I'm about to take a re-test for a pre-employment polygraph. After failing once already despite telling the complete truth, I was so depressed and miserable.   I have a question about the format of the polygraph used for the agency that is mentioned in the following post:

http://antipolygraph.org/cgi-bin/forums/YaBB.pl?board=Policy&action=display&num=...

It looks like the format used is R/I, until I saw this post:

http://www.antipolygraph.org/cgi-bin/forums/YaBB.pl/YaBB.pl?board=Proc&action=di...

which makes the polygraph test I took for that agency to look more like the GQT test because they asked me "Do you intend to lie to any of the questions on this test?"   

Now I also got Doug William's online book about the polygraph.  In his manual, it states those questions are most likely relevant questions, so do not show any reactions to that question.  But if the test is a GQT, then I actually should react to that disguised control question in order to "pass", right?  Can someone please comment whether the test format for that agency is most likely R/I or a GQT.  Thank you very much.   

psyduck,

According to the Defense Security Service, DoDPI last taught the GQT format c. 1997. I do not know whether the format was actually discontinued altogether or if it might have simply been re-named (perhaps after slight modification).

In any event, in the GQT format, the question "Do you intend to lie to any of the questions on this test?" serves as a disguised "control" question, and if one were to encounter this format, one would want to augment one's reactions to it.

However, in a probable- or directed-lie "control" question "test" (CQT), this same question would commonly serve as a "sacrifice relevant" question and would not be scored. Presumably, there would be no harm in augmenting one's physiological reactions to it. The sacrifice relevant question usually appears as the first relevant question in a series, following two irrelevant questions.

If this question were asked in the context of a R/I "test" administered by the NSA, it is conceivable that it might be intended to serve as a kind of "control" question against which to compare responses to the more specific relevant questions that are asked.

I agree with BEAR that you should download and carefully read through The Lie Behind the Lie Detector. Chapters 3 & 4 will be of special interest. I think you'll find that our explanation of the various polygraph techniques is more comprehensive than that provided by Doug Williams in "How to Sting the Polygraph." We also provide ample references for further reading.

Now, with regard to whether NSA uses the R/I technique or some variant of the GQT with disguised "control" questions, I don't know for sure.

As you know, the OTA report described it in 1983 as an R/I technique. In the mid-1970s, the then director of the NSA's polygraph program, Raymond J. Weir, Jr., described the NSA's R/I technique in a pair of articles published in the American Polygraph Association quarterly Polygraph. The description of the R/I technique at pp. 100-103 of the 2nd edition of The Lie Behind the Lie Detector is based in large part on Weir's articles.

While I suspect that NSA is probably still using the format described by Weir and mentioned in the OTA report, I have not found documentation of this.

BEAR,

It's conceivable that your polygraph interrogation may have included "disguised control" questions like those used in the GQT format, but I would make no such assumption, and I think it's rather unlikely, perhaps all the more so because your examination was not with a federal agency.

I think it's more likely that you encountered a conventional R/I technique. The question, "Do you plan on lying during this test?" was likely a "sacrifice" relevant question (and was likely the first relevant question in a series). And the question, "Did you lie on any part of your application?" is a very common relevant question in pre-employment polygraph screening. Note that this latter question is not used as a "disguised control" question in the GQT format.