pailryder wrote on Sep 25^th, 2014 at 1:36pm:


The point is that deception in humans is not correlated in any systematic way with any of the physiological indices recorded by the polygraph.

Doug and George echo an inconvenient truth that the polygraph industry, generally speaking, seems content to minimize or gloss over entirely.

The 2003 NAS study, The Polygraph and Lie Detection, was quite clear, saying:

...there is evidence suggesting that truthful members of socially stigmatized groups and truthful examinees who are believed to be guilty or believed to have a high likelihood of being guilty may show emotional and physiological responses in polygraph test situations that mimic the responses that are expected of deceptive individuals.

Given all of the variables in any given polygraph exam -- examiner skills, technique used, test subject personality and suitability, environmental factors, etc. -- the real-life error rate of polygraph testing is, in my opinion, significantly higher than typically represented by polygraph proponents.

1st4th5thand6th, in my opinion, it is because polygraph is first and foremost an INDU$TRY. 

Polygraph indu$trialists -- whether they're on the manufacturing side, in the polygraph school business, run a private practice, or are part of a government subsidized polygraph fiefdom -- certainly aren't about to kill the goose that lays the golden eggs, at least in my estimation. 

If the science is suspect, well...caveat emptor.

By the way, when it comes to the government's use of polygraph, it seems to me that utility trumps scientific validity, and for good reason: Polygraph utility absolutely "works." 

But it comes at a cost.

Collateral damage -- as typically manifested in false-positive results -- is simply a necessary by-product.

Again, this is just one polygrapher's opinion. 

1st4th5thand6th wrote on Sep 26^th, 2014 at 2:18pm:


Go get em' 1st4th5thand6th!   

Dan Mangan wrote on Sep 26^th, 2014 at 3:05pm:


Dan,

I would take issue with any characterization of the NAS report as endorsing the accuracy of polygraphy. The passage you paraphrased is from p. 214 of the NAS report. Here it is in its full context:

Quote:


Note the caveat, which I've highlighted above. Moreover, polygraphy hasn't been shown through peer-reviewed research to reliably perform at better-than-chance levels under field conditions (that is, in the real world, as opposed to the laboratory).

Indeed, a statistical analysis by Dr. Alan P. Zelicoff of the best available field studies suggests that "if a subject fails a polygraph, the probability that she is, in fact, being deceptive is little more than chance alone; that is, one could flip a coin and get virtually the same result for a positive test based on the published data."

Quote:
 

Sir if it's 55% accurate...you're a hair above a coin toss... you want this country to spend millions for 55%???   we can go flip a quarter for free.... 

Quote:
 

Dan,, what do you mean no manipulation?  You turn the box on don't you?  
and you know the box is nothing more than a prop.  So how can you say you're not manipulating them?    

Further, you're not going to turn that box on to anyone who knows that it's a farce... what would be the point?.  you'll hook him/her up... start asking questions, the polygraph wiggles you might say (hypothetically of course): "Ms so-and-so you've had a reaction to this 'crime' question - you wanna explain that?  Ms so-and-so can look at you and say... Sir, "there is no systemic way to map my physiological reactions to any particular emotion". "your wiggles mean nothing and you know it".    What do you say then?  I'll bet most will claim the person is lying or accuse them of hiding something and go on the verbal, psychological assault under that assumption..  For what?  The person made a truthful statement  because she knows your conning them... the whole thing falls apart.... 

Also, how could anyone who knows the con...even believe a polygrapher who says something to the effect of  "you're reacting to this question more than the others?"  

Does the polygrapher stop and show the different charts to the person?
I'll bet the answer  to that is NO!

But Even if they did, so what?... the chart means nothing because it too is just a prop an attempt to get the candidate to believe something that is not TRUE... 

This scenario and any variation of it is TOTAL MANIPULATION.... 
How can you possibly say otherwise?

So by virtue of turning on the box you are manipulating them.

Further, by deliberately refusing to test anyone who knows the truth about the polygraph you taint the sample set and thus the result set.  

Upon "finding out" that the person has read the books, visited the sites. you automatically conclude that these are dishonest, cheaters..and most likely act vigorously on that false assumption.  That person will end up in the reject bin....and your organization will claim if found a cheater!!!  
 
So if you even hook them up you're automatically biased.... . if she breathes, you suspect CMs.  Face it,  Polygraphers are totally biased in giving this test - which automatically renders it a manipulated sample set


There is no legitimate 3rd party "organization". or accounting group,  or organization that performs statistical quality analysis for anything that will bless a result from a biased manipulated  and then further pass it off as expert analysis.  

with the sole EXCEPTION being your profession...

1st4th5thand6th, I deliberately chose the 55% figure as a minimum threshold example of "more likely than not."  I agree that it's not much better than a coin flip, but, in the extreme, it has value for gross culling and screening purposes.

True, the box is often used as a prop, but not so, generally speaking, in specific-issue testing. The box measures and records physiological feedback from the test taker. Patterns are analyzed and a result is usually rendered.

I do not manipulate clients, but many examiners -- perhaps most -- do exactly that. And while I would never stop mid-stream to show a test taker the differences in their reactions vis-a-vis RQs and CQs, I often do just that after the test.

My testing model is "open book." Days prior to the test, I advise the client of the risks, realities and limitations of the test. My web site has a page designated "Recommended Reading," which provides a broad spectrum of opinions (pro and con) about polygraph. Yes, it contains a link to this very site.

Why do I do that? Simple. By thusly informing the potential client, it is the consumer who decides if there's what is known as a "value proposition" -- in business terms -- to be derived from their participation in the polygraph process. I take myself out of that loop.

I would be happy to demonstrate the machinations of polygraph testing to you at no charge. Should you find yourself in the New England area sometime, call me and we'll set up an appointment. You can participate in a mock crime exercise, after which you would be polygraphed on a particular aspect of that crime.

Based on my experience, if you actually committed the mock crime, there is about a 95% chance you would be detected. If you did not do the deed (such as taking money), there is a roughly 85% chance you will be found truthful.

Again, there is no manipulation in my tests. I've been doing open-book polygraph testing for a while now. Again, I tell the client the "truth about the test" (your phrase) in advance of the appointment and encourage them to explore the links found on the Recommended Reading page of my web site. Immediately prior to the test, I review CQT theory  and even identify the RQs and the CQs.

While such an approach makes many of my polygraph colleagues almost apoplectic, providing all of that allegedly sensitive information to the examinee doesn't seem to matter much. 

The guilty are usually found deceptive and the innocent are usually found truthful. 

All of that said, you are absolutely right about the lack of any central controlling authority. Unfortunately, polygraph abuse and victimization occurs far too often. Therefore, the key to consumer protection is consumer education.

That's why I have proposed a "bill of rights" for polygraph test subjects. Such a measure would sharply curtail the abuses that plague the polygraph industry. The polygraph establishment, however, wants little to do with a bill of rights for the person who's in the hot seat.  In my opinion, that's wholly unethical, and another glaring blemish on an industry with a very sketchy reputation.