posted 06-22-2006 06:56 AM            

1.What are the different points of view on this subject?
2.Are there certain situations or formats where this works better than others?
3. What does available research say?

posted 06-22-2006 08:51 AM            

There was quite a series of exchanges in the journal a couple of years ago - between honts and Abrams, I believe (now I'll have to check).

Abrams made some court testimony and publication that was, in my view, premature and based upon anecdotal experience and opinion that he handed off as research. Sorry to be so critical about this, I really like most of Abrams work and contribution to this field.

TES is a DLC test, but its structure and strict operational definition do not generalize well to situations outside its intended use. The TES is one of the best tests yet developed. You can find some info on these (unfortunately) on the anti site.

The Utah tests were DLC (or PLC) tests. You can look for Raskin's publications. Plus I think there is some info on these in Matte's book.

I am of the opinion that widespread use of the DLC might cause serious changes to the PLC test's useability. The DLC requires clearing telegraphing the control question. It also requires examiners re-think their over-reliance on FFF theoretical foundations, as the FFF response to DLC doesn't hold up. Other psychological and physiological explainations do a better job explaining why the DLC works, and can simultaneously explain PLC reactions. In science, on principle is parsimony - the ability to explain the widest range of phenomena with the simplest possible explaination. FFF cannot explain DLC, and does not therefore meet the requirement for parsimony.

Another complication with DLC is that telegraphing the control is an invitation to countermeasure.

I believe that use of a peripheral activity sensor (not just motion or movement - any activity in the perepheral nervous system) is critical with the DLC. The peripheral activity sensor tells us through data (not opinion) whether the data from the traditional component sensors is authentic autonomic nervous system data, or autonomic data that is compromised by peripheral nervous system activity.

I am aware of one examiner who had an very unfortunate false negative on a DLC test - conducted using ineffective DLCs (directed lie to neutral questions) and no peripheral activity sensor. Effective DLCs still require complexity and personal/ethical compromise. I am not at all convinced the false negative was due to the DLC.

r

------------------
"Gentlemen, you can't fight in here, this is the war room."
--(Dr. Strangelove, 1964)

posted 06-22-2006 08:55 AM            

National Polygraph Consultants, LLC

To print this page, click 'file' and then 'print'.
Information on ordering these items follows the list of titles.

Articles selected from search:
Title: A Critical Analysis of Matte's Analysis of the Directed Lie, FirstAuthor: Charles R. Honts, SecondAuthor: Anne Gordon, Date: 1998, Volume: 27, Number: 4, PageNumber: 241-252, Keywords: directed lie/DLC/criticism,

Title: A critical analysis of Matte's analysis of the directed lie, FirstAuthor: Honts, Charles, SecondAuthor: Gordon, Anne, Date: 1998, Volume: 27, Number: 4, PageNumber: 241-252, Keywords: directed lie techniques, validity,

Title: An analysis of the psychodynamics of the directed lie control question in the control question technique, FirstAuthor: Matte, James A., Date: 1998, Volume: 27, Number: 1, PageNumber: 56-67, Keywords: comparison question test, CQT, directed lie, PDD, validity,

Title: An Analysis of the Psychodynamics' of the Directed Lie Control Question in the Control Question Technique, FirstAuthor: James Allan Mattee, Ph.D., Date: 1998, Volume: 27, Number: 1, PageNumber: 56-67, Keywords: DLC/criticism,

Title: The Directed Lie Control Question, FirstAuthor: Stan Abrams Ph.D., Date: 1990, Volume: 19, Number: 4, PageNumber: 26-31, Keywords: techniques-DLCQ/ control questions,

Title: The hybrid directed-lie test, the overemphasized comparison question, chimeras and other inventions: A rejoinder to Abrams (1999), FirstAuthor: Honts, Charles R., SecondAuthor: Raskin, David C., ThirdAuthor: Amato, Susan L., FourthAuthor: Gordon, Anne, FifthAuthor: Devitt, Mary, Date: 2000, Volume: 29, Number: 2, PageNumber: 156-168, Keywords: directed lie techniques, validity,

