posted 08-26-2006 10:11 PM            

FYI, all researcher are bias. It's when they conclude they were wrong that you should place the most trust in them. You'll find some of those in the research.

Just so you know, the NAS did an independant review in 2003. They did no new research, but looked at some of what you plan to delve into. They said polygraph (single-issue tests) work and work well (86%). Other studies found it to be better.

posted 08-29-2006 05:35 AM            

Read Honts and Raskin's chapter in Murray Kleiner's Handbook of Polygraph Testing as another base document on which to start your search. They look at the high-quality research out there, something the NAS (surprisingly) didn't do. If I recall correctly, they discuss knowledge of the test and test accuracy.

posted 08-29-2006 08:00 AM            

The Psychophysiological detection of deception.

It's in Detection of Deception in Forensic Contexts, edited by Granhag and Stromwall, and it's published by Cambridge University Press.

It discusses your questions.

posted 08-29-2006 02:40 PM         

From your response, one question immediately comes to mind. Without having read anything from Rovner or Honts, if it is true that "it doesn't matter how much the [polygraph subject] knows" about how the technology really works, why aren't polygraphers up front with that information? Even if ignorance in the subject does marginally improve accuracy, there appear to be costs to the deception.

Bear in mind that my main interest is in the use of polygraphy for employment security screening. In such a situation, when people find out about the deception involved on the other side of the box, they might come away with the message that those in authority over them think that it is indeed okay to lie if you think you can get good things by doing so. (Subjects who take and pass a Probable Lie Comparison test, which needs the subject to lie, may also figure that lying is okay if it gets them good things)

posted 08-29-2006 07:53 PM            

There are some who have posted here saying knowledge of the test has hurt them. The research, for the most part, looked to see if knowledge could make it possible for the guilty to slip through the cracks as that is where the fear is.

posted 08-30-2006 01:44 AM         

Additionally, I am still curious about my prior question which, as you rightly point out, does not apply equally to all forms of polygraph testing. But so far as the probable lie test is concerned, if it is true that knowing how the exam works is irrelevant to the results a subject will produce, why does the polygrapher deceive the subject about the exam?

posted 08-30-2006 01:06 PM            

Exams designed to protect national security (your "security exams," I presume) are post-hire exams, and they typically use the TES, a DLCQ test.

The LEPET is used as a pre-employment screening exam. I don't know how many actually use that particular one any more though.

As to why PLCQs are popular, the answer is (simply) that they work. I've yet to have anybody admit (during the pre-test) he knows a lot about tests, and since they work regardless of what one knows, then why change things? (I don't care how much the person knows as it's not going to help him, so that's not probed too much.) Additionally, they are preferred because they tend to be slightly more accurate than DLCQs, which makes them more fair to the examinee.

You can run a PLCQ test without "lying" as you call it as PLCQs don't necessarily require a person to lie. You have spent too much time reading the anti-polygraph sites where the info they provide is very selective.

posted 08-30-2006 02:08 PM         

(1) In a PLC test, does the polygrapher lie to and/or deceive the subject in any way regarding how the test works?
(2) If yes, does the subject's knowledge of a how the PLC test works influence the results produced?
(3a) If not, then why is the subject lied to and deceived?
(3b) If it will, why is/was the PLC test used if there are, as you claim, other polygraph tests which avoid the pitfall in (2)?

These are pretty straightforward questions. (1) and (2) can be answered with a simple yes or no, though if you'd like to elaborate please do. I appreciate that your knowledge of polygraphy dwarfs mine and was acquired over much time and from numerous sources. But these questions are very narrow and, I think, important. I don't expect a totally comprehensive answer, but anything satisfactory would be greatly appreciated.

posted 08-30-2006 02:58 PM            

Question one isn't a simple yes or no question because it depends on the type of PLCQ test that is run. So the answer is yes, possibly, and no.

The answer to question two is no, which I already answered.

I answered question three already too. You don't know what the examinee knows, so you stick with the protocol you are used to using. If the examinee doesn't know about the test, then we expect it to work as it does in the research. If the person does know about the test (and doesn't speak up), he still knows what questions are important to him (to "pass"), which should make them more salient - exactly what happens with the naive examinee. The same theory holds true with DLCQs. PLCQs are much easier for most examiners because that's what they are used to using. There is no "pitfall" from the examiner's perspective.

posted 08-31-2006 06:39 PM         

(1) There are two ways to conduct a Probable Lie Comparison (PLC) test, first by giving deceptive instructions to the subject (let's call it "the deceptive method") and second by being honest with the subject about how the exam works (let's call this "the straightforward method").
(2) Both methods have an equal chance of producing accurate results.
(3) Regardless of which method is used, the subject's knowledge or ignorance of how the test works makes no difference in the results.
(4) However, if the polygrapher knows the subject is aware of how the exam works, he will probably use the straightforward method. If he knows the subject is ignorant, he will probably use the deceptive method. If the polygrapher doesn't know whether or not the subject knows how the test works, the polygrapher will use whichever method he is most comfortable with.
(5) Most polygraphers are more familiar with the deceptive method and so will normally opt for that if unsure what the subject knows.

