"The polygraph is merely a psychological rubber hose, not a way of detecting deception."

Dr. Phil,

You did a disservice to society by claiming that administering a polygraph could reveal the certain truth of whether Mikai is a sexual predator.

During your two shows, you gave the impression that the polygraph machine in the hands of a registered polygraphist is a scientifically based test which was not only accurate, but 100% percent accurate, or, as you put it, was "foolproof" with regard to whether Mikai was a sexual predator.

But, later your website posted a note from you that shows that the polygrapist, Mr Skeeters, uses the polygraph merely as an interrogation tool, a way to induce confession. This contradicts his own account of the polygraph during the show, as a 100% accurate way of determining the truth rather than being, as he puts it, a "tool" for "an interrogator".

The public needs to understand the real nature of the polygraph and how it is used. Those who promote the impression that it can directly reveal whether a person is lying or truthful are actually participating in a hoodwinking of the public and of those who agree to submit to a polygraphic procedure.

When polygraphers pretend to "explain" the results of a polygraph administration, there is one thing they rarely reveal. This is that the judgment about the polygraphed individual as to whether he or she is lying or telling the truth depends on the degree to which that individual is made nervous by two sorts of questions. One class of questions concerns deeds of which the suspect is accused, while another class of questions concern deeds that are unrelated to these accusations, but denial of which is either a partial or total lie. These two sorts of questions are called "relevant" and "control" (or "comparison") by the polygraph industry. The rationale of the so-called "test" rests on assessing the relative degrees of "nervousness" elicited by the two sorts of questions. This "nervousness" is indexed by certain autonomic or involuntary physiological responses such as blood pressure change and drop in skin resistance or galvanic skin response (GSR).

The fate of the suspect depends on this relevant/control comparison. If the autonomic responses, for example the GSR, elicited by the relevant questions and recorded by the polygraph are clearly greater than the GSR elicited by the "control" questions, the suspect is classified as "deceptive" or lying about the accusation. If it is the "control" questions that elicit the clearly greater GSR, a sign of greater nervousness, then the suspect is classified as "truthful" or innocent. If the GSRs elicited by the two sorts of questions don't differ much, the so-called "test" is classified as "inconclusive".

The point here is that there is nothing in the polygraph procedure that can discriminate between nervousness caused by anxiety in the innocent in response to a question, and nervousness caused by fear of getting caught in a lie in the guilty. So, contrary to the strong impression you and your registered polygraphist, Mr. Skeeters, gave to the audience throughout the two shows, the polygraph is not only not a foolproof test of truth (and, of course, no psychological test is 100% accurate); it may be completely misleading in the case of some individuals.

The post-polygraph section of the Mikai program illustrates that the polygraph is designed primarily not to detect deception, but to induce confession, which is precisely the admission Mr. Skeeters made in the note that you posted after the show that quoted him directly. The physiological recording that provides pseudo-scientific status serves only as an interrogatory prop. What we saw was a classic instance of the "post-test interview" stage during which time the polygrapher attempts to induce a confession (which itself, of course, can be used as evidence in any court of law).

Some viewers may say, what does that matter, under pressure Mikai confessed and the important thing was to get at the truth in this situation. But it is just as important that innocent people are not pressed to confess to acts they did not do on the basis of a pseudo-scientific procedure that is not more valid than the reading of entrails practiced by priests, and believed in by many ancient Romans.

By posting the note that hints at the limitations and practical use of the polygraph, you have perhaps somewhat undone the erroneous impression of the show itself. But there is still enough ambiguity in the statement posted in conjunction with the show, to perpetuate the false and often very damaging impression that the polygraph can reveal whether a person is lying or telling the truth, rather than merely reacting more or less nervously or not to different sorts of questions.

The conclusions regarding the polygraph that I have written so far are based on arguments that do not require knowledge of specific scientific areas that relate to this procedure. It is possible, however, to provide a more specialized set of arguments for which my expertise and specific research interests are relevant. I am a professor of psychology at the University of Toronto, and my main specialty is psychophysiology, of which the polygraph is a purported application. My conclusion regarding the polygraph, as it is practiced in North America, after some two decades of scientific consideration (my first publication on lie detection was in the 1986 Handbook of Psychophysiology), is as follows: "The so-called 'control' question 'test' polygraph is a technological flight of fancy. It is often used as a psychological rubber hose to induce confessions. It spreads distrust while posing as the path to truth."

This is a very harsh conclusion, but justified in light of the scientific evidence. The details of this position in scientific terms can be found in a 1996 paper published in the International Journal of Psychophysiology. The web reference for this paper is: http://psych.utoronto.ca/~furedy/Papers/North%20American%20Polygraph%20and%20Psychophysiology.rtf