Title: The Hybrid Directed-Lie Test, The Overemphasized Comparison Question, Chimeras and Other Inventions: A Rejoinder to Abrams (1999), FirstAuthor: Charles R. Honts, SecondAuthor: David C. Raskin, ThirdAuthor: Susan L. Amato, FourthAuthor: Anne Gordon, FifthAuthor: Mary Devitt, Date: 2000, Volume: 29, Number: 2, PageNumber: 156-168, Keywords: directed lie/DLC/criticism,

Title: Validation of potential response elements in the directed lie control question, FirstAuthor: Matte, James A., SecondAuthor: Reuss, Ronald M., Date: 1999, Volume: 28, Number: 2, PageNumber: 124-142, Keywords: directed lie techniques, DLCQ, control question tests,

Title: Validation of Potential Response Elements in the Directed-Lie Control Question, FirstAuthor: James A. Matte, SecondAuthor: Ronald M. Reuss, Date: 1999, Volume: 28, Number: 2, PageNumber: 124-142, Keywords: directed lie/DLC/criticism,

--------------------------------------------------------------------------------
To order articles from the journal Polygraph, contact the National Office of the American Polygraph Association, PO Box 8037, Chattanooga, TN 37414-0037. You can also contact the National Office by e-mail at apabennett@aol.com, or toll free at 1-800-APA-8037. These citations are provided as a courtesy of the National Polygraph Consultants, which is not affiliated with the APA.

posted 06-22-2006 10:44 AM            

The info in Matte's book is wrong on the Utah tests (according to Hontss and Raskin's writings and personal communications I've had with them).

You can use DLCQs on any Utah test. If you want to know how to run one, read Hont's and Raskin's chapter in Kleiner. I have it in PDF somewhere if anybody wants it.

What I don't like about them is they don't "feel" right. How's that for not the least bit scientific? In other words, you can't know a CQ is set until the guy passes the test, unlike the verbal and non-verbal cues you get when you use a well-developed PLCQ.

As Ray pointed out, the TES has the best numbers when it comes to a screening exam, and it uses DLCQs.

posted 06-22-2006 04:12 PM            

The only setting in which I might be very hesitant to use DLC's would be the crim enviroment. Why? Well, we all must agree that overcoming the relevant issue can be rather difficult, ever for an innocent examinee, so how does a traffic violation match-up to committing any number of horrible crimes.

That leave us with the screening enviroment, which is where the majority of examiners utilize the DLC. In my opinion, if one must or feels the need to use DLC's, this is certainly the setting.

posted 06-22-2006 04:45 PM         

The examinee was prepared during pretest to recognize that polygraph worked, that as long as one was truthful, there would be no (consistent, significant) response during testing. When properly vented during pretest question discussion and review, an examinee was expected to display little to no consistent physiological response to the RQs during testing. The pretest was a little longer and more detailed than what most examiners experience today primarily because of their reliance on and presentation to the examinee of ‘cognitive dissonance’ as the reason for responsiveness during testing vice the FFF thought by so many today to be the reason polygraph works.

In this early DLC environment, the examiners were observing the RQs much more than the CQs and as long as there appeared to be ‘normal’ responsiveness at the CQs (and no exaggeration indicative of deliberate manipulations), they were confident that the process was working. When RQs were consistent, timely, and significantly more responsive than CQs, the posttest began. When the opposite was true, the examinee was deemed to be NDI. This is a simple explanation of the technique and one that has not been well understood by many who implemented testing and use following Lou Fuse’s paper in 1982.

Most of the polygraph world was used to using PLCs so when they began to experiment with and employ DLCs, the assumption was that they needed to be important, significant, or emotion evoking. All of which is untrue. The examinee needs to be aware that you’re paying attention to his/her answers and they must pay attention to avoid mis-answering. They should also be asked to have something specific to which they are lying. They should not be asked to play mental gymnastics in their mind with the question for a period of time before answering. These concepts are misguided and dangerous in the hands of an inexperienced examiner who might want to accept all resulting tracings as legitimate physiological response.

Research has demonstrated that ‘personal involvement’ in the DLC in question provides better results than DLCs which have no personal involvement (“Did you ever violate a traffic law?” vs having a person lie to “Are the lights on?”). Interestingly, the folks who devised this technique did just that but they didn’t have the benefit of research.

