A transcript of Dr. Richardon's remarks is now available at:

http://antipolygraph.org/nas/richardson-transcript.shtml 

Audio recordings of the 17 October meeting are now available on the committee's website in RealAudio format at:

http://www4.nas.edu/webcr.nsf/MeetingDisplay5/BCSS-I-00-01-A?OpenDocument 

Those not familiar with Dr. Richardson's previous history of blowing the whistle on polygraph "testing" may be interested in the following documents:

• Dr. Richardson's opening statement at a 29 September 1997 hearing of the U.S. Senate Committee on the Judiciary's Subcommittee on Administrative Oversight and the Courts;
• Dr. Richardson's memo the the director of the FBI laboratory dated 13 September 1999
• Dr. Richardson's follow-up memo dated 25 October 1999

Note that in reward for his candor, the FBI prohibited Dr. Richardson, their top scientific expert on polygraphy, from testifying in court on polygraph matters.

The federal polygraph program is rotten to the core, and it's high time the fraudulent pseudoscience of polygraphy was abolished. It is to be hoped that the NAS report on the scientific evidence on the polygraph will be the wake-up call that leads to the ultimate abolishment of polygraphy.

In the highlight of the meeting, Dr. Richardson proposed a mock experiment to test the veracity of the polygraphers' claims.  He suggested that he could train 2-3 members of the panel in polygraph countermeasures in a short time period.  He then challenged the 4-5 esteemed polygraphers in the room to polygraph a number of the committee members and come up with reliable results.  The polygraphers would be allowed, he said, to go into a private room and collaborate and attempt to determine who was using countermeasures using their secret techniques.  He wagered that the polygraphers would be defeated, that they would be unable to detect the attempts at countermeasures better than chance, and that they would falsely accuse one or more subjects who knew nothing about countermeasures! To no one's surprise, none of the polygraphers chose to accept the challenge.

In another unsupported statement similar to Dr. Ryan's, DOE polygraph chief David Renzelman commented from the audience that all DOE polygraphers have been trained in recognizing polygraph countermeasures and that there had been a number of recent examples where a determination of "countermeasures used" was made by an examiner and a subsequent confession followed.

Note:  In The Lie Behind the Lie Detector, we make it abundantly clear that the only way that polygraphers can reliably detect countermeasures is to bluff subjects into admitting their usage.  This happens most often in the post-test interview.    

I concur 100% that anyone making such an admission is stupid and probably not intelligent enough to employ these simple techniques anyway.  I considered speaking up and adding my own comments.  In particular, I wished to ask how these polygraphers can say with straight faces that they caught numerous individuals who followed our advice and then attribute behavior to the examinees that clearly runs contrary to one of our most important caveats.  I decided against it, because those on the NAS panel are clearly capable of drawing this conclusion themselves.  

Dr. Richardson replied to Renzelman's boast with, "So then you'll take my challenge."

Clearly agitated at his card having been pulled, Renzelman angrily snapped in response "Did I interrupt you when you were speaking?" He then proceeded to continue on describing how polygraphers have an effective-but-secret "counter-countermeasures" program.  He did not at any point address Dr. Richardson's challenge.

Dr. Andy Ryan once again spoke from the audience during this segment, suggesting that he believes that there is a correlation between attempts at countermeasures and other undesirable behavior.

One of the committee members responded by pointing out that those who seek federal law enforcement positions tend to be those who have pursued them their whole lives and could be expected to do nearly anything if it meant getting the job.

Note to Dr. Ryan and all the others: People practicing countermeasures are not subversive or otherwise dishonest.  They are simply intelligent.

Who are these individuals?  First, these are people who have researched polygraphy.  I hope Dr. Ryan is not saying that the government should not hire anyone who has looked up what is involved in a polygraph "test" on the Internet.  If intelligent, naturally curious people are being turned away in favor of those who believe in "trust us, we're the government..." we are all in trouble.  

Second, after researching polygraphy, these individuals have come to the conclusion that polygraphy is unreliable and that they stand a substantial chance of "failing" even if they tell the truth.  This is the conclusion a vast majority of intelligent researchers will come to, since the polygraph has never shown better than chance accuracy under field conditions and all the studies purporting high accuracy rates were done not by impartial scientists but by those with a vested interest in polygraphy.

Last, these people have chosen to employ simple techniques to protect themselves from a false positive outcome.  I find this hardly a breach of integrity, especially since the flawed "test" depends on numerous deceptions on the part of the examiner.

Dr. Richardson also noted that because of countermeasures, polygraphs have a negative, not a neutral effect on national security.  He continued by stating that anyone who uses simple countermeasures to pass a counterintelligence polygraph essentially receives a five year insulation from suspicion.  

Next, Dr. Richardson presented some very interesting information from an unpublished study done by DoDPI.  The study apparently shows that blacks have a substantially greater chance of being falsely accused of deception by polygraphers than whites.  I eventually plan to reproduce the charts that Dr. Richardson handed out and display them on this website. 

Dr. Andy Ryan attempted to counter this study by noting that the study had been done using student examiners, not those who graduated from the DoDPI program.

Note:  If, as Dr. Ryan implied, what we see in this study is due to examiner errors, how come the examiners are making far more errors with exams conducted on blacks?  If false positives are occurring simply because of examiner errors, they should be expected to occur equally for tests on all races.  In my opinion, this explanation just does not seem plausible.