However, I still do not understand why the polygrapher would ever lie to the subject of the exam. Because he is more familiar with that method? Okay. But why is he or she more familiar with that method? Why would it ever be used, let alone used to the point that the polygrapher isn't comfortable being honest? Your answer, as I read it, gives no reason for ever deceiving the subject. You say that the deceptive method does not increase accuracy and you claim no other advantages for it whatsoever. But, as I argued in an earlier post, lying to the subject appears to have drawbacks (namely, sending mixed messages about lying, perhaps making the subject feel insulted, etc). Unless you deny there are any costs to using the method, surely you must be able to point to benefits that will outweigh these costs.

I feel my question still remains. Why would a polygrapher ever lie to the subject of a polygraph exam?

posted 08-31-2006 08:06 PM            

What is "deceptive" about asking the CQs?

That is an idea propagated by the anti-polygraph crowd, but it's inaccurate. They somehow think that if an examiner weren't completely truthful in his approach, that would be unethical, and therefore invalidate the test. That's a philosophical issue, and in our society, it's wrong. The police can and do lie to people all the time, and the courts support it. The confession is no less valid. If it is helpful to "lie" to an examinee to give him a better chance of passing the test, then the issue is one of fairness and necessity.

There are questions that are considered PLCQs that require nothing but an examiner asking them and plugging them in to the test. There's no deception there.

To date, there is no all-encompassing theory as to why the test works, so the "straightforward method" is not accurate either. The methods are described in Honts and Raskin, so start there.

2) Since you can't seem to grasp the issue in 1, this question doesn't mean much.

3) That's what the research shows.

4) I explained this as best as I could above. Since you don't seem to be following me, I'm not sure what to say. I'm not going to write a book here. I think you are talking about DLCQs vs PLCQs - two different types of tests, and the DLCQ is the test of choice, for those who know how to do it, of those who fully understand the PLCQ test; however, even if one does, you can still run a PLCQ test. The decision would depend on a number of factors. Since they'll both work anyhow, it really doesn't matter.
5) Most examiners who run CQ tests run PLCQs - not DLCQs. I never said only the inexperienced would deceive a subject. Why is saying "it is only done some of the time is no answer"? It is the truth. Are you asking me to lie?

Why would a polygraph examiner ever lie to a subject? What if he believes the person is guilty? I have before. I have to be neutral in my test or the results will be worthless. I tell the person I haven't made a decision on truth or deception, but sometimes I can't help myself. It would be unethical for me to accuse the person as the chances of passing go down. Instead, what is ethical is to assure the subject you are neutral, and you run the test that way. And guess what? Sometimes we are wrong, and the person is innocent. We could have scrapped the test all together and concluded guilt based on case facts, but instead we did the right thing in order to find the right person.

Why is it wrong for an examiner to lie if he so chooses to conduct a valid test? How many times did you lie this past week? The research shows most people lie two to 25 times per day, but only polygraph examiners are prohibited from doing so even with good intentions? Have you ever told somebody he or she looked good when that wasn't true. Why would you do that? Have you ever complimented a person but didn't mean it? Why?

I suspect you have come here with motives other than what you have presented, but I don't know that for sure. In any event, I will do the best I can to help you grasp this if you are sincere, but prove it to me by reading a good sample of the research. I started on the anti site before I became an examiner, and then I read the other stuff out there. The rest is history.

posted 09-06-2006 12:07 AM            

Do all examiners lie to set up a PLCQT of course not , does the test require a lie in response to a question to have a control that works ,the answer to this is no as well uncertainty about the truthfulness of the answer will work as well.

Even if you understand all the nuts and bolts about the PLCQT , if you are asked a question you have to make a concious decision to lie to answer truthfully or you may not know if your answer is truthful or not but you dont recall anything specific so do you mention this or just keep it to yourself. If you identify a control then does it matter as you realise this is a important question that can affect test outcomes so that in itself should elicit a response.

All the research on that subject comes to the same conclusion prior knowledge does not affect test outcomes.

It is fairly obvious you are just trying to have a debate not really asking a question that you really need an answer to.

posted 09-27-2006 07:31 PM            

Here are the references to the studies done on this subject Dr Rovner and Dr Honts looked at prior knowledge as a way to "beat" the test their findings concluded it did not effect accuracy rates compared to those who did not have prior knowledge the differences if any are not significant.

In my own practice it has been my experience to have clients "fail" exams after applying countermeasures they read about off the internet and correctly identifying the controls, and applying cms, they confessed to this after some discussion that they did in fact try this and I asked them what question they thought the control was and they had picked the right ones but the outcome was not effected.

However that is a different issue to your question just knowing how the test works and what purpose each question serves will have no influence on test outcomes.Some of these articles then take the next step of applying countermeasures.

Scientific Journal articles:

Honts, C. R., Raskin, D. C., & Kircher, J. C. (1994). Mental and Physical countermeasures to reduce the accuracy of polygraph tests. Journal of Applied Psychology, 79, 252-259.

