Latest Study Indicates "Lie Behind the Lie Detector" Hurts Innocent, Doesn't Help Guilty

[sombody]

The Lie Behind the Lie Detector is being downloaded something on the order of a thousand times a week. AntiPolygraph.org is not receiving the kind of feedback we would expect to receive were our suggested countermeasures hurting our readers.

I think you would agree that this falls well short of "empirical data" and is not be based in science as to the validity of countermeasures.  When someone uses a product and they do not get the desired result, the lack of complaint to the distributor does not prove the product works.  Retailers know this and make a killing when they sell "manufacturer's guarantees".  Americans are lazy and would most often rather spend their time doing something they enjoy instead of complaining when a product fails to live up to its promises.  

If you were to have scientific proof that countermeasures actually hurt innocent examinees, would you cease advising people to use them?  I understand that you may find faults in any research study that debunks your countermeasure teachings and I doubt that you would have an open mind in any event, but, hypothetically speaking, if someone were to perfect such a research study in which you could find no fault........what would you do?

You are obviously a smart guy and very passionate about your cause, but I only ask that you to keep an open mind.  I would hate to think that your passion against and anger towards the polygraph community would have collateral damage.  "Friendly fire" is the word that comes to mind.  In an effort to "help" innocent people, it appears that there is a distinct possibility (however slight you may think it is) that you may be creating more victims in your crusade.  Many of the posts on this message board talk about one false positive being one too many, but what if your instructions (countermeasures) caused one false positive?  Standing behind the excuse of an individual's "informed" decision will not deflect you from blame.  Most of the people who will download TLBTLD have no experience in polygraph and would most likely believe in your claims that countermeasures would be their only hope of passing a test.  

I think that if I were in your position, I would have to rethink my position, or at least put a disclaimer in the book stating that there is some research (however fallible you believe that research to be) that indicates countermeasures may actually cause a false positive, just to be "fair and balanced.  

[nonombre]

The Lie Behind the Lie Detector is being downloaded something on the order of a thousand times a week. AntiPolygraph.org is not receiving the kind of feedback we would expect to receive were our suggested countermeasures hurting our readers.

I think you would agree that this falls well short of "empirical data" and is not be based in science as to the validity of countermeasures.  When someone uses a product and they do not get the desired result, the lack of complaint to the distributor does not prove the product works.  Retailers know this and make a killing when they sell "manufacturer's guarantees".  Americans are lazy and would most often rather spend their time doing something they enjoy instead of complaining when a product fails to live up to its promises.  

If you were to have scientific proof that countermeasures actually hurt innocent examinees, would you cease advising people to use them?  I understand that you may find faults in any research study that debunks your countermeasure teachings and I doubt that you would have an open mind in any event, but, hypothetically speaking, if someone were to perfect such a research study in which you could find no fault........what would you do?

You are obviously a smart guy and very passionate about your cause, but I only ask that you to keep an open mind.  I would hate to think that your passion against and anger towards the polygraph community would have collateral damage.  "Friendly fire" is the word that comes to mind.  In an effort to "help" innocent people, it appears that there is a distinct possibility (however slight you may think it is) that you may be creating more victims in your crusade.  Many of the posts on this message board talk about one false positive being one too many, but what if your instructions (countermeasures) caused one false positive?  Standing behind the excuse of an individual's "informed" decision will not deflect you from blame.  Most of the people who will download TLBTLD have no experience in polygraph and would most likely believe in your claims that countermeasures would be their only hope of passing a test.  

I think that if I were in your position, I would have to rethink my position, or at least put a disclaimer in the book stating that there is some research (however fallible you believe that research to be) that indicates countermeasures may actually cause a false positive, just to be "fair and balanced.  

Somebody,

You are obviously a highly intelligent and skilled debator.  As such, please allow me to provide a bit of advice.  I have asked various versions of the same question you have posed NUMEROUS times on this site.  Neither George Maschke or any of his minions will answer it.  They can't, for to do so honestly and objectively would be to admit that there is a significant chance, in fact a strong certainty that the "advice" offered on this site in the past half dozen or so years has caused more good, honest candidates to be disqualified from consideration then all the other causes of "false positives" combined.

