The methodology used in the Relevant/Irrelevant technique is addressed at p. 177 ff. of the 4th edition of The Lie Behind the Lie Detector and a countermeasure approach is discussed at p. 151.

nonombre,

Is this a common test format and which agency uses it more often?

Quote:


Opp,

The RI is utilized in several forms by a growing number of state and local police departments (I'm sorry, I don't have an actual list.) 

In addition, I know for sure that RI is the primary procedure for several of the federal intelligence agencies and is a secondary procedure for virtually ever federal law enforcement agency.

I hope this helps.

Nonombre

nonombre wrote on Aug 4^th, 2005 at 3:16am:

Nonombre,

I am certainly not trying to answer for Drew, but I thought I’d add my opinion…

There is no way of being certain you can “beat” the test for an RI exam, because you can be “failed” for any of a number of reasons totally unrelated to actual deception.  In that way it is just like the other forms of polygraph testing.  I know from personal experience that being completely truthful and not withholding any information does not guarantee that you can “beat” the test.   In fact doing so resulted in a failure rate of 75% for the four polygraphs I had to endure.

To me, this cuts to the heart of the matter regarding polygraphs.  If a person can take a polygraph, tell the complete truth and not withhold anything even remotely relevant, and still fail three out of four times that indicates a fatal flaw in the test itself.

I’m not interested in comparisons with the subjective oral board interviews or the subjective background investigation.  The polygraph is supposed to obtain damaging admissions and/or provide data to conclude if the subject is being truthful or deceptive with regards to the relevant questions.  That is its function, and it is largely ineffective in fulfilling that function.

If the subject does not believe in the myth of the polygraph as a “lie detector” then no damaging admissions will be forthcoming.   

Anything less than 100% accuracy in the truthful vs. deceptive assessment is worthless.  If you want to use an accuracy rate of 75% (which I believe is significantly higher than reality, but I’ll use it for purposes of discussion) then for every hundred subjects who are labeled as deceptive, twenty-five of them were actually truthful.  Which ones?  It’s impossible to tell, and that’s where the polygraph falls short with its less-than-complete-accuracy.

If a truthful subject can be branded as deceptive a significant percentage of the time, and a deceptive subject can be labeled as truthful a significant percentage of the time, then what is the final utility of the polygraph?  Other than as an interrogation intimidator to be used for extracting confessions from unwitting subjects, I can’t see any legitimate use.

Quote:


George,

I have a question regarding the info on page 151. It states that polygraphers--during th the Rel/Irrel test--are looking for significant changes to patterned relevant questions, and  (quoted from the page) "one can prevent such a pattern from occuring by simply producing responses to two differing groups of relevant questions within the different chart presentations." What exactly does this mean? Are you supposed to produce varying a responses to to groups of relevant questions to stop a pattern from forming? If so, will this not alert the examiner? What is the the best CM approach for Rel/Irrel test?

Thanks!  

Nonombre,

Perhaps in my absence and prior to my fully answering your question (there is a general approach for all examinees and a different targeted approach for those who would seek to insure non-deceptive charts while being "guilty" (beat the test) vs. those who are innocent (overcome the inherent error associated with this form of testing) with an RI exam--I will discuss only the latter of the two targeted approaches for obvious reasons), you might address the issues I raised.  You suggested an examinee should freely discuss his concerns with his examiner.  I have stated that “lie detection” is about deception and that the initial (perhaps only) deceiver (the examiner) should not be trusted nor sought out for advice.  I have given several examples of this type of deception (http://antipolygraph.org/cgi-bin/forums/YaBB.pl?board=Proc;action=display;num=99... ).  Would you comment on these areas that I have labeled as misrepresentation, falsification, exaggeration, outright deception, etc.  Would you agree that they are?  If not, why not, and if so, in light of this behavior why would you suggest any examinee should make himself/herself vulnerable to such an examiner through the sharing of problems and confidences?  Regards,

Nonombre,

I believe the notion that the polygrapher is the deceiver is derived from the fact that the test is only partially explained to the examinee.  If the test were to be explained in whole, the examinee would certainly have a much better idea of how the test functions.  As examinees we are told to never hold anything back and provide full explanation and complete truth.  However, if an examinee does so then the test is unable to function as proposed.  Therefore deception on the examiners part is crucial in an attempt to make the examinee believe the test is actually accurate in detecting deception.