The Australian Broadcasting Corporation (ABC) television program Catalyst examined lie detection in an episode titled, “To Catch a Liar”:
How good are you at spotting a lie? We’re lied to up to 100 times a day, yet research shows fewer than one in a 1000 people can reliably detect lies. But science has declared a new war on deception. In this special half hour report, Catalyst crosses the world to put the latest in lie detection technology to the test. We’ll see an actual lie forming in the brain, visit the US Department of Defence’s “Booth of Truth”, learn the truths and myths of spotting lies, find out when lying is good for you, and why humans have evolved to be so vulnerable to liars. Along the way, Catalyst reporter Dr Jonica Newby exposes one of her deepest secrets to the lie detectors – “ would you answer questions about your integrity, plastic surgery, even fidelity on national television? The stakes are high, or the tests won’t work. This week on Catalyst – can science catch a liar?
A transcript of the show is available at the above-linked page, and it seems that the video will eventually appear on-line, too.
catalyst
SOME POINTS ARISING FROM WATCHING THIS 25-MINUTE AUSTRALIAN SHOW
In general it was as informative as I’ve seen on the issue of detecting lies by psychophysiological means, but of course there are points of ambiguity that are potentially relevant to the interests of this list.
I’ve numbered the points below to facilitate further discussion, which is probably best done if replies specify to which of the points they are addressed.
All the best, John
The points:
1. The blood flow to the face by DODPI employs the pseudo-scientific CQT, rather than the science-based GKT (for distinction see http://www.psych.utoronto.ca/~furedy/polygraph.htm). However, the most important practical point for antipolygraph is that none of the countermeasure that involve physiological augmentation of responding to the “control†questions in the CQT will work, as this sort of augmentation control is not possible with facial blood flow. Nevertheless, the psychological countermeasures (e.g., thinking exciting thoughts or getting angry during the “control†questions) may work, but this is very much an empirical, and as yet unresearched question.
2. The placement of DODPI on an equal scientific footing with the other three genuinely scientific labs is bad news, especially as the facial blood volume (FBV), if it were based on sound methodology (which the CQT is not), has the potential of being applied without the examinee’s knowledge. In terms of potential dependent variables, DODPI appears to be a scientific leader. The reasons for this perceived prominence include the failure to discriminate between CQT and GKT methods, and, more generally, the failure to understand that technological developments in higher-tech dependent variables do not result in any significant real advances in detection, unless the independent variables are properly manipulated in a controlled, standardized way. So it doesn’t matter if a CQT uses the highest. tech dependent variables, as long as is not standardized, and the relevant and “control†question do not differ only on the dimension of guilt. Given that even the blue ribbon NAS committee has not understood the difference between CQT and GKT methods of detecting guilt, the prospects for North American society continue to be poor.
3. Even though the FBV measure used in the very well equipped DODPI lab is too expensive to be used in the many field polygraphs that are given in North America, I can see a day where cases that have attracted media attention, and are therefore important, are “verified†in the super-duper DODPI FBV facility. This will lend considerable surface validity to what still remains an entrails reading procedure for detecting guilt.
4. While, at the end of the show a couple of GKT methods were mentioned, that terminology was not used, and it was also not made clear that of all the procedures they did look at, they were based on CQT methodology. It is interesting to note that while the other three labs all had principal investigators with good scientific credentials, none has expertise in experimental psychophysiology, that branch of psychology that uses physiological measures to study and differentiate psychological functions.
5. In discussing the Mallard case, the show did say that the polygraph was ruled as inadmissible because it was unreliable (see http://www.psych.utoronto.ca/~furedy/Papers/ld/Mallardverdictpol.doc), but then went on to assert that since Mallard has now been acquitted (on the basis of new evidence and also some misbehavior on the part of the police), this suggests that, after all Van Apperen and his polygraph-based claim of innocence was right. Of course the two issues are quite separate, but especially as the media have been on Mallard’s side during the last few years, I think most listeners will gain the impression that the Mallard case has confirmed the validity of the polygraph. Buttressing this conclusion is that Van Apperen himself cuts a good image on TV, where the fact that in logical terms he clearly lost (see judge’s summing up in the Mallard case I cited above). Doesn’t count. I see him as Mr. Australian polygrapher, whenever the media want to consult an “expertâ€.
6. The show did end properly with the admission that the way in which it tested validity was not a scientifically controlled study. Nevertheless, the take home message to most viewers was that as technology improves for dependent measures, the (CQT) polygraph will become more and more accurate, although all admitted that it will probably never reach 100% accuracy. But then, what test reaches 100% accuracy?
7. Finally, the basic problem that the CQT polygraph does not differentiate between the anxious innocent and anxious guilty was not addressed, and still remains as the most obvious flaw in this procedure/ Note, by the way, that the GKT is not subject to this flaw.