Bottom line:  These guys are a menace to the very people they claim to serve.  

Regards,

Nonombre :-/

[Twoblock]

Nonombre

You are correct. You have asked NUMEROUS questions and, as well as I remember, they have been answered albeit not the way that you wanted.

On the other hand, over the years, you have been asked NUMEROUS questions that went unanswered or made a feeble attempt at skirting the issue.

If the information offered here is causing you to catch so many using CMs then you should be happy. Instead you come on here and warn prospective agents LE not to pay any attention to what they learn here.

Two more question: 1. Over the last 3 or 4 years what has been your percentage increase of CM users over the prior 3 or 4 years? 2. Over the last 3 or 4 years what has been your percentage difference in pass/failure rate?

I am not trying to be snippy with this post. I am just curious as to why you would warn people not to pay attention to the info on this site and in the percentage differences.

It has been stated here that anyone who would use CMs would not make good LEOs. Therefore, you should be happy that you're catching them. We are all looking for good officers to protect us. I just donated blood to a good officer badly wounded in the line of duty.

[Sergeant1107]

Nonombre,

If you truly believe the results of this study are true, then it would appear you were wrong all the times you compared providing CM information with leaving a loaded shotgun on a subway platform, or in a schoolyard, or whatever.

These test results indicate that CM's don't help the guilty, so if you believe the results you would have to believe that nothing on this site could possibly be compared to leaving a loaded shotgun on a subway platform, or in a schoolyard, or whatever.

I think that either you owe George an apology, or you should go ahead and admit that the study referenced in this thread has fatal flaws in its methodology.

[George W. Maschke]

The Lie Behind the Lie Detector is being downloaded something on the order of a thousand times a week. AntiPolygraph.org is not receiving the kind of feedback we would expect to receive were our suggested countermeasures hurting our readers.

I think you would agree that this falls well short of "empirical data" and is not be based in science as to the validity of countermeasures.  When someone uses a product and they do not get the desired result, the lack of complaint to the distributor does not prove the product works.  Retailers know this and make a killing when they sell "manufacturer's guarantees".  Americans are lazy and would most often rather spend their time doing something they enjoy instead of complaining when a product fails to live up to its promises.

Certainly, reader feedback is not scientific data, but that doesn't mean that it's not a useful indicator (and I very much disagree with your characterization of Americans as "lazy"). Polygraph examinations are usually about something important, and the results can have life-changing consequences for the examinee. Again, we're not receiving the sort of feedback one would expect were The Lie Behind the Lie Detector hurting our readers.

If you were to have scientific proof that countermeasures actually hurt innocent examinees, would you cease advising people to use them?

Yes, most certainly. But there is no such proof, and the Honts & Alloway study tells us nothing about the efficacy of countermeasures under field conditions.

I understand that you may find faults in any research study that debunks your countermeasure teachings and I doubt that you would have an open mind in any event, but, hypothetically speaking, if someone were to perfect such a research study in which you could find no fault........what would you do?

I am prepared to revise my views in the face of new evidence.

You are obviously a smart guy and very passionate about your cause, but I only ask that you to keep an open mind.

Indeed we should all keep an open mind, but not so open, as Richard Dawkins has put it, that our brains drop out.

I would hate to think that your passion against and anger towards the polygraph community would have collateral damage.  "Friendly fire" is the word that comes to mind.  In an effort to "help" innocent people, it appears that there is a distinct possibility (however slight you may think it is) that you may be creating more victims in your crusade.  Many of the posts on this message board talk about one false positive being one too many, but what if your instructions (countermeasures) caused one false positive?

Certainly, misapplication of polygraph countermeasures (for example, mistakenly augmenting reactions to a relevant question instead of to a "control" question) can lead to a false positive outcome. But there is no evidence to support the counterintuitive notion that countermeasure use as a rule increases the risk of a false positive outcome.

Standing behind the excuse of an individual's "informed" decision will not deflect you from blame.  Most of the people who will download TLBTLD have no experience in polygraph and would most likely believe in your claims that countermeasures would be their only hope of passing a test.