The argument over which is better, DLCs vs. PLCs, may go on for some time but I think it’s clear both work well and perhaps even have their own arenas where one may provide some benefit over the other (screening vs. specific issue/criminal testing). As Barry noted, some of the recent research by Kircher, et al, indicated a slight edge to the PLC but that was in specific issue testing (I believe). In other projects, particularly several of the screening projects, the DLC modules outperformed the PLC modules (note the TES studies mentioned above and the studies cited within those reports).

One last comment for tonight (I’m getting tired), interestingly, while I was taught that DLCs should not be used in issue testing (for ‘balance’, etc., in situations where there was already some emotion evoking issue and perhaps interrogation…PLCs were preferred where I grew up), the studies done by Honts, Kircher, and others usually involve specific issue situations, yet they usually reported highly accurate results using DLCs.

Barry mentioned, “they just don’t feel right.” For most who do specific issue testing of folks who are suspected, accused, have been interrogated, the thought of using a CQ which is not “set” is difficult to swallow. And in my opinion, in that situation, a DLC ought not to be used.

I don’t think DLCs are the panacea for all polygraph testing but I do think they bring a valuable tool to non-suspect screening which, when conservatively evaluated, lends itself to highly accurate results and the ability to posttest, develop information, and resolve the examination with a high degree of confidence.

Now, I’ve done it. I mentioned conservative evaluation. DLCs were never intended (by the original designers) to be evaluated in the same manner as PLCs. Their view of pneumos was critical and primarily to gauge Cm attempts. Research by Horowitz et al. and again more recently by Kircher et al. support this view. Both reported that different rules are necessary for evaluating DLCs. Perhaps a lack of understanding of this need for conservative evaluation can be blamed for the perceived problems with FPs when using DLCs.

Recall Don Krapohl’s ‘successive hurdles’ approach and use DLCs as a very conservative measure for initial series screening. This will (in my opinion) lend itself to a politically acceptable, yet very sensitive to deception screening tool for the initial series. Problems leading to interrogation and information development can always be resolved later by break-out testing using PLCs to balance the test for those who have problems passing, have provided information, and are ready for further confirmatory testing to resolve the problematic issue(s).

This is just one guy’s opinion most of which was included in a paper in the Journal a few years back to set the record straight and give credit to the guys who devised this technique in the 1960s.

Pardon the typos…the old eyes and fingers are failing. Boy what a rambling piece. I’ll keep quiet for a while.
Regards,

posted 06-23-2006 07:40 AM            

I would be very interested for someone who actually uses DLC to provide a brief script for introduction of a DLC, to help me get a grasp on using this technique.

My mind is still open, but there may not be much room left for new stuff so hurry

posted 06-23-2006 10:52 AM            

You start with a numbers test (known), and you tell the person you know what it looks like when he lies, and that he responds well when he does. You tell him you will use that reaction to compare with the RQs and the CQs on the test (not for greater size even though that is what you'll be doing). You explain that the two question types should look different because he's lying to the DLCQs and - so he says - truthful to the RQs. If the question types look different, and the DLCQs look like the numbers test lie (because you need to know he's maintaining an ability to respond...), then you'll know he's truthful. If, however, both question types look the same, then you'll know they're all lie responses.

That creates a fear in the truthful, as they are now motivated to (and fearful of not) respond(ing) well to the DLCQs as they have to in order to convince you of their truthfulness. The liar on the other hand, doesn't want to cooperate (by following your instructions on how to deal with the DLCQs, which I won't discuss now). He is still more concerned about the RQs.

I hope that makes some sense. In short, the truthful fear the DLCQs because if they don't react to them, they won't pass the test, something they are highly motivated to do. (I say "fear," but fear really isn't necessary in a polygraph test, but that's a different topic we've discussed already.)

Scoring is still the same for a PLCQ and a DLCQ. Use the Utah rules. Breathing is different, but they haven't yet validated a different set of rules for their evaluation. You will get some crazy pneumos on a DLCQ. If you looked at them as you would on a PLCQ, you'd suspect CMs, and that is one of the drawbacks.

posted 06-23-2006 01:57 PM            

Well said.

r

------------------
"Gentlemen, you can't fight in here, this is the war room."
--(Dr. Strangelove, 1964)

posted 06-24-2006 09:41 AM         

However, for historical reasons, I still have to point out there are some subtle differences between what is done by many today (because it's now been 'researched' using the method of treating DLCs like PLCs; having examinees 'fear' them; or examiners believing they must make the DLCs important and even emotional) and what was done by those who used them initially. Their method was not well understood when introduced by Fuse, primarily because there was little to no exposure to the concept (as they used it) and when the researchers began testing the idea, they truly did not comprehend the sophistication used by Fuse and his mentors (Hitchcok & Dubiel).

