I shouldn't have let Mr. Warner's conclusions go without comment:

Quote:


I agree with Mr. Warner here. I don't think anyone is arguing that that  polygraphy can't help investigators obtain relevant facts. To the extent that it leads to independently corroborable admissions/confessions that might not otherwise have been obtained, it certainly may.

However, because CQT polygraphy is without validity, to the extent that polygraph chart readings are relied on, it may also pose an impediment to law enforcement investigators obtaining the complete facts and may help the guilty to evade justice.

Quote:


True, but it's in the same league as other interrogational ruses such as falsely telling a suspect that he was caught on a hidden surveillance camera or that one of his co-suspects has fingered him.

Quote:


Al-Qaeda has long been aware that the lie detector is a sham. Nowadays, anyone with Internet access can discover this with a few minutes of research. As knowledge of the truth about lie detectors rapidly spreads, the utility of the polygraph for obtaining admissions can only wane.

While such independently verifiable admissions/confessions as may be obtained should be welcomed, because it is well documented that CQT polygraphy is without scientific basis, it is a dereliction of duty for any investigator to place any reliance on polygraph chart readings.

Quote:


Mr. Warner does not establish in his article that polygraphy helps solve crimes or prevent future tragedies "more often than not." As Drew Richardson pointed out, Mr. Warner fails to consider the negative utility associated with polygraphy. It would appear that the FBI Polygraph Unit's archival research study to which Mr. Warner referred also failed to consider this negative utility. Why might that be?

While reading SA Warner’s article, I noted that the inaccuracy and the false-positive rate of the polygraph were not discussed.  These are the core issues in my opinion.  The fact that so many law enforcement and governmental agencies base vital decisions on an inaccurate “test” which misidentifies as many truly deceptive people as it labels honest people as “deceptive” is the heart of the entire polygraph problem.  To simply gloss over that matter and switch the argument to whether the polygraph has any use at all in an interrogation setting with an unwitting examinee is specious reasoning.  It is analogous to arguing that an experimental medication which has been shown to worsen patient’s symptoms as often as it eases them is still useful because some patients improve when given the medication simply because they believe it will work (a.k.a. the placebo effect.)

I fully agree that the polygraph can be useful in an interrogation setting.  If the subject’s knowledge of the polygraph comes solely from movies and television (like most people) then he or she will most likely believe that any lie they tell will be detected.  In cases like that, the use or threat of a polygraph “test” may elicit a confession or admission that otherwise might have been withheld.

Having said that, I would have to qualify my opinion by saying that if I was interviewing a subject who grew up believing that crystal balls, tarot cards, or palm readings were effective at detecting falsehoods, each one of those would be just as effective as a polygraph in eliciting confessions or admissions from that subject.  Even SA Warner writes in his article: “any technique that examinees believe to be a valid test for deception likely can produce deterrence and admissions.”  However, SA Warner contradicts that statement when he writes: “Investigators do not want to waste their time with a lie detection technique that yields little more than speculative results.”

I must confess that I’m confused.  I thought the focus of the article was that, without delving into the accuracy or false-positive rate of the polygraph, it can still be useful as an interrogation tool because some people believe it will detect lies.  But the latter quote from the article seems to imply that the polygraph is superior as an interrogation tool because it is a lie detection method that yields better than speculative results.  Which is it?

With regards to the “archival research study” mentioned in the article, I must be more cynical than the author.  I would tend to give little if any credibility to a pseudo-scientific study conducted by FBI polygraph examiners on FBI polygraph examiners.  Perhaps we could also ask Big Tobacco to conduct further studies to see if nicotine is addictive?  Or we could ask McDonald’s to do some research to see if a high-fat, high-sodium, high-sugar meal is good for kids?

In the close of the article, SA Warner poses a question that is designed to make the reader stop and think.  The question is: if your child was kidnapped and the police had a suspect, would you want the suspect to be given a polygraph exam?  The dilemma inherent in that question, in my opinion, is not what would happen if the suspect was questioned and confessed or admitted to some involvement.  The dilemma lies in what would happen if he didn’t confess and in fact came back with a “no deception indicated” chart.  The police would most likely write him off as a viable suspect!  If he was a viable suspect to begin with he should never be cleared by a polygraph, crystal ball, tarot cards, or any other interrogation intimidator.  It simply should not happen.  But it does, in part because the polygraph has a false aura of credibility.  The same incorrect knowledge gleaned from movies and television that make the polygraph an effective interrogation tool on some people also make it a danger to use because of its inaccuracy.  In the example above, if the suspect passed the polygraph exam there is a real possibility that, in the minds of at least some of the investigators working the case, the suspect would no longer be investigated with the same zeal, because he’d already been “proven” innocent.  Witnesses might even change their story, convinced they’ve made a mistake because the person they thought they saw “passed” his polygraph.  The only way to counter that possibility would be to make widely known the details of the polygraph examination and why it works, in order to educate law enforcement officers and the rest of the public that it is ONLY valid as an intimidator during interrogations, and has no actual use as a “lie detector.”  If that ever happened, it would also educate the people who might someday have to take a polygraph exam, which would effectively nullify the value of the polygraph as an intimidator.