Perhaps you should read The Lie Behind the Lie Detector more carefully. Nowhere do we claim that countermeasures are the reader's "only hope of passing a test." From the beginning we have suggested a "complete honesty" approach that involves explaining to the polygrapher that one is fully informed about polygraph procedure and countermeasures. The following is quoted from Chapter 4 of the 4th edition of TLBTLD:

Complete Honesty

A second approach is to be completely honest with your polygrapher. Tell him that you know the lie behind the lie detector. Explain to him that you understand that the true purpose of the "stim test" is to dupe you into believing in the validity of polygraphic lie detection. Tell him that you understand the trickery behind "control" question "tests"—whether probable- or directed-lie. Explain that you understand the difference between "control," relevant, and irrelevant questions and that you have studied and know how to employ polygraph countermeasures. Give him a printout of this book to prove it in a way that he will not be able to later deny. Explain to him that you are not a suitable candidate for polygraphic interrogation, and request that your polygraph "testing" be waived.

One of the authors of this book knows of a Department of Defense employee whose polygraph screening was waived when he explained to his polygrapher that he understood how polygraph "tests" work and that he had received training in how to defeat them.

But beware! While the Wizard of Oz may have meekly admitted to being a humbug once the curtain was drawn aside and his humbuggery laid bare, your polygrapher might not be so accommodating. One graduate of DoDPI has cautioned that if a subject were to follow this "complete honesty" approach, the polygrapher would probably go ahead with the polygraph interrogation anyhow and arbitrarily accuse the subject of having employed countermeasures.

Maureen Lenihan is a case in point. She worked as a research assistant with the federal Commission on Protecting and Reducing Government Secrecy, also known as the "Moynihan Commission."²⁵ She later applied for employment with the CIA. She explained to her CIA polygrapher that she had researched polygraphy while working with the Commission. The polygrapher proceeded with the interrogation anyhow, and later accused her of having employed countermeasures. (Weiner, 1999)

When one of the authors of this book specifically asked then president of the American Polygraph Association, Mr. Milton O. "Skip" Webb, Jr., how an APA member should proceed if a subject were to reveal that he/she has read The Lie Behind the Lie Detector and understands the psychological manipulations involved in both the "stim test" and the "control" questions, Mr. Webb declined to provide an explanation.²⁶

---
²⁵ The Commission's report is available on-line at:

 http://www.fas.org/sgp/library/moynihan/index.html

²⁶ See George W. Maschke's e-mail exchange with Mr. Webb, which is available
on-line at:

 http://antipolygraph.org/read.shtml#informed-subjects

Perhaps now that Skip Webb is a registered user of this message board (and indeed, the originator of this message thread), he might at long last be willing to answer the question I put to him?

I think that if I were in your position, I would have to rethink my position, or at least put a disclaimer in the book stating that there is some research (however fallible you believe that research to be) that indicates countermeasures may actually cause a false positive, just to be "fair and balanced.  

The next edition of The Lie Behind the Lie Detector may well include a discussion of the 2007 Honts & Alloway study. But as I explained earlier in this message thread, the study is beset by serious methodological shortcomings that make generalization to field conditions all but impossible.

Speaking of disclaimers, perhaps polygraphers who administer polygraph screening examinations should include a disclaimer noting the National Academy of Science's finding that "[polygraph testing's] accuracy in distinguishing actual or potential security violators from innocent test takers is insufficient to justify reliance on its use in employee security screening in federal agencies."

[1904]

Nonombre,

That post was like painting a target on your head.

[raymond.nelson]

George W. Maschke:

<snip>

But there is no evidence to support the counterintuitive notion that countermeasure use as a rule increases the risk of a false positive outcome.

Mr. Maschke,

I believe you may be forgetting about his section on page 140 of the NAS report.

There is evidence that some
countermeasures used by innocent examinees can in fact increase their
chances of appearing deceptive (Dawson, 1980; Honts, Amato, and Gordon,
2001).

I just thought I'd point out the oversight as I know you are interested in your readers having access to good information.


r

[raymond.nelson]

George W. Maschke:

Certainly, reader feedback is not scientific data, but that doesn't mean that it's not a useful indicator (and I very much disagree with your characterization of Americans as "lazy"). Polygraph examinations are usually about something important, and the results can have life-changing consequences for the examinee. Again, we're not receiving the sort of feedback one would expect were The Lie Behind the Lie Detector hurting our readers.

