Lie Detection: The Makings of a Pseudo-Psychology
Delving into the relationship between testing and the preservation of human life.
Although lie-detector tests cannot be used as legal evidence in the case of a suspect’s guilt or innocence, polygraphs may be described as “an important tool in the extraction of false confessions”(Yant, 2012, para. 2). To illustrate, if a suspect fails a lie-detector test, police may utilize the results to obtain a confession; moreover, police may tell a suspect that they failed the exam (even if they did not) in an effort to obtain an admission of guilt (Yant, 2012).
For example, in 2011, the West Memphis Three (Damien Echols, Jason Baldwin and Jesse Miskelley) were released after 18 years in prison after being wrongfully committed of murdering three young boys in a small Arkansas town (Steel, 1998). When the boys were initially convicted, Miskelley took a polygraph test in which the examiner denoted that he was showing “signs of deception”, this information was then used to extract a false confession from Miskelley (Steel, 1998). It was then later reported that the “interrogating officers had used intimidating methods during the interrogation”(Steel, 1998, Confession, para. 9). In this case, the detectives used the polygraph information (whether true or false) to withdraw a confession.
On the flip side of the coin, the American Polygraph Association conducted a meta-analysis in which they reported that the polygraph test has a high accuracy rate at 87% with an inconclusive rate of 13% (American Polygraph Association, 2010-2011).
So, what are we supposed to believe? Do lie-detector tests have relevancy in the field of psychology and criminality? To answer these questions, it may be important to revisit the basic structure of a psychological test?
What is a psychological test?
As per Reynolds and Livingston (2012), a psychological test is “a procedure in which a sample of an individual’s behavior is obtained, evaluated and scored using standardized procedures” (9). This definition seems to correspond to a polygraph in that, a lie-detector or polygraph test records and measured physiological measures e.g., blood pressure, pulse etc. In other words, a lie detector test is measuring a sample of an individual’s behavior.
However, a lie-detector test does not satisfy the definition of a psychological test when it comes to the evaluation and scoring. As per Reynolds & Livingston (2012), a person must be qualified by education and training to administer psychological assessments. Namely, because the polygraph test is one dealing in psychometrics and psychophysiology, one might then suppose that polygraphers are trained psychologists. However, this is not the case, polygraphers are trained outside of psychology and in fact, most polygraphers have been educated by people with a background in law enforcement.
Moreover, a lie-detector test may take a sample of an individual’s physiological response but does it measure what it claims to measure? (i.e. to distinguish between a participant’s deceptive and non-deceptive responses). Where the polygraph really fails is in terms of test-criterion evidence for validity; in other words, how can we know that the test predicts performance on the criterion it is supposed to be measuring— truthfulness/deception. To collect test-criterion evidence for validity, Reynolds and Livingston (2012) suggest using predictive/concurrent studies in which, the test is administered and then the criterion is measured. The question then becomes, is the test able to predict the criterion? For example,
Scholastic Assessment Test (assessment) — — — — — — — — — > GPA scores
This could then be translated with the lie-detection test to:
Polygraph (assessment) — — — — — — — — — — — > ?
The question mark here represents the “best existing measure of the construct of interest” (Reynolds and Livingston, 2012, p. 166). But what is the best measure of “truth”. It there even a valid existing measure of the construct especially since “truth” is dynamic and not static i.e., it depends on point of view, past experiences, etc.
In sum, lie-detection tests raise a lot of questions regarding validity to real-world problems. Is it possible that when dealing with human lives even the suggestion of an error level is not acceptable? Perhaps in the future, lie-detection will cease to be a simple “arousal measurement” and may progress to measuring brain responses with fMRI. The question remains however, do we want individuals’ guessing at your thoughts, intentions? — Things that remain part of our private self.
Penny for your thought(s).
American Polygraph Association. (2010-2011). Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques: Report Prepared For The American Polygraph Association Board of Directors: Nate Gordon, President. Retrieved from http://www.polygraph.org/section/validated-polygraph-techniques/executive-summary-meta-analytic-survey-criterion-accuracy-val
Reynolds, Cecil. R. & Livingston, Ronald. B. (2012). Mastering Modern Psychological Testing: Theory & Methods. New Jersey: Pearson Education Inc.
Steel, Fiona (1998). A most heinous crime. Turner Entertainment Digital Network, Inc. Retrieved from http://www.crimelibrary.com/notorious_murders/famous/memphis/index_1.html
Yant, Martin. (2012, February 23). Re: The Polygraph and false confessions [Web log Message]. Retrieved from http://wrongfulconvictionsblog.org/2012/02/23/the-polygraph-and-false-confessions/
