Barry_C wrote on Dec 26^th, 2007 at 3:16pm:


You know full well what I'm talking about.  Telling someone that he can't get a job if he's ever told one lie ever, when such is not the case, is deceptive.  Cut the crap.

Barry_C wrote on Dec 26^th, 2007 at 3:16pm:


I guess you just got the results of that study in the past few weeks, huh?  You didn't know what the answer was two years ago?  Seriously, why did it take over two years to get that simple yes/no answer?  Did you honestly not know the answer back then?  You acted as if you did.  At what point did you discover the answer to the yes/no question in question?

Everyone, please forgive the Reverend Barry.  He didn't read the part about letting your "yes" mean yes and your "no" mean no.  Anyway, he's just acting out his training; he's programmed to not answer straight questions that might lead to difficulties for his guild.

Barry_C wrote on Dec 26^th, 2007 at 3:16pm:


Hmm.  Never answer someone's questions forthright and then call him ignorant.  Anyway, my main point (that there is certain knowledge the possession of which decreases the usefulness of the polygraph exam) is pretty much confirmed every time you play these little games to avoid answering questions.  If the premise was false, you could quickly dispatch it by revealing what you struggle so mightily to conceal.

Oh, Barry, I see that you haven't logged on in about three weeks.  I hope you're still here with us, reading posts without logging in.  I also hope it's not too cold up there around Windham, but with a name like that it's gotta be.  Anyway, congrats to your Patriots for making (another) Super Bowl and come back to us soon!

Lethe,

I've (and others) responded to you ad nauseam, but you don't want to listen.  You have an agenda you can't look past it objectively.

Quote:


First, what do you consider "deception"?  It's not a crazy question.  Do a little research, and you'll see that there is no consensus in the psychological literature as to how to define it (or a "lie").

Again, not all examiners use the "deception" to which you refer (without defining, I might add).

Quote:


Again, your intellectual laziness is apparent once again.  Read the study.  That is the finding.   You may not like it, but that's what they found.

Quote:


I'm not sure what you mean, but it sounds like you're asking the right question to begin a study of your own.  Why don't you try that?  Maybe you could contribute knowledge to the field of the detection of deception rather than slinging mud based on your rather uninformed opinions?

Quote:


Can you list those insults here?  Should I avoid the truth or candy-coat it to avoid hurting your feelings?  Wouldn't that be "deceptive"? Is pointing out inconsistent logic insulting?  Is pointing out that you are often intellectually lazy wrong?  My intent is not to hurt your feelings.  If I did so, I'm sorry.  I appreciate the "tell it like it is" approach to skirting issues, which is why I don't take offense with your passion - as ill informed as you are at times.

Dr. Lethe;

It appears to me the reason the polygraphers come onto this forum is to attempt to discredit George Maschke, with the purpose to taint the influence of this site on their "profession".  I asked a sincere question about whether or not Georges discreption of the control questions asked in a poly exam was correct, and none would either confirm the correctness of what has been written in LBLD, or discredit it.  Of course, it is better to be silent than either confirm that George is correct, or obviously lie. (since their real identities are now known).

I predict they will soon all run away, as they are hurting their cause more than helping, because when they attempt to engage in their plan to discredit George, they result in looking like fools.

Barry_C wrote on Dec 20^th, 2007 at 12:19am:


Good grief.  It's been, like, two years that I've been trying to get this thing to answer my question, does the deception used by polygraphers have a purpose?  It's a simple yes or no question and, apparently, Chaplain Barry has finally, after numerous personal insults, indicated that the answer is yes.  Thank you.

According to the chaplain, the deception does not improve the accuracy of the exam itself but it "help[s] better identify the deceptive" in some unspecified manner.  Now, why would it take two years to get that answer?  I wonder.

In any event, how does it help identify the deceptive?  Simply claiming that it does is hardly an answer at all, one would hardly expect the practice to help identify a good place to have lunch.

Additionally, I'm sure that thinking the practice is benign to the innocent helps you sleep at night, but it seems unlikely to be true.  Think about it.  Is the average person more or less likely to be honest when someone is actively trying to trick her?

