been shown to exhibit cardiovascular patterns associated with threat, including increased myocardial contractility, decreased cardiac output, increased total peripheral resistance, and increases in blood pressure (Blascovich, 2000; Blascovich et al., 2001b).
These studies suggest that stigma may affect polygraph test accuracy. Specifically, they suggest that if either the examiner or the examinee bears a stigma, the examinee may exhibit heightened cardiovascular responses during the polygraph testing situation, particularly during difficult aspects of that situation such as answering relevant questions, independently of whether he or she is answering truthfully. Such responses would be likely to increase the rate of false positive results among examinees who are members of stigmatized groups, at least on relevant-irrelevant and comparison question tests.15 (In Chapter 4, we discuss the very limited empirical research examining the effects of stigma-related characteristics of examiners and examinees, such as race and gender, on the accuracy of polygraph diagnoses of deception.)
Expectancies have been a subject of social-psychological research for the past 40 years. In the early 1960s, Robert Rosenthal began one major line of research, examining the social psychology of the research situation; he hypothesized and verified the so-called experimenter expectancy effects. He demonstrated that experimenter biases affected the results of experimental psychological studies in many situations, even when the experimenters had no intention to do so. Expectancy effects have been tested outside the research situation hundreds of times in a variety of settings (e.g., Rosenthal and Jacobson, 1968; Rosenthal and Rubin, 1978; Harris and Rosenthal, 1985; Rosenthal, 1994; McNatt, 2000; Kierein and Gold, 2000). The most familiar example of expectancy effects is the so-called “Pygmalion effect,” in which teachers’ initial expectancies about specific students’ potential can affect the students’ future performance in the classroom and on standardized tests.
Expectancies in the polygraph testing situation have the potential to affect the validity of such testing.16 It is reasonable to assume, for instance, that an examiner’s belief, or expectancy, about examinees’ guilt or innocence in a criminal investigation setting may cause the examiner to behave differentially—for instance, in a more hostile manner—toward examinees believed to be guilty or deceptive. Such behavior would plausibly create differential emotional reactions in examinees that could affect physiological responses that are detected by the polygraph. These emotional reactions would plausibly be strongest in response to questions about which the examiner expects deceptive responses, thus possibly