 1973; Mischel, 1968). This is equally true when predicting criminal behavior (Alison et al., 2002; Andrews & Bonta, 2006; Bennell & Canter, 2002; Gendreau, Smith, & French, 2006; Woodhams & Toye, 2007).

[…]

In sum, there is no compelling scientific evidence to support the positive view of CP that dominates popular opinion. Far from cutting edge science, CP approaches are based on typologies that lack empirical support and are often based on an outdated understanding of human behavior. In addition, professional profilers often produce predictions that are not significantly more accurate than nonprofilers. Given this state of affairs, one might wonder why police officers continue to request the assistance of profilers. Whereas some police officers report using CP because they believe that it works (e.g., Copson, 1995; Jackson et al., 1993; Pinizzotto, 1984), there are likely other officers who use CP but do not believe that it works. We suspect that some of these officers might use CP because they believe (or are instructed) that it is their duty to use all available investigative techniques. Others may believe that they have nothing to lose in seeing what a profiler can offer to an investigation. It is not known, however, whether CP is helpful or harmful to police investigations. As positive beliefs about the validity and reliability of CP are not supported by empirical evidence, the rest of this article is devoted to explaining why people might believe that CP works.

There’s much more to the article, and it makes a fascinating read. Indeed, if I were still in academia, I would assign this article as required reading.

Pseudoscience poses a serious risk of harm to us all. We may waste time and money on treatments that do not work, we may conclude people are lying or guilty of crimes when they are not and send them off to jail for crimes they did not commit, and while we expend precious resources on pseudoscientific techniques, we may miss the opportunity to develop methodologies that might really be of value.

In any event, Snook et al.’s review is yet another example or reminder to us all as to why we should not just blindly accept the opinions or statements of all “experts” we see interviewed in the media, no matter how convinced they may seem or how much they may genuinely believe in the value of their analyses or methods. I would love to see reporters or journalists ask their experts, “Is this technique accepted in the scientific community as being reliable for this purpose?”

5 Responses to “Is criminal profiling a pseudoscience?”

  1. Will I Be Free says:

    I was surprised to find that your site required an email address since you are for privacy. The world we live in … I guess.

    I am convinced that the various pseudo sciences are consistently bad for our social structure and our moral health. I find the article compelling in many respects. This behavior is a sigil of the deterioration of our moral fiber and intellectual integrity in America. It is akin to the reliance on magic and formulaic repetition which comes from belief in invisible deities to explain things that were beyond our control in the dark ages of reason.

    The paradigm of “self help” books popularized by the media and given some credence by subscription of “experts” is in the same category and has about the same level of reliability – pop psychiatry or pop psychology or the recurring theme of “body language” as a way of reading people like a book. All of these activities generally lead our society toward the dissolution of the ability to reason and apply scientific principles in rigorous testing of reality. The more people who are willing to accept such rubbish as established science, the fewer people there are left to take the path of education and enlightenment which produces minds capable of deduction and the search for truth.

  2. dissent says:

    I go back and forth about requiring an email address, even though it does not appear publicly. One thing commenters concerned about their privacy can do (here and in general) is to use a throwaway email account that is only used for commenting on blogs or forums and that is not tied to any real accounts or their ISP’s IP.

    As to your main point: some “self-help” books are useful — the ones that are actually based on research and are presenting material in a way that can reach more people. For example, years ago, Harper and Ellis wrote “A New Guide to Rational Living.” It was a self-help book based on rational-emotive therapy (RET). RET had an empirical basis. Similarly, I’ve seen some self-help books on overcoming Obsessive-Compulsive Disorder that are based on research-validated treatments in Cognitive-Behavioral Therapy/Exposure – Response Prevention. Those types of self-help books can be a boon to those who cannot afford therapy or who live in areas where qualified therapists are not available. But not all self-help books are grounded in research or supported by research, and therein lies a problem.

    I see the “body language” books, which you refer to specifically, as a mixed bag for a number of reasons. There’s no licensing specific for “body language analyst” that I know of, which means that anyone might be able to call themselves a “body language expert.” Then, too, if you look at the books on Amazon.com and check to see which ones refer to actual research, you’ll find that while some make a point of referencing research, others just make assertions without any empirical support other than their own experiences or claims about others’ experiences. While experience is a good teacher, there has been research on nonverbal communication and it’s fair to ask whether what these people are selling is consistent with that research.