Maybe I'll have an answer before the start of the next decade!  Oh, and Barry, I hate to break it to you, but I'm not going to be sending you a Christmas present this year.  (That doesn't bode well for me getting an answer before the next presidential administration, Barry only answers questions from his good buddies and not from people who don't uncritically accept whatever answer he grudgingly proffers)

If you want to find out where I'm going with this, answer the questions.  Thanks.

Quote:


What you'd find if you were to read the study is that your "BS" approach does nothing for the truthful, which is why I've told you time   and again, it's unnecessary.  Don't mistake that with useless, however, as they found that it does help better identify the deceptive.  Now that your house of cards has fallen, will you be honest or will you continue spewing your unsupported theories (that are contradicted by Dr. Rovner's work)?

By the way, why do you have such a problem with the way some examiners introduce the CQTs?  You act as if you are the ethics police.  Is is wrong when researchers "mislead" their subjects during an experiment?

Mr. Maschke,

Thank you for clarifying the cite for Patrick and Iacono 1989. 

I don't think it really impresses knowledgeable people to quote Lykken, who refers to the Forman and McCauley study - for which the investigators did not numerically score the results, regarding a technique which has been largely uninteresting and unimpressive. Surely Lykken must have been aware that the absence of numerical scores would prevent the authors from any ability to understand the level of significance at which they permitted a decision. Its enough to make one think.

Also Patrick and Iacono made little effort, to evaluate their decision rules as a source of bias, other than changing the symmetrical inconclusive zone, to , despite the fact that they were well aware of the differences in the distributions of scores of deceptive and truthful persons. As they had every ability and opportunity to evaluate those distributions, using the very simple chi-square tests they used elsewhere in their study - one has to wonder why not. It is evident from their data that they were well aware of the asymmetry of their data, so this oversight seems to have been somewhat unnecessary, don't you think? Almost as if they didn't want to know.

Also, I'm curious what you think about their findings, from the DA involving their STAI scores, that it seemed higher levels of anxiety or arousal were not associated with errors.


r

nopolycop wrote on Dec 16^th, 2007 at 4:40pm:


Are any of the poly examiners going to answer this honest question?  Mr. Webb?  Mr. Nelson?  Mr. Cushman?  Mr. Johnson?   

Can I assume by your silence that this is in fact correct, and you simply don't want to acknowledge it here?

SanchoPanza wrote on Dec 14^th, 2007 at 8:35pm:


Sir, 

If myself and an eminent group of like-minded people conducted a mock-situation study of witches & That study was peer reviewed by another eminent group of like minded people - and they then came out in support of the witch study. Based on that peer reviewed research, 
would you believe that under the guidance of a full moon, that 
women could wear pointed hats and take-off on brooms ???

Perr reviewed research of bs = verified bs.

Respectfully,
JP

nopolycop wrote on Dec 16^th, 2007 at 4:40pm:


Sir,
The book TLBTLD is a very useful and accurate manual iro the workings of the polygrapf and the deception practised by the examiner.
It iss true that if the responses (poly tracings) to the CQ are greter than those for the Relevant Issue questions, then one has 'passed' the test. Wether the examiner will actually be truthful and advise you that you have passed, is another issue altogether.

As you have read here many times, polygrapf examiners are generally dishonest people, supporting dishonest technologie, with dishonest bs methods that are handed down through the ages.

Respectfully,
JP

Question.  Where the report says that some of the subjects had the control questions "explained" and others didn't, is that talking about the nomal, B.S., deceptive "explanation" that polygraphers normally give?  Or is it talking about a real, truthful, and in line with reality explanation on how the control questions really work?

If the former, that is very strong evidence that the subject needs to be deceived in order to produce accurate results (and thus that people who know the truth will produce charts that are little better than chance).  If the later, it should convince you polygraphers to cut the crap and be truthful with people since it won't harm accuracy and will lead fewer innocent people to try countermeasures.

So, which is it?  Or are you guys going to pull your normal crap and insist I spend $40 for the mere possibility of maybe, perhaps having my question answered in the actual report?

Okay, someone who has had polygraph training please confirm or discredit the following information.  In TLBLD, George states:   

"The dirty little secret behind the polygraph “test” is that while the
polygrapher admonishes the examinee to answer all questions truthfully,he secretly assumes that denials in response to certain questions—called “control” questions—will be less than truthful."

