Mr. Webb.  Please allow me to thank-you for your previous time, but your input is no longer needed or requested.  I was able to obtain a copy of the aforementioned study, and it is much easier to read and understand for a layman such as myself.

What I gleaned from my reading the original study is the following:

40 subjects were given a polygraph, 20 each who were actually innocent of the faux crime of stealing some movie passes, and 20 who were actually guilty of the same faux crime.  Of the 20 "innocent" subjects, the polygraph results actually showed that 7 of those 20 were accused of being guilty.  Of the 20 subjects who were actually guilty of the faux crime, the polygraph found 16 of the 20 subjects guilty.

So, according to Mr. Hont's study, a polygraph found 7 innocent people guilty, and 4 guilty people innocent, resulting in an error rate of 27.5%.

My comparison study...

It has been said that the results of a polygraph are no better than a flip of a coin.  To determine if this hypothesis is correct, I took the last 5 minutes and flipped a quarter 40 times.  The results of my 40 flips were 22 heads and 18 tails.  I guess the polygraph is in fact more accurate than a flip of a coin, but certainly not to the accuracy quoted by many.

Any thoughts?  

1904:

Your suggestion seems pretty valid to  me.  But, before countermeasures are studied in an unbiased manner, perhaps a study, (or a series of studies) actually attempting to scientifically measure the ability of the the polygraph to actually detect deception would be in order.

I find it amusing that the examiners who have now re-registered under their real names after being outed refuse to partake in my discussion regarding Hont's study showing the polygraph to be less than 75% accurate.  Perhaps I am not worthy of such discussion, but failure to logically counter this claim (that the study purports a poor performance regarding accuracy of the polygraph) indicates that the claim is valid.  One can tell who is reading the forums by looking at the bottom of forum home page, and I have seen several of the names as active on the forum.  In fact, as I write this, myself, 1904 and EJohnson are on line.

1904,

There's no raging debate among scientists about polygraphic lie detection. There is broad consensus that it is has no scientific basis. As Gino has noted before, the only remaining debate pits those who understand that polygraphy is junk science against those who don't (yet) and those with a vested interest in the perpetuation of polygraphy. See Chapter 1 of The Lie Behind the Lie Detector and the sources cited there for a thorough debunking of polygraphy.

EJohnson wrote on Oct 24^th, 2007 at 2:26pm:


Mr. Johnson:

Are you going to address my questions/claims about the accuracy of the polygraph as it was revealed in Mr. Hont's study, or just argue with George?

Mr. Johnson,

The National Academy of Sciences' Committee to Review the Scientific Evidence on the Polygraph did not endorse the validity of polygraphy. On the contrary, the panel reached a very different conclusion (at pp. 212-13 of its report, emphasis in the original):

Quote:


That's about  as strong a condemnation as is possible in a consensus scientific report. Speaking more plainly, the Commitee's chairman, Professor Stephen Feinberg stated, "Polygraph testing has been the gold standard, but it's obviously fool's gold."

I note that Charles Honts, a co-author of the article that is the topic of this discussion thread, was discredited by a federal judge after he attempted to characterize the NAS report as endorsing the validity of polygraphy.

EJohnson wrote on Oct 24^th, 2007 at 2:56pm:


Addressing the last two lines, when I first logged on to this site, I read the lead paragraph about using an alias, which I took to be good advise.  As far as any personal information, forget it.  Any personal information about me opens me up to personal attacks, which I am not interested in fostering, nor participating in.  I think is fairly disingenious of you to suggest that because I use an alias, that fact somehow puts me into the "troll" catagory, given your past history.   

Moving on...

As I stated in a previous post, scientific studies can expose results which are ancilary to the original purpose of the study, but are still significant in their own right.  Given the study in question, it appears that this is the case.  I notice you do not challenge my assertion that the study indicates the accuracy of the polygraph (in this test anyway) is no better than 73%.   I will take that lack of challenge as agreement to this assertion.

Given that fact then, with an error rate of over 25%, how can any honest reviewer conclude that the study proves anything regarding the effectiveness of countermeasures, when the polygraph itself could not detect blatant deception?  Add in the variable of self-reporting, ("uh, yah, I used some of those things I read about in the lie book, can I have my movie passes now?") while it sounds pretty impressive, it really is just a good example of the type of junk-science that seems to be passed off for true research in this field.

George W Maschke: Quote:


I would respectfully submit that you have been in violation of your own posting policies for some time when you make this assertion.

There is no such broad consensus. The presence of a 400 page review of the science of polygraph speaks to the complexity of the scientific basis for the test. There is in fact a scientific basis for polygraph test, in terms of a number of important psychological theories and constructs, known physiology, and the basic principles of inferential and bayesian predictions in the field of testing, diagnostics, signal detection, and decision theory.

Whether you like that fact, are satisfied with the present answers or even disagree that we have a fully developed or a mature understanding of the scientific basis for polygraph is another matter. There are a number of scientific theories underlying the polygraph test. 

I believe you are deliberately providing inaccurate information to your readers here, and that appears to be a violation of the posting policy.

Quote:


Would you please correct the errors of your previous postings, so that you are no longer in violation of your posting policies.

Thank you.


r