Here they provide subjects with detailed information about how the test works and offer possible countermeasures.

Rovner, L.I., Raskin, D. C. and Kircher, J.C. (1979) Effects of information and practice on detection of deception. Psychophysiology, 16, 197-198.

Horowitz, s.w., Kircher,J.C.,Honts,C.R.and Raskin,D.C.(1997) The role of comparison questions in physiological detection of deception. Psychophysiology,34, 108-115

Rovner,L.I.(1986) The accuracy of physiological detection of deception for subjects with prior knowledge. Polygraph , 15, 1-39.

There are more but that is all I can recall from past conversations about this subject.

hope that gives you aplace to start.

posted 05-21-2007 02:37 AM            

Honts, C. R., Raskin, D. C., & Kircher, J. C. (1994). Mental and Physical countermeasures to reduce the accuracy of polygraph tests. Journal of Applied Psychology, 79, 252-259.

Here they provide subjects with detailed information about how the test works and offer possible countermeasures.

Rovner, L.I., Raskin, D. C. and Kircher, J.C. (1979) Effects of information and practice on detection of deception. Psychophysiology, 16, 197-198.

Horowitz, s.w., Kircher,J.C.,Honts,C.R.and Raskin,D.C.(1997) The role of comparison questions in physiological detection of deception. Psychophysiology,34, 108-115

If a subject has prior knowledge will this effect the accuracy the studies above conclude it does not.

Examiners follow guidelines they are given, researchers have examined the affect this has on test outcomes studies on the subject published in mainstream scientific journals(listed above) conclude it does not change outcomes when a subject has intricate knowledge of the process.

Either you dont accept the research and draw a conclusion based on an assumption like the NAS do. The National Academy of Sciences (NAS) after culling through the numerous studies of the accuracy of polygraph detection, the NAS identified 57 that had “sufficient scientific rigor”. According to you they were not satisfied with the existing research despite them making the above statement, then go and draw a conclusion without research to back it up making assumptions and calling it fact.

show me your research that states the opposite that your assumption has been tested and proved that knowledge does effect outcomes.

If your point is to try and state polygraph examiners that use PLCQT are all liars you are entitled to your opinion. Only a $#!* veils an insult with a question.

The question of prior knowledge has been the subject of research and all the studies conclude the same it has no effect.

If that is not good enough for you then you have a problem which we cant fix as you beleive your assumptions are better than research.

posted 05-21-2007 11:44 AM         

Your post was a fine one, but it didn't address the question: What is the purpose of the lies told during PLCQ exams if it is not to increase the accuracy thereof?

posted 05-21-2007 12:15 PM         

My point is not to state that every polygrapher who conducts PLCQ tests is a liar. It might be the case that they all must lie; certainly it is the case that the literature I have read which is put out by members of the polygraph community describes a lot of deception. What I think you don't understand about what I'm getting at here is that admitting that lies are frequently used by examiners would be the beginning--not the end--of the discussion.

You seem to assume that my point is that it is never, ever okay to lie no matter what. That is not my point. Indeed, I think there are circumstances under which it is okay.

Dietrich Bonhoeffer was a Christian theologian who said just that. He was a German who lived during (but not through) the Holocaust and gave the example of a family who was hiding Jews in their basement. When the SS men came and asked if they knew of any Jews in the area, the family would be doing a great evil if they told the truth and revealed the Jews they were hiding. But neither could they simply not answer the question, that would surely reveal that they were hiding something by not answering (take note) and would probably lead to the house being searched. So, Bonhoeffer said, it is not only okay but good for you to lie in such a circumstance.

Is the polygraph such a circumstance? I have no idea. To know that I would need to know the purpose of the lies, which I do not. Why don't you answer the question, and let me reach my own conclusion on the matter? What is the purpose of the lies told during PLCQ exams if it is not to increase the accuracy thereof?

If you will not or cannot answer the question under any circumstances, just say so. Maybe say "I appreciate your interest, but that is a trade secret that we cannot reveal. You might as well ask Coca Cola for the coke recipe." That'd be a perfectly legitimate answer.

Your last post amounted to "knowledge of the exam doesn't decrease accuracy" which seems to be the same as "deception doesn't increase accuracy." But I already am assuming that in the question. I'm asking "What is the answer to this question if it is not X?" and your response is "The answer is not X." Well, I've already assumed that.

Please either answer the actual question that I'm asking or indicate that you cannot do so, providing a reason for the inability if possible. If you can't answer, we're kind of wasting each other's time here, and there's no need for that.

posted 05-21-2007 12:24 PM            

quote:

---

The fact still remains that examiners attempt to deceive their subjects. Why?

---

If they only attempt to deceive, then they don't deceive; therefore, they can't be liars - only hopeful liars.

I ran a CQT this morning, and no deception was involved. If I ask you a relevant question and a comparison question and you choose to lie to either, then why am I the liar? Just because I can set up the test in such a fashion doesn't mean I have to do so.

quote:

---

I'm not looking for proof since, philosophically, it is impossible to prove anything other than your own existence; I'm looking for reasons to believe....

---