He also states:

"The polygrapher scores the test by comparing physiological responses
(breathing, blood pressure, heart, and perspiration rates)
to these probable-lie control questions with reactions to relevant
questions such as, “Did you ever use an illegal drug?” If the former
reactions are greater, the examinee passes; if the latter are greater,
he fails. If responses to both “control” and relevant questions are
about the same, the result is deemed inconclusive."

Are these statements true?

Quote:


Sancho Panza,

My comments (I presume those quoted above are the ones you are referring to) had nothing to do with the study you refer to.  They dealt specifically with what I believe to be a recent phenomenon and trend of misrepresenting David Lykken's opinions regarding lie detection (in particular with the scientifically uncontrolled and unstandardized nature of the PLCQT).  I rest assured that David Lykken's opinion will not change posthumously, and unless the basic nature and format of the CQT has changed in my absence, the study that you refer to will not address the concerns that I raised.  That having been said if you think there is purpose in my reading the study, I would be happy to do so following the holidays.  Best, Drew Richardson

Dr. Richardson, 
My criticism of Dr. Maschke's comments was based on the fact that he was attempting to evaluate the value of a research project he had not  read by using generalizations obtained from another researcher who had also not read the research project. Now it seems that you are doing the same thing. 

Nothing in your post indicates that you have reviewed the Offe study so your conclusions are like Dr. Mascke’s, mere supposition. 

It makes both of you appear to be against further polygraph research and you seem set against any attempts by researchers to address the criticisms raised in the NAS committee report. Had the Wright brothers believed that aerodynamic research had gone as far as possible, our world would be quite different today. If you are so sure that your conclusions about polygraph are correct why are you so resistant to further research?

Even though the Wright brothers achieved flight, I would offer that their research methods would not withstand modern peer review since neither of them achieved a formal education beyond high school. Still, somehow airplanes fly.   

It doesn't take a doctorate to see that if one has a poor experience driving a Chevrolet Malibu that it doesn't necessarily mean that all Malibus are bad cars and that nothing about the experience has any relevance at all to Chevrolet Impalas without careful review and comparison. 

It just seems that you are offering your criticism prematurely. I sincerely look forward to your comments after you have an opportunity to review the complete study

Sancho Panza

Raymond,

Lykken's endnote for the relevant study (which I omitted from my quotation above) is: "C. Patrick and W.G. Iacono, Psychopathy, threat, and polygraph test accuracy, Journal of Applied Psychology, 1989, 74, 347-355 (specifically pp. 348-349)."

Mr. Maschke,

Will you please provide a more complete citation for your reference to the Patrick and Iacono study with prison inmates?

A careful look at their other study, the one listed in the reference lists at this site, will reveal that they reached their conclusions while failing to control for or evaluate their own procedural and decision rule bias in their blind-scoring experiment. One would almost guess that kind of blind-spot or oversight to be a form of bias itself.


r

Yes, nopoly4me...the comparison questions of the present control question test are the same as were previously (prior to about 15 years ago) referred to as control questions.  Although the polygraph community is to be commended for more correctly referring to what these questions are (and by omission what they are not—a form of scientific control), they are in no way excused from the necessity of providing scientific control for what they purport to be a "test."

Dr. Richardson, could you address my question, is a Comparison Question Test the same as a Control Question Test?  If not, what are the differences?

Thank-you.

Thank you for the correction, nopoly4me.  I will make the change to reflect that which you have noted.

[quote author=4D7B6C7E565B606A61687B6D7A6667090 link=1197651321/0#8 date=1197726601 I feel a bit like Ed Muskie listening to Dan Quayle speak about JFK.      
[/quote]

Point of correction.  I believe you are referring to LLoyd Bentsen, not Ed Muskie. 

Every now and then, when I hear some of the musing that comes forth about David Lykken and what he would think, I feel a bit like Lloyd Bentsen listening to Dan Quayle speak about JFK.  David and others have spoken about the weaknesses of probable lie control question tests (PLCQTs) in largely two contexts.   