Quote:
If you were to have scientific proof that countermeasures actually hurt innocent examinees, would you cease advising people to use them?


Yes, most certainly. But there is no such proof, and the Honts & Alloway study tells us nothing about the efficacy of countermeasures under field conditions.

Quote:
I understand that you may find faults in any research study that debunks your countermeasure teachings and I doubt that you would have an open mind in any event, but, hypothetically speaking, if someone were to perfect such a research study in which you could find no fault........what would you do?


I am prepared to revise my views in the face of new evidence.

Mr. Maschke,

In delayed response to your interest in some additional response to your commentary re Honts & Alloway (2007) methodology, you must recognize that there are some who would consider Mr. Webb's rejoinder to your critique to be not without credibility. Mr. Webb simply pointed out some obvious facts about the expressed concerns.

Terms like "Serious methodological shortcomings," and "fatally flawed," are vacuous hyperbole (the equivalent of a verbal rubber stamp) and have become over-used accusation when one simply wants to discount or discredit a study or findings which one does not like. They are simple-minded cudgels, with which to bash on and reject information which is inconsistent with an apologetic position. When used properly, the term "serious methodological shortcomings," is a generic description of any set of conditions which would reasonably prevent the formulation of any conclusions.

It is an expectation of all investigators to provide some description of the weakness of any project, and attempt to account for such weaknesses. Honts & Alloway have done that. I would note that your concerns about motivation, and other concerns about the generalizability of laboratory data are adequately accounted for in their report. You have not pointed out anything which they have not already stated. Your criticism adds or corrects nothing. Raising those criticisms is a moot point, and is a simple exercise in fault finding. You could, of course, contact Honts and Alloway to inquire whether the lack of description about examiner blindness to the conditions is an editorial oversight based on the implicit obviousness of such a design feature in  project of this type, or is a feature of there experimental design.

You are free to disagree with the idea that the information from the study can provide us any further information regarding the important questions of polygraph validity in the context of prior knowledge. You are also free to formulate your conclusions in advance, and even from your personal experience. But that's not science, and tends to extinguish all opportunity for the addition of new and information into  our existing knowledge base.

No study, by itself, will definitely answer all, or perhaps even any, questions about anything. The rejection of information from laboratory studies as not generalizable due to artificial conditions is an exercise in closed-mindedness, just as the rejection of data from field studies as wholly uninformative due to the  tenuous information about causality and the impossibility of controlling all variables. These are dilemmas that are inherent to all fields of applied scientific study. The common approach is to base conclusions on multiple studies, and to scaffold new knowledge onto existing knowledge. We use field studies to understand correlations, and laboratory experiments to gain a better understanding of causality. We use meta-analytic procedures to aggregate the volumes of data. Interestingly, common practice in meta analytic study is to include all studies – even those of less than ideal design – and simply weight strength of that information in the meta-analysis. Editors of science journals expect a similar practice, when they require discussion about the potential informational shortcomings of any particular project.

Of course, all studies have their limitations, and no single study is adequate to define the limits of our potential knowledge regarding any particular subject. Some would suggest that defining the limits of knowledge is not an objective of research, while defining the threshold of our knowledge.

Reading a research report with a critical concern is an important skill. One can do so with the objectives of finding fault, rejecting the study out of hand, and simultaneously rejecting all opportunity to advance our knowledge about the subject matter. Or one can read a report with critical concern, with the objective of advancing our knowledge when the data and the merits of the study, with consideration for identifiable limitations, support the cautious inclusion or juxtaposition of information with existing knowledge. It really depends on whether or not one is interested in accumulating or rejecting knowledge. Certainly, some people will find some knew knowledge objectionable.

With that in mind, your suggestion that the study is small and lacks statistical power might be considered plainly inaccurate and misleading to your reader. Please be careful.

There is no description of an a priori or post hoc power analysis, and there quite often is none with small studies. To suggest, however, that small studies lack statistical power simply because they are small is a little overly simplistic. Remember that power, in statistical jargon, is the ability of an mathematical experiment to reject a null hypothesis that is false.  There are certainly a number of ways to state any hypothesis and corresponding null hypothesis, and perhaps equally as many ways to design corresponding experiments. While great emphasis is placed on the formulation of null hypothesis in academic training settings, it seems common for published reports to lack an articulated null hypothesis, and include only an implied understanding of such, through the description of the study objectives. Some studies, like this one, investigate a number of related hypotheses.

In this study, some of the the hypotheses seem to have been some version of:

H1: Polygraph scores will be significantly different for guilty and innocent subjects, regardless of prior knowledge.

with the null-hypothesis being:

H01: There is no significant difference in polygraph scores for guilty and truthful subjects.

and

H2: Polygraph scores will be significantly different for guilty or  truthful subjects with and without prior knowledge.
.
for which the corresponding null-hypothesis is:

H02: There is no significant difference in the scores of guilty or truthful subject with and without prior knowledge.

In their discussion of the results, Honts and Alloway provide:

That analysis revealed that CPSp|t values for guilty participants, M=0.40, SD=0.33,
were significantly lower than were the CPSp|t values for innocent participants,
M=0.72, SD=0.33, F(1; 39)= 9.24, p=.004, h2=0.2. Neither the main effect for
information nor the interaction of guilt and information were significant,
F(1; 39)=0.02, ns, h2=0.01 and F(1; 39)=0.19, ns, h2=0.005, respectively.

You may correct me if I am wrong, but it appears there experiment did have sufficient power reject H01, while H02 was not rejected.

This is a good example of the fact that even small experiments can have some sufficient statistical power to reject a null hypothesis and provide interesting information.

Rejection of H01 in this laboratory study would justify the cautious consideration of H1 alongside our other knowledge about the effects of prior knowledge on polygraph results.

Non-rejection of H02, simply means that we would be unwise to cautiously accept the premise of H2 (that prior knowledge will significantly affect polygraph results for guilty or truthful subjects).

In consideration of the possible errors inherent in these conclusions, you should keep in mind that when we reach a statically significant results that allow the rejection of a null hypothesis with small datasets, there is no inherent reason to suspect that lack of statistical power is related to some risk of Type 1 error (rejecting a null-hypothesis that is true).

H02 was not rejected, because the results of the study pertaining to that experiment were not significant. There is some risk of a type II error here (failure to reject a null-hypothesis that is false) - that small sample size masked a real difference within guilty or truthful subjects with and without prior knowledge. But what he have so far is insufficient to support the rejection of the idea that there is no difference. Perhaps a larger experiment would help. We don't know one way or the other, and it is not only unwise but empirically irresponsible to reach conclusions about this hypothesis based only on the sample size and methodology of this experiment. The researchers could not reject, with any stastistical significance, the notion that prior knowledge makes not difference.

That's still a far cry from concluding they will make a difference.  This is simply what we have for now, based on this study. Our task is to thoughtfully consider this information alongside existing knowledge.

It is what it is. Its a small study. Its  laboratory study. Its a single study. But it is not completely uninformative – unless one desires strongly to be uninformed or cannot tolerate the results.





r

[sombody]

WOW!!!!!!!

r.nelson, i may just be some dumb cop, but that sounded brilliant, although I must admit I didn't understand most of it.  

It appears though, based on the past posts and George's reply to my post, that whatever scientific data we are prepared to present which is against his crusade will be dismissed out of hand without any consideration.  He appears to be totally convinced that preaching countermeasures harms no one based on the fact that he has had little feedback from people that have downloaded TLBTLD even when faced with empirical data and scientific studies that prove otherwise.  

George, before you reiterate that fact that you find significant flaws in the studies, we get it......you will search long and hard to find any reason to discount anything that goes against your belief system, but there is no way anyone can believe that the study was so flawed to get ALL the information incorrect.  Even if there was some examiner bias, you can not for a minute believe that all the innocent people who read TLBTLD and were categorized as "false postive" were a result of such bias.  There had to be at least one who failed the test simply because they read your book.  

r.nelson, as I stated above, there appears to be no talking sense into George, but maybe if we stay here long enough we can shed some light for the other innocent people who visit this site.  Maybe we can save some of them even as George is destroying them...............

[George W. Maschke]

George W. Maschke:

<snip>

But there is no evidence to support the counterintuitive notion that countermeasure use as a rule increases the risk of a false positive outcome.