Quote:


Finally, you quote this stat
Quote:


Anyone who peruses this board will find time after time that especially in the case of multi issue screening exams, many of the people who have educated themselves about polygraph assign it about chance accuracy in this format. chance accuracy is another way of saying a coin flip or 50%. YOUR STATS SAY THAT OUT OF 2600 CASES, 1300 CONFESS TO SOMETHING. the irony of your proof is amazing. yes, im sure that a portion of the 1300 who didnt admit to anything are lying, but what %? the point is you have no idea and if you thing that wrecking the aspirations of every 3 or 4 out of 10 applicants is aceptable, then to hell with the FBI.

i think this sums it up
Quote:


at least you admit its a prop

Those of us whose FBI careers have been unjustly destroyed before they started by this machine would like you to answer this question, Mr. Warner. What do you estimate your false positive rate is for Multi-issue CQT format screening exams? My FBI polygraphers all told me 0% when I asked them. I wonder if that answer would stand up to their own test?

also you mention this
Quote:


anytime the FBI would actually like to INVESTIGATE your allegations of drug use or sale against me, I would welcome it. If you find 1 shred of evidence that I ever took any drug or sold any drug, I'll buy you a brand new GSR monitor. But you dont investigate people do you Mr. Warner?, you just trash their dreams and goals if god forbid their finger sweats when an FBI agent in a blank room asks them if they ever took any drugs.

Mr. Warner, you write, among other things:

Quote:


The only way to objectively prove that such rules for admissibility have been met is to record the polygraph interrogation in its entirety. Why then does the FBI Polygraph Unit stubbornly refuse to audio- or videotape polygraph interrogations? Why are you (collectively speaking) afraid of creating an objective record that would make it clear to judge and jury whether or not FBI polygraphers used force, threats, or promises to obtain confessions?

Some recent high profile cases suggest that FBI polygraphers do in fact sometimes use such coercive techniques. Take, for example, the FBI polygraph interrogations of Dr. Thomas C. Butler (sleep deprivation, false promises alleged) and Abdallah Higazy (implied threats alleged).

You refer to a study recently carried out by the FBI Polygraph Unit:

Quote:


Would you or one of your colleagues be so kind as to send a copy of this study to AntiPolygraph.org? We'll be happy to make it available on-line in its entirety.

You pose the following question to critics of the polygraph:

Quote:


The parent of a missing child often becomes the prime suspect in the child's disappearance. If I were the parent of a missing child and asked to submit to a polygraph "test" to prove my innocence, knowing "the lie behind the lie detector," I would indeed reply that the polygraph technique is unreliable and demand that you find my child through real investigation rather than voodoo science.

That said, I would not object to the polygraph being used with a suspect as an interrogational prop. But I would strongly object to the polygraph interrogation not being recorded.

Moreover, in such a situation, if the polygraph were to be used at all, I would urge that a Concealed Information or Guilty Knowledge Test be employed instead of the invalid  "Control" Question "Test" preferred by the polygraph community.

Finally, if CQT polygraphy is to be used for its utility in getting admissions/confessions despite its lack of validity, then why not switch to voice stress analyzers? They're cheaper than polygraph instruments and require less operator training.

An article entitled "Polygraph Testing A Utilitarian Tool" authored by FBI Special Agent polygrapher William Warner and published in the April 2005 edition of the FBI Law Enforcement Bulletin (http://www.fbi.gov/publications/leb/2005/apr2005/april2005leb.htm#page10) has been brought to my attention.  The author appears to realize that the debate regarding validity and reliability of polygraph testing (presumably CQT testing in its various applications) is a losing battle if not a closed subject for the majority of scientists in the relevant community(ies) who have examined it.  The thrust of Mr. Warner's article appears to be that polygraphy should not be dismissed simply because it is invalid and unreliable, but through the various examples he offers, he believes it (polygraphy) merits consideration because of its utility.  Although I would suggest through an equally impressive listing of polygraph mistakes, that Mr. Warner consider the negative utility or lack of utility of the polygraph, I believe the most salient point that Mr. Warner fails to consider in his article is the relationship between validity and utility.  The admissions and confessions that Mr. Warner considers to be evidence of utility (He further fails to consider or at least discuss that same might be obtained in the absence of polygraphy) are (if related to polygraphy at all) dependent on an examinee believing that polygraphy has some validity as a diagnostic test, i.e., that it works.  Now that the cat is out of the bag so to speak and it is largely clear to the scientific community that no such validity exists, with the advent of various forms of widespread and rapid communication fostered by various vehicles such as this site (Mr. Warner should be congratulated on citing it in his bibliography), the general public is not far behind.  The ignorance that has been the foundation and the bulwark of any polygraph utility is crumbling and eventually will collapse altogether.  If Mr. Warner is not now aware of the relationship between validity and utility, he most certainly will be then.  Perhaps Mr. Warner might care to join us and continue a discussion of his article and the topics of polygraph validity and utility.