    I’ve always been particularly concerned about claims that we can know with certainty and accuracy that any one individual is lying based on body language, since reviews of the research that I’ve read suggest that we cannot know that without any doubt, as the body language may be indicative of stress as opposed to deception. Some of the research on investigative techniques that I read is really quite interesting as it showed those who relied on body language were generally less accurate than those who relied on the verbal content of the suspect’s interview. And a criminal defense attorney pointed out another problem to me: once an investigator gets it in his or her head that the suspect is lying (perhaps because they bought the myth that body language conclusively indicates deception instead of stress), the police may focus exclusively on that individual and look to find other evidence of guilt instead of continuing to look for other possible suspects.

    Joe Navarro does a good job about being clear about the point that body language cannot be used to confidently determine deception or lying and that it indicates stress and not necessarily deception, but other so-called body language experts continue to make statements that I think are unsupported by the research or even contradicted by research.

    Social psychologist Dr. Gary Wood has blogged about body language myths and body language “experts” a number of times on his blog, PsyCentral Blog. You might particularly appreciate these comments that he published in June 2009:

    One of the most important things we ‘know’ about body language is not true! It’s based on a distortion of research by Albert Mehrabian. Body language does not account for 55% of the message in all communication. The 55% figure is only relevant when we are forming an attitude (like or dislike) of someone. The fact that some ‘experts’ incorrectly trumpet this blatant misreading of the research (intentionally or through ignorance) simply distorts our perception of the importance of body language over words, and over context in the case of pap snaps. It really means very little in celebrity snap shots and would have to be so obvious that we wouldn’t need an expert to decode it, such as one person strangling the other. Yes, it offers an example of a particular body language sign but to say it actually applies in a particular case (such as a photo) is at best guess work and most likely flim-flam! It’s there to entertain and titillate not to inform or educate!

    Setting aside the fact that psychologist shouldn’t be speculating about the private lives of celebrities, body language (non-verbal communication) isn’t as exact as the ‘experts’ would have us believe. Context and congruence are all important. One ‘classic’ signal may conflict or be overridden by other signals. A snapshot cannot possibly provide all the information necessary to make an educated guess let alone a definite statement. We need to take a video approach over the snapshot approach. To gain any insight into the state of a relationship the signals we need to consider a broad range of signs and behaviours over a longer period of time, rather than cherry pick based on a snap shot. It’s worth remembering that a turd with a cherry on the top is still crap!

    I’ll be blogging more about this in a separate blog entry.

  3. Will I Be Free says:

    Thanks for the reference to Dr. Woods. I enjoyed reading his blog and will be adding him to my usual haunts.

    I agree that some self-help books serve a useful purpose and are well written and based on quality scientific principles and research. I should have generalized less in my original comment.

    I actually know many books of that ilk which are real gems and which I would have no qualms in giving a recommendation on. One such is “Games people play” by Dr. Eric Berne (even though it is from the 1970s it is still relevant and the research backing it has continued to progress). So, I agree with everything that you stated in your answering comment.

    I find it disturbing that no body regulates the dissemination of inaccurate and misleading information under the guise of science. Real harm is possible when someone believes in the workings of some psuedosciences and uses them to drive decision making in their life or in the lives of others. True science is sometimes abused in a some what similar fashion when people advocate its use in situations where it is not suitable – such as body language and other pop psychology.

    I am for freedom of speech and the free flow of ideas, but there must be accountability when so called experts cause social harm by subscribing to self-help concepts irresponsibly to gain fame or to make a fast buck.

  4. Will I Be Free says:

    Equally disturbing to me is the use of the legal system, or the threat there of, to compel those who would force these charlatans to be drug kicking and screaming into the light of rigorous scientific scrutiny to be silent or remain silent. The legal system should fight back by offering a penalty equal to the judgment sought by these charlatans when it is obvious that they have wielded a suit at law for the express purpose of muzzling the proponents of change or scientific inquiry.

    Therefore, if a suit is brought for – say a million dollars – against someone for speaking out about a charlatan and the charlatan loses the suit then they should be fined the million dollars or it should cause a reverse distribution of that amount to the party having been sued. That would quickly put a stop to nuisance suits with the sole intent of forcing people to remain silent about a particular product, service, or unprofessional behavior and would have the benefit of slowing down the social harm caused by these people or companies. The current legal practice of simply assessing court costs against the losing party in some of these instances is not enough to slow down this practice.

    I am all for some method of fostering true comity where there is a whistle-blower, if you will, and a social ill being targeted.

  5. dissent says:

    In the U.S., some states have SLAPP statutes and judges may impose fines (I think) or penalties against the party who filed the SLAPP suit in addition to awarding court costs to the defendant. I’m not sure how often that has occurred, though, and would have to ask around unless one of my readers knows and can jump in.

    In some sense, your comment is equally or even more appropriate to the blog entry on libel tourism where you have companies or individuals running to another country to file a defamation lawsuit that would never stand a chance here as a way of squelching critics.

Panorama Theme by Themocracy