Can activity sensors reliably detect the use of physical countermeasures?

I would like to hear your thoughts on the following information taken from the John E. Reid: Investigator Tips website:

<sup>An important part of an examiner's training is to identify those subjects who are at risk for producing erroneous results. The ideal subject is one of average, or above average intelligence, who is not mentally impaired and is in reasonably good medical health. It is also important that the subject's emotional state is relatively stable during the examination. A leading cause for false positive results (reporting a truthful subject as deceptive) is anger within the subject. This anger may be caused by prior interrogation or an accusatory pre-test interview. Inadequate development or selection of control questions is another cause for false positive results. 

False negative results (reporting a deceptive subject as innocent) may be the result of improper question formulation in which relevant questions are phrased in such a way that they do not present a significant threat to the guilty person. Subjects with a lower intelligence may also increase the risk of a false negative result. In some cases, a guilty subject may engage in countermeasures that the examiner fails to recognize during chart analysis. Identifying countermeasures is a critical part of a competent examiner's training. In our office almost 25% of guilty suspects are identified primarily because of their use of countermeasures. Innocent subjects allow their body to respond normally during a polygraph examination. It is the guilty subject who will consciously try to manipulate the polygraph recordings. In this regard, agencies should be very cautious when using an examiner who relies exclusively on computer analyzed charts. None of the software developed for this purpose is capable of identifying specific subject countermeasures.</sup>

http://www.reid.com/Investigatortips.html?serial=321090728120738

First, even John E. Reid & Associates' ideal subject who is "of average, or above average intelligence, who is not mentally impaired and is in reasonably good medical health" is "at risk for producing erroneous results." Polygraphy has not been proven by peer-reviewed scientific research to work at better than chance levels under field conditions.

The late John E. Reid is generally credited with the development of the probable-lie "Control" Question "Test." Reid co-authored a book about polygraphy with Fred E. Inbau titled, Truth and Deception: The Polygraph ("Lie-Detector") Technique, which is one of the main reference books on polygraphy. But Reid & Inbau devote relatively little space to what they term "evasive tactics," and they don't offer any counter-countermeasures that would detect the kinds of countermeasures described in The Lie Behind the Lie Detector.

They do provide portions of sample charts of polygraph subjects who did such things as breathing slowly and evenly, or who took unusually deep breaths, etc. Some people who do not understand the assumptions based on which polygraphers infer truth vs. deception may believe that the polygraph "test" can be passed by coughing, or flexing one's arm, or wiggling one's toes, or by breathing slowly and evenly. Such manipulations might well be detected by an alert polygrapher, and are unlikely to help a person in passing. But none of the countermeasures described in The Lie Behind the Lie Detector are obvious to the trained eye, whether that eye is observing the subject or poring over a polygraph chart, and there is no evidence in the peer-reviewed scientific literature that polygraphers can detect the use of these countermeasures at better than chance levels of accuracy.

Indeed, in a recent non-peer-reviewed report by Charles R. Honts, Susan L. Amato, and Anne K. Gordon titled "Effects of Spontaneous Countermeasures Used Against the Comparison Question Test" (Polygraph, Vol. 30 (2001), No. 1, pp. 1-9), polygraphers were unable to identify even unsophisticated countermeasures attempts at better than chance levels:

Quote:


In Honts et al.'s laboratory study, which involved a mock crime scenario, 89.6% of the programmed guilty subjects and 45.8% of the programmed innocent subjects reported the use of spontaneous countermeasures, that is, although they had received no training in polygraph countermeasures, they did things that they imagined would help them to pass. (We can be confident that the subjects in Honts et al.'s study had not read The Lie Behind the Lie Detector, as the study was conducted before the 18 September 2000 release of our book.)

Honts et al.'s study suggests that unsophisticated countermeasures attempts are quite common -- even by truthful subjects -- and that polygraphers cannot detect them at better than chance levels. The countermeasures described in The Lie Behind the Lie Detector would arguably be much harder to detect than unsophisticated countermeasures attempts, as they are quite subtle, replicate the reactions deemed by DoDPI doctrine to be significant, and those applying them will be well aware of the distinction between "control" (also known as "comparison") questions, relevant questions, and irrelevant questions and will know to augment their reactions only to the "control" questions.

If John E. Reid & Associates believe they can reliably detect countermeasures, then perhaps they'll accept Dr. Richardson's Polygraph Countermeasure Challenge.