ated with deception, or the fear of deception, were involuntary and quite large in comparison to other anxieties aroused by the test (Marston, 1917). Consistent with this line of thinking, theories of the psychophysiological detection of deception by polygraph assume that relevant, in contrast to comparison, questions are more stimulating to those giving deceptive than truthful answers. Interpretation of a polygraph test has typically been based on the relative size of the physiological responses elicited by relevant questions and the associated comparison questions (e.g., Podlesny and Raskin, 1977; Lykken, 1998). If the assumptions about large and involuntary responses to relevant questions are true, the polygraph test would be characterized by high sensitivity and specificity—it would discriminate very accurately between deception and truthfulness—and it would be immune to countermeasures.

Such assumptions are not tenable in light of contemporary research on individual and situational determinants of autonomic responses generally (Lacey, 1967; Coles, Donchin, and Porges, 1986; Cacioppo, Tassinary, and Berntson, 2000a) and on the physiological detection of deception in particular (e.g., Lykken, 2000; Iacono, 2000). There is no unique physiological response that indicates deception (Lykken, 1998). If deceivers in fact have stronger differential responses to relevant questions, it does not necessarily follow that an examinee who shows this response pattern was lying (see Strube, 1990; Cacioppo and Tassinary, 1990a) because differences in people’s anticipation of and responses to the relevant and comparison questions other than differences in truthfulness can also produce differential physiological reactions. For example, relevant questions are sometimes inherently more threatening than comparison questions. Asking a weapons scientist “Have you committed espionage?” might generate a stronger response in some innocent examinees than “Have you ever taken something that did not belong to you?” Also, as noted above, individuals who have experienced punitive outcomes from being wrongly accused in the past or who believe the examiner suspects them of being the culprit may, in theory, be more reactive to relevant than control questions even when responding truthfully. No independent evidence has been reported in mock crime studies to verify that relevant questions are more stimulating than comparison questions to those giving deceptive answers or that comparison questions are equally or more stimulating than relevant questions to those giving truthful responses.

Most comparison question testing formats face the difficult challenge of calibrating the emotional content of relevant and comparison questions to elicit the levels of response that are needed in order to correctly interpret the test results. It has been argued that an unethical examiner could manipulate the questions and the way they are presented to produce

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