One is “What is the proper way (environment) to analyze witchcraft?" consideration.  Within this context, issues of a lack of external validity with laboratory studies and the discarding of data points (the discarding of false positive polygraph results (obviously) not followed by confessions) with confession-based (i.e., ground truth determined through confessions) real case studies come into play.  Both of these  phenomena make questionable the results of any studies so impacted.  These considerations are however not at the core of the more serious shortcoming of this practice.   

As has been said time and time again quite eloquently by John Furedy, the probable-lie control question test neither contains any form of scientific control nor resembles any sort of recognizable test.  It is centered on a roughly 30 minute to hour and a half unstandardized interview.  The interview has several purposes, some largely unrelated to the administration of the test (e.g., collection of demographic data) and at least one serious and universal confound (developing theme material to be used in an interrogation of a subsequently found to be deceptive examinee following the in-test phase) but largely focuses on the “setting” of control questions in relationship to relevant questions.  This is a vague notion and has no objective measures for ascertaining whether examiner lies regarding question types are believed by the examinee nor more generally whether this “setting” has been accomplished.   

As previously stated, this witchcraft is further characterized by being unstandardized.  Every test is different.  One hundred exams (more specifically the pre-test interviews) given to 100 examinees about the same specific incident are all different.  Any two exams about the same issue given to the same person are different.  As opposed to the ability to give tape-recorded instructions and an administered test with a guilty knowledge test, the currently practiced PLCQT is nothing more than an unstanderdized interview administered differently by every examiner on every occasion. This may be poor art but it is most assuredly not good science.

I believe it is in regard to this latter set of considerations that David was referring when he said that which is quoted on the home page, "...the theory and methods of polygraphic lie detection are not rocket science, indeed, they are not science at all."







     

Okay, I am a little confused.  (EJohnshon, insert your impending smarmy comment here).  Anyway, I believe I understand that a "Control Question Test" relies upon the testee answering one more questions on the test that are designed to elicit a lie, then the physiological responses as measured by the polygraph machine are compared against the relevant question, (the subject area being tested).

My confusion stems around the phrase "Comparison Question Test."  Is this the same as the "Control Question Test?"

Dr. Maschke,, 

We both know that it is an absolute impossibility for David Lykken to ever publish a review of the Offe study.  I do not think it is reasonable to state that the quoted passage is germane to the Offe study until you have the time to review the study and its findings to illustrate a FULL comparison of the differences and similarities that may or may not exist between their study and the studies analyzed years before by David Lykken. Especially since he died before the Offe study was released. 

Whether or not one agrees with any or all of anything David Lykken ever wrote or said, it is fair to give him credit as a driving force behind continued polygraph research. He managed to inspire those who agreed with him regarding polygraph and challenge those who disagreed, which in my humble opinion is a greater legacy than any conclusion he ever reached. To invoke his name in such a haphazard fashion is a disservice to what he stood for. 

Although I cannot conceive that it was your intent; a premature statement like the one you made might lead someone to conclude that the Offe study somehow possessed whatever flaws David Lykken believed existed in the other studies that he analyzed for his book.   

It is just as likely that upon review of the Offe study, David Lykken would have changed his mind completely and retracted the very comments you quoted.  If I had made the presumptuous statement that the Offe findings would have changed his mind absent a careful analysis and supporting argument, I believe that you would have been among the first to criticize my conclusion, and justifiably so.   

I look forward to your informed analysis of this study after the first of the year. 

I know you are busy and I am still,
Sancho Panza

Skip,

Thank you for posting this. (I would point out that following the citation, the first paragraph of your post is the article abstract. The text that follows is your commentary on the article, yes?) I'd like to read the entire article before commenting on it, and don't expect to be able to do so before the new year. However, based on the excerpt you've shared here, it seems that Lykken's summation of CQT laboratory studies published in scientific journals (at pp. 132-33 of A Tremor in the Blood: Uses and Abuses of the Lie Detector, 2nd ed., endnotes omitted) is germane:

Quote:

I am looking forward to George's analysis of this study, as I am not convinced that it is worth MY money to review at this time.

For thise unfamiliar with Springer the link is

http://www.springerlink.com/content/7x837227611x7734

It's quite interesting, but I will be very surprised if Mr. Maschke posts a copy on this site.  Since it is so unlikely that it will appear here, I would like to say that Mr' Webb's summary is quite accurate. But if you are curious, it is worth the money.

Sancho Panza