Speaking of DoDPI studies, during the meeting Dr. Richardson also repeated a statement that he had made publicly before.  He said thatthe Department of Defense Polygraph Institute is hardly a disinterested party and has absolutely no business conducting validity studies on polygraphy.  This, he said, was akin to "having big tobacco companies placed in control of lung cancer research."  DoDPI research, he said, should be limited to studies testing the "effectiveness of using question #3 before question #4, etc." General studies on predictive validity, etc., should be done by impartial scientific institutes, not by those who depend on polygraphy to make a living.

Dr. Richardson's presentation closed with a segment on "Brain Fingerprinting."  He turned the floor over to Dr. Larry Farwell, the pioneer of the technique.  Brain Fingerprinting definitely appears to have some promise.  Dr. Farwell noted that he had achieved 100% accuracy during a study identifying FBI agents from a group of non-agents.  Dr. Farwell noted that it is important to remember that his techniques are not applicable to screening situations (like drug testing, etc.).  Rather, they simply determine whether a certain piece of information is familiar to a certain individual.  Nonetheless, this technology does appear to have significant promise in the counterintelligence arena.  If specific information known only to members of a terrorist group (and not the general public) is uncovered by investigators, brain fingerprinting tests could be conducted to determine with a high degree of accuracy whether or not this information is familiar to suspects.





Overall, Dr. Richardson's talk was extremely informative and enlightening.  I look forward to hearing him speak publicly in the future once again (should he choose to do so).

Here are my notes from recently retired FBI Supervisory Special Agent Dr. Drew C. Richardson's 10/17/01 presentation to the National Academy of Sciences Study to Review the Scientific Evidence on the Polygraph.  This information is accurate to the best of my recollection.   Apologies in advance to Dr. Richardson and others if it contains any errors (the talk lasted nearly 2 1/2 hours and was dense with information).




Dr. Richardson opened his presentation to the committee by bringing up the topic of scientific control and establishing how internal standards (controls) in a serious scientific discipline (forensic toxicology) allow one to determine (1) whether an experiment worked at all, and (2) to draw justified inferences and conclusions about the relationship between internal standards and the things being measured.  He illustrated his point through an example involving science-based urine tests for cocaine and its metabolites. 
Continuing, he then described control question "test" polygraphy and noted that the pseudoscientific discipline fails miserably at both tasks.

Dr. Richardson then went on to describe how there is no well-supported theory explaining why polygraphy works.  He mentioned that even Dr. Charles Honts, a supporter of specific issue polygraphy (but not employment screening) has conceded that even he does not know how it works.  

According to Dr. Richardson, polygraphers often argue that polygraphy is valid because subjects being "tested" have a "fear of detection" (this probably explains why polygraphers extensively trick subjects to create this fear).  He then went on to explain that he did not think that this is a plausible explanation.   

Dr. Richardson next brought up a possible alternative theory to explain polygraphy, "fear of consequences."  While he feels that this theory better explains what is going on, Dr. Richardson noted that plenty of flaws were present here as well. 

Dr. Richardson illustrated this point with a case where a death row prisoner was given a last chance at clemency contingent on passing a polygraph the day before he was to be executed.  He said that the chances of an innocent person passing a polygraph in this situation were virtually zero.  Knowing that any reactions to the relevant questions would result in his impending death (consequences) would be enough to sensitize him to those questions and cause a reaction.  

In the close of this segment, Dr. Richardson questioned how polygraphy could possibly be expected to work on those who know the trickery behind the "test."  Specifically, he wondered how polygraphers expect the silly "control" questions to cause greater reactions than the relevant questions in a person who knows that the controls are meaningless and reactions to the relevant questions will lead to consequences. He then stated that it was his belief that those submitting to a "test" who know the trickery behind polygraphy were bound to fail unless they employed countermeasures.

Dr. Richardson then moved into a discussion of polygraph countermeasures. He said that he believes that they can be employed very effectively by individuals who have merely read instructions from a book.  He then mentioned The Lie Behind the Lie Detector as an effective source of countermeasures information.  He also recalled the previous appearance by Dr. Honts, who stated that he did not believe that polygraphers could detect countermeasures like those contained in The Lie Behind the Lie Detector at better than chance levels.

Dr. Richardson next noted that a friend in the Secret Service told him that "the way to get good at identifying counterfeit money is not to look carefully at hundreds of fake bills but at hundreds of real ones."  He went on to analogize that the way to pass a polygraph is to look at publicly available Department of Defense Polygraph Institute (DoDPI) information regarding "passing" charts and then to manipulate reactions to the control questions in order to mimic them.  The recently retired FBI scientist said that this is something that he and others could do with ease.  

During this segment of the talk, Dr. Richardson exposed the widespread misbelief that the way to beat the polygraph was to conceal one's reactions to certain questions.  He noted that only certain people (those involved in Yoga, meditation, etc) are capable of suppressing the reactions measured by the polygraph.  Furthermore, he continued, even these individuals are incapable of manipulating their reactions during the short time periods when polygraph questions are asked. He explained that this topic is moot, because suppressing one's reactions to the relevant questions is not necessary to create a "passing" chart. In actuality, all that is necessary is to augment reactions to the control questions (which is easily done by anyone)..

A number of polygraphers in the audience responded by asserting anecdotal evidence where they allegedly detected countermeasures and the subject employing them confessed during the post test interrogation.  Dr. Richardson's response was that any person who actually admitted use of countermeasures was simply "stupid" and probably not capable of employing them anyway.

Fortunately, one of the panel members spoke up and asked the examiners if they had any studies supporting their dubious claim. The examiners quickly retreated to their usual statement of "it's classified."