Mr. Maschke,

I believe you may be forgetting about his section on page 140 of the NAS report.

There is evidence that some
countermeasures used by innocent examinees can in fact increase their
chances of appearing deceptive (Dawson, 1980; Honts, Amato, and Gordon,
2001).

I just thought I'd point out the oversight as I know you are interested in your readers having access to good information.


r

Raymond,

No, I didn't forget that section of the NAS report. This is something you more than once alluded to when you were posting as "Ludovico," and to which I responded here:

https://antipolygraph.org/forum/index.php?topic=3696.msg25991#msg25991

and here:

https://antipolygraph.org/forum/index.php?topic=3707.msg26282#msg26282

Again, the "evidence that some countermeasures used by innocent examinees can in fact increase their chances of appearing deceptive" (emphasis added) did not involve the kinds of countermeasures described in The Lie Behind the Lie Detector. As I mentioned earlier, in the Honts, Amato & Gordon study the "countermeasures" were things that subjects ignorant of polygraph procedure did on their own in the belief that it might help them pass the polygraph. Such countermeasures are not comparable to those suggested in The Lie Behind the Lie Detector.

As for the 1980 study by Michael E. Dawson ("Physiological Detection of Deception: Measurement of Responses to Questions and Answers During Countermeasure Maneuvers," Psychophysiology 17 (1), 8–17), as explained in the article abstract: "All subjects were trained in the Stanislavsky method of acting and were instructed to use this method to appear innocent on the polygraph test." Again, this is nothing at all like the countermeasures suggested in The Lie Behind the Lie Detector.

Once more: there is no evidence to support the counterintuitive notion that countermeasure use as a rule increases the risk of a false positive outcome.

[nopolycop]

In doing more research, I have come accross the following reference to a study by Honts stating that countermeasures worked about 50% of the time.  What gives?

"Effects of countermeasures on the CQT polygraph test were examined in an experiment with 120 subjects recruited from the general community. Subjects were given polygraph tests by an examiner who used field techniques. Twenty subjects were innocent, and of the 100 guilty subjects, 80 were trained in the use of either a physical countermeasure (biting the tongue or pressing the toes to the floor) or a mental countermeasure (counting backward by 7) to be applied while control questions were being presented during their examinations. The mental and physical countermeasures were equally effective: Each enabled approximately 50% of the subjects to defeat the polygraph test. ... Moreover, the countermeasures were difficult to detect either instrumentally or through observation.

C. R. Honts, D. C. Raskin, and J. C. Kircher, "Mental and Physical Countermeasures Reduce the Accuracy of Polygraph Tests," J. Appl. Psych., v. 79, n. 2, 1994, pp. 252-259. "

[Barry_C]

This is the study that gives Doug Williams his ammo.  The CMs "worked" for those who received hands-on training (practice), and the examiners could didn't take any steps to "neutralize" the CM attempts, and they scored the charts regardless of what was there.  (You can't move with a motion sensor without getting caught, and they didn't have one.)

Charles position is that reading about CMs won't help you. (See his chapter in, I believe, Granham (sp?) for details.) The (healthy) fear is with spies where it may well be possible for them to receive hands on training, practice, feedback, etc, but that needs more research.

[1904]

This is the study that gives Doug Williams his ammo.  The CMs "worked" for those who received hands-on training (practice), and the examiners could didn't take any steps to "neutralize" the CM attempts, and they scored the charts regardless of what was there.  (You can't move with a motion sensor without getting caught, and they didn't have one.)

Charles position is that reading about CMs won't help you. (See his chapter in, I believe, Granham (sp?) for details.) The (healthy) fear is with spies where it may well be possible for them to receive hands on training, practice, feedback, etc, but that needs more research.

The mental and physical countermeasures were equally effective: Each enabled approximately 50% of the subjects to defeat the polygraph test. ... Moreover, the countermeasures were difficult to detect either instrumentally or through observation.

Which part of the above did you not understand?
You just cant seem to deal with reality....shame.

[Barry_C]

I presume your question is in regard to the bold text.  I fully understand it.  I'm sorry to have confused you.

[Brettski]

Was there ever a pdf copy of the research posted to this thread that I can read?