The early DLC method was to convince an examine (remember: a non-suspect subject looking for job/clearance/or continued access andnot someone who had been interrogated re: a specific crime) that he/she only had to be truthful and cooperative to pass the test. Part of the conditioning took place with the ACQT/Stim test, which in their day, was an unknown number test done in a manner which made it very difficult to miss (see Barland, 1981 & Fuse, 1982). The functions of the DLCs were related to the response activity seen on the ACQT when explaining the questions, to insure the subject recognized the importance of the DLC.

Comments such as, "you need to pay attention to avoid misanswering during testing..." when introducing the DLCs, were sufficient to emphasize the importance without requiring the examiner to push the issue of comparison or the need to generate fear in the subject about not passing the test. In fact the entire pretest was done to convince the innocent person that as long as he/she was truthful, they would pass the test. Hence the DLC introduction made clear to a properly prepared subject that they had nothing to fear because they knew the examiner would be able to see where they were lying during testing (and in their case, it would be to the DLCs).

Their methods were admittedly more conservative because they were using it in screening and the technique was extremely unintrusive. They had no problem questioning someone respoding to a relevant question and retesting to confirm information developed. Therefore, if there was any significant response to RQs, they questioned. Later on, a numerical analysis was added, I was told, "to help their new examiners learn to evaluate DLCs."

While conditioning is important in any pretest, it may be even moreso in the DLC pretest, as originally designed. In this test, there is no emotion-evoking PLC on which to rely to balance things if other parts of the pretest are week. Somehow this philosophy and concept never was emphasized when the DLCs began their entry into the broader community and it is one that is not easily embraced. Especially by examiners who have been used to using PLCs which require the laying of foundations, setting, and some emotion evoking value. The idea is a little different.

Actually some of the subtle differences lie at the bottom of the discussion between Honts and Abrams several years back in the APA's Journals and Newsletters. One can get into trouble if DLCs are "pushed" and you don't recognize the distortions which are frequently generated by some of the instructions provided to an examinee during DLC introduction.

I pointed this out in the 2004 paper because of some of the concerns voiced across the community about FNs and in some quarters, FPs, when using DLCs. This is more for historial concern rather than trying to point out there is only one way to do it.

I've seen DLCs used in a variety of ways, I've noted the research explanations, and have talked with the researchers and those who originally used the "weak controls," as they called them. Some of these 'old guys' complained to me that they felt too many examiners use DLCs today because they are so strong and are less difficult to use (perhaps mistaken thinking). Given how they used the DLCs, I understand their thinking (but am less apt to lean to this judgement since I am aware of the methods espoused in the literature since Fuse's paper). That's how today's examiners have been taught.

Their method worked for them. It was their usage of DLCs noted in Barland's 1981 study (hope I'm right on that one!) and the 1989 Barland, Honts, Barger study that awakened the interest of researchers. What has become standard practice in some quarters is not what was practiced by the original designers. And no, I'm not advocating a return to use of the unknown numbers test (and yes, I know that test had some value...).

Hopefully, this isn't too disjointed. I'm just trying to add some background and clarity, not preach about the only way to do things. This is a technique that was never really understood as originally used. For whatever that's worth.

Again, I'll apologize for typos and rambling but thank you for the forum!

PMM

posted 06-25-2006 11:33 AM            

You're exactly right, but Charles is going to say he wrote exactly how to conduct such a test in Kleiner, which he to me I could feel free to distribute to anybody who wants it. Just email me, and I'll send it to whoever wants it. (The stimming between charts takes place on a Utah PLCQ test too. It's just done a little differently, but that's another topic.)

posted 06-26-2006 05:04 AM            

Score the test using the Utah scoring criteria as it has been validated on both DLCQ and PLCQ tests.

Remember that the set-up of the test is crucial, so be sure your pre-test is correct, which, again, you'll find in the chapter in Kleiner.