The empirical research on countermeasures has not provided enough information to determine whether specific countermeasures have the specific physiological effects that would lead a polygraph examiner to judge an examinee as nondeceptive. Consequently, it is difficult to determine why specific countermeasure strategies might or might not work. We would not expect specific countermeasures (e.g., biting one’s tongue) to have uniform effects on all of the chart readings obtained during a polygraph test, and studies that focus exclusively on the effects of countermeasures on accuracy do not allow one to determine why specific approaches might work or fail to work in different contexts.



Studies of the effects of countermeasures on the outcomes of polygraph examinations have yielded mixed outcomes. Studies on the effects of drugs, already discussed, are a good example. An early study by Waid et al. (1981) suggested that the use of the drug meprobamate reduced the accuracy of polygraph examinations, but subsequent studies (Iacono, Boisvenu, and Fleming, 1984; Iacono et al., 1992) suggest that similar drugs, such as diazepam (Valium) and methlyphenidate (Ritalin), have little effect on the outcomes of polygraph examinations.

It is difficult to draw firm conclusions from research on the effects of drugs and alcohol on polygraph examinations for two reasons: there are relatively few studies that provide data, and these studies share a central weakness that is endemic in most of the polygraph research we have reviewed—a failure to articulate and test specific theories or hypotheses about how and why drugs might influence polygraph outcomes. These studies have rarely stated or tested predictions about the effects of specific classes of drugs on specific physiological readings obtained using the polygraph, on the examiner’s interpretations of those readings, or of other behaviors observed during a polygraph examination. Different classes of drugs are likely to affect different physiological responses, and the effects of one class of drugs (e.g., benzodiazepines used to treat anxiety) might be qualitatively different from the effects of alcohol or some other drug. Research on drug and alcohol effects has not yet examined the processes by which these substances might influence polygraph outcomes, making it difficult to interpret any studies showing that particular drug-based countermeasures either work or fail to work.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement