Editorial

The Psychophysiologist as Innocent Bystander: Ethical Mismatch

Robert J. Barry

Department of Psychology, University of Wollongong, Wollongong 2522, Australia

At the 7th International Scientific Meeting of the International Organization of Psychophysiology (IOP), held in Greece in September-October 1994, I was appointed to the Chair of the Ethics Committee of the IOP. From this

perspective, I was especially pleased when John Gruzelier asked me to be the action editor for John Furedy's paper, "The North American Polygraph and Psychophysiology: Disinterested, Uninterested, and Interested Perspectives",

which is published in this issue of the International Journal of Psychophysiology. Furedy raises a number of important issues in this paper, in what can be recognized as his signature style when writing on contemporary issues -- logically-developed ideas presented in a relatively impassioned and provocative fashion.

The thrust of Furedy's paper is that the psychophysiological research community must adopt an impartial (disinterested) but active adjudicative role (as distinct from being merely uninterested) in the debate about the use of certain paradigms in the detection of deception.

As was explicitly noted by Gershon Ben-Shakhar in the review process, Furedy's call to action must be recognized in the wider context of our ethical role as scientists in the international community. As scientists with relevant expertise -- experts in Psychophysiology -- we are morally compelled to enter the debate about the merits of any activity or procedure which our expertise leads us to believe is of dubious scientific value. It is part of our ethical duty to ensure that the scientific consensus about the validity and reliability of any procedure or practice is clearly communicated to the end user.

For many of us, this is a relatively novel idea. We have tended to see our scientific role with the emphasis on investigator -- as the provider of information which we make available, through scholarly publication, to the community for its use. We leave the practical implementation of our discoveries, and their consequences, up to others. Unfortunately, this separation of the content of our science from the consequences of its application is ethically indefensible. We cannot continue to pretend to be innocent bystanders. Because of our expertise, we cannot claim innocence; in the absence of innocence, we cannot remain mere bystanders. This moral imperative must be accepted by IOP and its members. If we do not actively embrace it, we negate the objectives of IOP as they relate to "the benefit of humanity".

The ethical side of our scientific activities clearly demands more attention and open discussion in order to foster the development of the profession of Psychophysiology. The International Journal of Psychophysiology is available as a forum for such debate. Your general comments on ethical issues or specific comments on Furedy's paper are invited, and should be addressed to the Editors.


Int. J. of Psychophys, 1966 (spring/summer)

The North American Polygraph and Psychophysiology: Disinterested, Uninterested, and Interested Perspectives

John J. Furedy*

University of Toronto, Department of Psychology, Toronto, Ont. M5S 1A1 (Canada; email: furedy@psych.utoronto.ca)

Keywords

"Control" question "test" (CQT) polygraph; disinterested, uninterested, and interested perspectives; detection versus differentiation of deception; experimental versus control comparisons; specifiable tests versus unspecifiable dynamic interviews; independent- versus dependent-variable considerations.

Abstract From both a scientific and an applied psychophysiological point of view, the related but different ideas of using physiological measures to differentiate and detect deception are of considerable potential interest. This paper's primary concern is with psychophysiological detection, and it is mainly focussed on the North American "Control" Question "Test" (CQT). The treatment is disinterested in the sense that there is an insistence on employing fundamental terms in a logically consistent way. Following a detailed description of the CQT, and an analysis of it and related psychophysiological deception procedures, it is suggested that, by and large, the North American research psychophysiological community has failed to measure up to the standards of disinterestedness with respect to the psychophysiological detection of deception. Instead it has adopted an _uninterested_ perspective, which has allowed the _interested_ community of professionals who employ the CQT to hood-wink both themselves and others (including the American Psychological Association) that the CQT is a controversial, but scientifically-based, test for detecting deception. As the most cognate organization, the international psychophysiological research community needs to take a more active and disinterested role in this salient purported application of psychophysiology--the detection of deception.

The North American Polygraph and Psychophysiology: Disinterested, Uninterested, and Interested Perspectives The subtle changes in physiological functions that psychophysiologists study appear to show considerable promise in the area of deception, because they are not under voluntary control. Plausible also is the initial rationale of the polygraph, according to which deception will be revealed by the greater guilt or anxiety that we feel when we lie, in contrast to when we are truthful. However, a scientific analysis has to go beyond such vague intuitions, and consider the issues in a more analytic and disinterested mode. A disinterested approach includes making basic distinctions, providing a clear description of relevant procedures, and eschewing the use of systematically misleading terms, no matter how popular these may be in both the popular and technical literature. The first section of this paper provides such a disinterested psychophysiological perspective for what is the most common embodiment of the psychophysiological detection of deception--the North American so-called "control" question "test" (CQT) polygraph. The second section addresses the response to this issue by a major scientific society into the province of which the CQT most clearly falls.

Deception and Psychophysiology

If psychophysiology is conceived as the study of psychological processes by means of observing subtle changes in physiological functions (see [1]; for a different view, see [2]), then the psychophysiological study of deception involves an experimental isolation of the phenomenon. This analysis distinguishes it from other psychological processes (i.e., confounding variables), provides evidence on factors that affect the deception phenomenon, and begins to formulate theoretical explanations for the phenomenon.

Deception as a psychological process is of particular scientific significance. From an evolutionary perspective, it may be viewed as the human development of the hiding response, which is itself the important alternative adaptation to the fight-or-flight response. In contrast to the applied perspective which views deception as something immoral to be detected in individuals, the scientific, basic-research perspective views deception as a process with evolutionary significance, and as one of the possible markers of the distinction between human and non-human animals. Psychophysiological experimental work to _differentiate_ deception as a psychological process has begun relatively recently [3], and has not attracted as much attention as the differentiation of other psychological cognitive (attentional-orienting) and non-cognitive (e.g., emotional) processes.

In contrast, the detection of deception involves classifying _individuals_ as deceptive or truthful, and it is clear that its main significance is applied. It is this purported application of scientific psychophysiology that is the main concern of those many people in North America whose lives are directly or indirectly affected by the Polygraph or Lie Detector. Moreover, as technology spreads around the world, although modified by cultural factors, it is clear that non-North American lives may also be affected by the Polygraph.

Currently, the dominant procedure is the North American "Control" Question "Test" (CQT) Polygraph, which I shall describe in some detail. It is particularly important to provide such a basic description in the case of the North American CQT, because many descriptions of the procedure are systematically misleading or deceptive, as regards fundamental terms like "control", "quantitative", and "test".

For example, because the procedure is not really a "test"_1_, any description can only be in outline form, the CQT being specifiable only to the extent that relatively unstructured interviewing procedures are specifiable. Within those limits of specifiability, it is possible to characterize the CQT as a four-phase procedure that is preceded by a pre-

CQT period during which time the examinee has become a suspect to a particular crime, agrees "voluntarily" (see e.g., [4], for problems of "voluntariness" and whether a consent given is truly "informed") to be "tested", and the examiner is informed about the examinee (and hence has a view about the examinee's honesty before the CQT proper begins). _The North American CQT Polygraph: Description and Significance_.--The first, "pre-test-interview" phase lasts from about 30 to 60 minutes. During this interview, the examiner seeks to convince the examinee that the polygraph is infallible, raises various topics (e.g., mother's death) that may be useful during the later "post-test-interview" (confession-eliciting) phase (e.g., "your poor dead mother would have wanted you to confess"), asks various medically-related questions, and discusses the formulations of both the relevant and "control" questions with the examinee. The relevant questions are reformulated until they are unambiguous to the examinee and s/he can clearly answer "no" to them (i.e. indicating innocence); the "control" questions are reformulated until a version is arrived at to which, in the examiner's view, the examinee's answer ("no") is either totally deceptive or at least not confidently truthful.

The next phase is the "test" phase, which lasts for about half an hour, and is begun by connecting the physiological recording equipment to the examinee. Modern CQT polygraphers typically use a machine that records four channels, namely the electrodermal response (GSR), relative blood pressure obtained through having the pressure cuff inflated between systolic and diastolic levels (cardio), and two channels of respiration taken from the chest and stomach. Changes in these functions following the various questions constitute the basic psychophysiological data, and the basic rationale is that if the relevant questions are followed by greater physiological changes than the "control" questions, then the examinee is deceptive.

The "test" phase proper is often preceded by the card or "stim" test, in which the examiner uses the instrument to detect, from a number of cards, the card of which the examinee is thinking. The purpose of the "stim" test is to convince the examinee of the polygraph's infallibility, so, even though this sort of physiological detection is relatively easy to perform, the stim test is usually rigged to ensure that correct detection always occurs. Following this demonstration of infallibility, the "test" proper is presented. It usually comprises a set of ten questions of which three each are relevant questions (e.g., Did you take the money?) and three are "control" questions (e.g., "Aside from the crime under investigation, did you ever take more than $5 that was not yours?")._2_ The questions are separated by about 30 seconds, and each repetition through the list is called a "chart". After three "charts", the examiner can decide whether or not to give one or two more.

The third phase is initiated by the examiner's decision to give no more "charts". The examiner leaves the examinee alone for some 20 minutes, in order to score the physiological records. The scoring method most commonly in current use is the "numerical" one. In this method, responses to pairs of adjacent relevant and "control" questions are compared

separately for each of the four physiological channels. If the examiner finds a "marked", "clear", or "slight" difference, the numbers 3, 2, and 1, respectively, are assigned to the comparison. The algebraic sign of the number is determined by the direction of the relevant-"control" difference. If the algebraic sum of these comparisons (of which there would be 4 x 3 x 3 in a CQT session using a 4-channel recorder, three pairs of relevant and "control" questions, and three "charts") is less than -5 or -6 (the latter criterion being used by more "conservative" examiners), the examinee is classified as deceptive. If the algebraic sum exceeds +5 or +6, the examinee is classified as truthful. Finally, if the algebraic sum falls between -5 or -6 and +5 or +6, then the CQT is considered to have been "inconclusive".

This third "scoring" phase also has an obvious, though difficult-to-specify role in the fourth and final phase, which occurs if the examiner has decided that the examinee is guilty. On returning to the room where the examinee has been left alone to ponder his or her fate, the examiner administers this "post test interview". The purpose of this interrogatory phase (the existence of which most examinees are unaware when they agree to be "tested" by the polygraph) is to induce a confession._3_ The phase can last from 10 minutes to several hours, and is terminated either by a confession or by the examiner's decision that no confession can be obtained from the examinee.

The CQT and Related Psychophysiological Deception-related Procedures.

Turning now to an analysis of the CQT and related procedures, Table I summarizes some information about the CQT, an alternative detection-of-deception procedure, the GKT (Guilty Knowledge Test), and the DDP, which is a _differentiation_-of-deception procedure. The table also provides examples of items in each procedure, as well as specifying the psychophysiological response comparisons on the basis of which inferences are made. A description of what goes on in the first, CQT procedure has already been provided. It is also worth noting that the example of a relevant (R) question given in the table is one with some degree of _surface_ validity. That is, it may at least be plausible for some to believe that an R>C result may indicate guilt. In real life, however, R questions are frequently of the sort that it takes a huge leap of faith to even suppose that an R>C outcome indicates guilt. For example, in a 1984 Canadian case in which I was involved ([4], p. 108), the R question which was put to a 74-year old crossing guard with no prior criminal record or deviant sexual history was "Did you lick X's vagina?", where X was a 4-year old; the comparison C question was "Did you ever do anything you were ashamed of in your life?". Yet the polygraph examiner had enough faith in his procedure and machine to testify that, in his professional opinion, the obtained R>C outcome did strongly indicate that the examinee was guilty. However, even with the more plausible sorts of R questions indicated in the figure, basic terms like "control" and "test" are used in ways that are not consistent with normal usage. For experimental psychophysiologists, it is the Alice-in-Wonderland usage of the term "control" that is most salient. There are virtually an infinite number of dimensions along which the R and the so-called "C" items of the CQT could differ. These differences include such dimensions as time (immediate versus distant past), potential penalties (imprisonment and a criminal record versus a bad conscience), and amount of time and attention paid to "developing" the questions (limited versus extensive). Accordingly, no logical inference is possible based on the R versus "C" comparison. For those concerned with the more applied issue of evaluating the accuracy of the CQT procedure, it is the procedure's _in-principle_ lack of standardization that is more critical. The fact that the procedure is not a test, but an unstandardizable interrogatory interview, means that its accuracy is not empirically, but only rhetorically, or anecdotally, evaluatable. That is, one can state accuracy figures only for a given examiner interacting with a given examinee, because the CQT is a dynamic interview situation rather than a standardizable and specifiable test. Even the weak assertion that a certain examiner is highly accurate cannot be supported, as different examinees alter the dynamic examiner-examinee relationship that grossly influences each unique and unspecifiable CQT episode.

In addition to these in-principle lack-of-standardization considerations, there are also two specific features of the CQT that arise from the general lack of standardization, and that render valid accuracy evaluations impossible. One of these features is contamination. Judgments made by the examiner on the basis of the psychophysiological information are significantly (although to an extent that is not precisely measurable) affected by prior information about the examinee that is available to the examiner. The second feature is the uncertainty about the way in which the "control" questions are perceived by the examinee. As part of the CQT protocol, the examiner tries to persuade the examinee that performance on the "control" questions is as important as performance on the relevant questions, because if the examinee is caught out in a lie on the "control" questions, this will result in being perceived as being deceptive in general, and hence also on the relevant question. This raises an ethical problem with the CQT, because the examiner's thesis, in this respect, is a lie. More importantly, however, it introduces a source of error which is impossible to precisely estimate, but which can be assumed to increase with time. That is, as there is greater public familiarity with what is really critical for "passing" a CQT (i.e., producing a clear "C">R result), the CQT becomes increasingly ineffective with those examinees who (through publicly available channels which now include even popular newspaper articles) are not deceived by the examiner's lie about the function of the "control" questions in the CQT.

In contrast, and as has been detailed elsewhere (e.g., [5]), the Guilty Knowledge Test or GKT is a detection-of-deception procedure wherein the terms "test" and "control" are appropriately used. The conditions under which a GKT may be used are stated in Table I just above the examples of R and C questions for the GKT. A minimal requirement for such conditions is that the police keep at least some details of the crime being investigated from the public. The example questions do allow the inference of the presence of guilty knowledge with an error rate that is, in principle, specifiable. In fact our information on the actual accuracies for the GKT is limited, because sufficient empirical research is lacking, especially in the field. North American polygraphers have largely eschewed using the GKT, and two recent Israeli studies [6, 7] appear to constitute the only available methodologically sound pieces of field research. Still, in contrast to the CQT, the GKT is, in principle, a scientifically based procedure which is specifiable, and where the experimental/control contrast constitutes appropriate terminological usage. Finally, even the GKT does not assess deception directly, but only allows the indirect inference of deception (about whether the examinee committed the crime) indirectly through the presence of guilty knowledge. Only the Differentiation of Deception Paradigm or DDP, the third procedure in the table, assesses deception as a process in the experimental psychophysiological sense. As detailed elsewhere (e.g., [3]), and as summarized in the table (paragraph immediately above DDP response comparison), the DDP is a laboratory procedure which contrasts items that, at least on the face of it, differ only with respect to whether they are deceptively (D) or honestly (H) answered. And like any psychophysiological phenomenon that is specified in terms of the experimental > control assessment, obtaining the D>H effect identifies the deception process, while if another manipulation affects the D>H effect, one may infer that this factor influences the process of deception.

The Society for Psychophysiological Research (SPR)'s Uninterested Approach to the Detection Deception

There are at least two reasons why SPR should have a disinterested or expert approach to the psychophysiological detection and differentiation of deception. In the first place, in terms of such measures as the scientific prestige of its journal, _Psychophysiology_, SPR is the acknowledged leader among societies concerned with the science of experimental psychophysiology. Secondly, SPR is predominantly American in membership, and lie detection (employing the CQT) is most frequently practiced in the U.S. Indeed, in many European countries, psychophysiological lie detection is either outlawed and/or is regarded with the same level of skepticism that is reserved for, say, tea-leaf reading. In claiming that SPR has, in fact, an uninterested attitude, I am, of course, asserting a statistical generalization which certainly has its individual exceptions. Nevertheless, the evidence does support the view that, _by and large_, SPR has failed to measure up to the standard of disinterested analysis as regards the psychophysiological detection of deception. _The detection of deception_. There are three lines of evidence that suggest an uninterested approach by SPR to the procedure that is predominantly used in the psychophysiological detection of deception in North America: the CQT.

1. The Society's first attempt to handle the polygraph was, in the mid-seventies, a committee of expert members to report on the scientific basis of what was, and is, psychophysiology's most salient purported application, especially in North America. For whatever reason (perhaps because the committee included, as members, David Lykken and David Raskin, who were certainly then, respectively, the most prominent psychophysiological opponent and proponent of the CQT), the committee broke up without submitting any report. Since then SPR has not set up another committee to deal with this issue, even though many other committees have issued reports on less psychophysiologically salient matters._4_

2. The second way in which SPR has dealt (or, rather, failed to deal) with the polygraph is to have individual proponents conduct surveys of members concerning their "attitude" towards its scientific basis. Such a survey was carried out in the eighties, and another one has recently been reported [8]. Surveys of this sort are meaningful only if the respondents are familiar with what the CQT procedure actually entails, including the deceptive use of basic terms like "test", "control", and "quantification". If the respondents are not aware of these facts, the results of such surveys represent only the opinions of a professional but ignorant group of respondents._5_

3. Consistently with survey opinions, and providing a more authorative boost for the CQT, has been some of the contents of SPR's official journal, _Psychophysiology_, which has implicitly accorded the CQT relatively high status by publishing articles that, by implication, treat the CQT and the GKT as if they were alternative and roughly-equal-status experimental procedures for the psychophysiological detection of deception. This equal-status treatment by _Psychophysiology_ is particularly important, because the journal has a reputation for considerable methodological rigour, and is the clear first choice of most psychophysiological researchers for publishing their experimental papers. Yet as late as 1994, O'Toole et al. [9], published a paper entitled "Alcohol and the Physiological Detection of Deception: Arousal and Memory Influences" in which the CQT and GKT were treated as alternative procedures of equal methodological status, and the various methodological critiques of the CQT were not even alluded to in the reference list. Also significant is the fact that one of the co-authors had, in 1991, reviewed our book on deception [5] in _Contemporary Psychology_. In what was a generally positive review entitled "Another Critical Assault on Lie Detection", he stated that the CQT's so-called "control" questions are "not controls in the normal scientific sense of that term" ([10] p. 863). The North American CQT Polygphers and Their Interested Approach.

The CQT polygraph is part of warp and woof of North American society. It is appealed to when truth is difficult, and yet important, to establish, in much the same way as duels were appealed to in European society some two hundred years ago. It is therefore clear why the polygraph profession, as most saliently represented by the American Polygraph Association (APA), should be vitally interested in promoting the status and use of the CQT. This interest has produced a genuine conviction by the APA that the CQT, like a magic potion, is, in fact, universally applicable. So, although even polygraphic texts indicate that the CQT is valid only for unambiguous acts, in fact the CQT is employed by polygraphers to resolve far more ambiguous situations.

In my own experience, I have testified against a polygrapher, who, in a Canadian case in 1982, was convinced that he could use the CQT to determine not the actions, but the state of mind of an accused who served as a get-away driver in a robbery. The charge was murder not only against the robber who shot the shop keeper, but also against the get-away driver. Although all other evidence suggested that the get-away driver did not conspire with the robber to kill the shop keeper (there was even some doubt as to whether he knew the robber had a gun), the polygrapher testified that the CQT allowed him to determine that the driver's state of mind was such as to render him guilty of murder. Again, in 1993, in a military court martial for rape (which carried a 40-year sentence), I testified against a polygrapher (actually two of them) who purported to use a form of the CQT to determine whether the accused was involved in rape or consensual sex. The parties agreed that the act--felatio between a male sergeant and a female private in his platoon during Desert Storm--had occurred, and that there had been no physical force used, or even threats of physical force uttered by the alleged rapist. Eight hours of "post-test-interview" interrogation produced a confession to rape, i.e., an assertion about a state of mind that existed about a year ago in the accused._6_ So the CQT polygraph is really used like a magic potion (good for any occasion), rather than as a scientifically-based procedure which is useful only under specified conditions. In a sense, then, polygraph professionals appear to be deceiving themselves when they employ the CQT.

However, the North American lay and professional communities, as well as professional organizations, also appear to be hood-winked by the CQT. The printed and visual media are replete with appeals to the polygraph by parties involved in disputes, and it is clear that the side that allows itself to be "polygraphed" has a rhetorical advantage in any argument. Moreover, qualified professionals are not immune. In the dispute between Clarence Thomas and Anita Hill, the latter, who is a professor of law and therefore is familiar with the polygraph's doubtful status as legal evidence (even in the U.S.A.), nevertheless used as an argument to bolster her case the fact that she had submitted to and "passed" a polygraph._7_ Again, although the Canadian Supreme Court ruled in 1987 that the polygraph was inadmissible as evidence, a few years later, one of the Chief Justices, while questioning a witness of whose probity he was uncertain, threatened the witness with a polygraph if he "did not tell the truth".

Still, the above examples are of individuals who may be considered to have been under considerable pressure to "get at" the truth, and have therefore availed themselves of a procedure that they, in cooler moments, know to be unscientific. However, professional communities have also supported the CQT. In 1988 the U.S. Congress banned the polygraph in its industrial use, and the American Psychological Association, which acted as a supportive information source for this motion, considered itself to have struck a blow for science-based technology by condemning the industrial version, but (by implication, at least) accepting the non-industrial, CQT version. That view, however, is self-deceiving, because the U.S. bill permits "specific-incident" (i.e., the CQT) form of polygraphy, under the following "severe restrictions" that (a) the "test" last at least 90 minutes, (b) the examinee get the questions in advance, and (c) the examiner have at least 6 months of polygraphic training. It is obvious once one considers what the CQT involves that these "severe restrictions" are irrelevant, and do not change the fact that the CQT is, like tea-leaf reading, an unstandardizable, motivationally dynamic episode rather than a scientifically-based, specifiable and objectively scoreable test. This is why this peculiarly North American flight of technological fancy constitutes, in my view, that society's second most serious social disease, after that of AIDS._8_ Nor is it clear that other cultures are totally immune to the CQT, because everyone faces situations where it is important, yet difficult, to know the truth. The truth about the CQT, however, is easy to know once one cuts through the deception and self-

deception that is involved. The responsibility for doing this is particularly relevant for the scientific area that is most relevant for evaluating the CQT's scientific status--experimental psychophysiology. More generally, it is experimental psychophysiology that should be foremost in the investigation of both the detection and differentiation of deception. So far this challenge has barely been taken up by the international psychophysiological community.

Acknowledgment This article originated from an Invited Plenary Address at the Congress of the International Organization of Psychophysiology, Thessalonica, Greece, October, 1994, but has been considerably revised in both tone and content in the light of helpful criticisms from two anonymous reviewers, G. Ben Shakhar, and R. Barry.

References

[ 1] Furedy, J.J. (1983). Operational, analogical, and genuine definitions of psychophysiology. _International Journal of Psychophysiology_, 1: 13-19.

[ 2] Obrist, P.A. (1981). _Cardiovascular Psychophysiology: A Perspective_. New York: Plenum Press.

[ 3] Furedy, J.J., Davis, C., and Gurevich, M. (1988) Differentiation of deception as a psychological process: A psychophysiological approach. _Psychophysiology_, 25: 683-688.

[ 4] Furedy, J.J. and Liss, J. (1986) Countering confessions induced by the polygraph: Of confessionals and psychological rubber hoses. _The Criminal Law Quarterly_, 29: 92-114.

[ 5] Ben-Shakhar, G. and Furedy, J.J. (1990) _Theories and Applications in the detection of deception: Psychophysiological and cultural perspectives_. Springer-Verlag, New York.

[ 6] Elaad, E. (1990). Detection of guilty knowledge in real-life criminal investigations. _Journal of Applied Psychology_, 75: 521-29.

[ 7] Elaad, E., Ginton, A., & Jungman, N. (1992). Detection measures in real-life criminal guilty knowledge tests. _Journal of Applied Psychology_, 77: 757-767.

[ 8] Amato, S.L. and Honts, C.R. (1994) What do psychophysiologists think about polygraph tests? A survey of the membership of SPR. _Psychophysiology_, SPR. Abstracts: S22.

[ 9] O'Toole, D., Yuille, J.C.., Patrick, C.J., and Iacono, W.G. (1994) Alcohol and the physiological detection of deception: Arousal and memory influences, _Psychophysiology_, 31: 253-263.

[10] Iacono, W. (1991) Another critical assault on Lie Detection. [Review of theories and applications in the detection of deception: A psychophysiological and international perspective]. _Contemporary Psychology_ 36: 862-864.

[11] Anastasi, A. (1988). _Psychological Testing_, 6th ed. New York:McMillan.

[12] Furedy, J.J. (1986) Lie detection as psychophysiological differentiation: Some fine lines. In M. Coles, E. Donchin, and S. Porges (Eds.), _Psychophysiology: Systems, processes, and applications--A handbook_. Guilford, New York, pp. 683-700.

[13] Furedy, J.J. (1965) Reinforcement through UCS offset is classical aversive conditioning. _Australian Journal. of Psychology_, 18: 255-261.

Footnotes

_1_Many writers refer to the CQT as a "test", but since the formulation of the questions (especially the "control" questions) depends on the individual examiners and their interactions with the examinees, the procedure is not standardized. In addition, the scoring (see below) is not fully objective. Hence, in the light of standard psychological text definitions of a psychological test (e.g., "an objective and standardized measure of a sample of behavior" [11, p. 23], the CQT is not a "test".

_2_The remaining four questions comprise the first-presented "buffer"question which is supposed to "habituate out" the orienting reaction, and three interspersed "irrelevant" questions. However, for deciding whether the examinee is "deceptive", only the responses to the relevant and "control" questions are scored.

_3_Polygraphers wrongly regard such confessions as certainly true, because they do not recognize that elementary psychological principles of motivation indicate that some confessions may be false [4]. However, the fact is that, unlike polygraphic evidence, confession-based evidence is generally admissible, and generally carries considerable face-validity weight in the courts. As an anecdotal, but striking, piece of evidence for the value that polygraphers place on the confession-inducing functionof the CQT, a Toronto police polygrapher described himself in court not as a polygrapher with interrogation skills, but as an interrogator with polygraphic skills.

_4_For example an SPR July, 1994 mailing included a committee report entitled "Summary of the Report of the Ad Hoc Enhancement Committee". The topic of the solicited committee report dealt with "whether the Society is meeting the needs of women, minority, and student members", and complained that the Board did not appropriate more funds for an additional survey of the full membership concerning these problems.

_5_A few years before I discovered what the CQT actually involved [12], and many years after my first publication in psychophysiology [13], I was part of this ignorant (of the CQT) group of psychophysiologists.

_6_In this case, it is interesting to note that, as I indicated to the Court, a "control" question to the Mexican-American, Catholic examinee was, "Did you ever try to persuade a woman to have sex with you when, at first, she did not want to?". Because this sort of question is likely to produce far less emotional impact in an examinee compared to rape-related relevant questions, it is little wonder that "the machine" showed a very clear "deceptive" (i.e., R>"C") outcome, and hence encouraged the two polygraphers to bear down on the examinee to produce his confession.

_7_This also constitutes a "magic-potion" use of the CQT, because the specific issue being "investigated" was not whether a relatively unambiguous act like theft, but the much more ambiguous issue of whether, some 13 years ago, sexual harassment had occurred.

_8_Of course the consequences of AIDS are far more serious, but in terms of comparative risk at least in North America, it appears that the risk of being polygraphed is greater both quantitatively and qualitatively. Quantitively, anyone in the workplace can become a potential suspect, and the more responsible the position, the more risky it is to refuse to "take" the polygraph. Unlike AIDS, there appear to be no parallels either to abstinence or "safe-sex" risk-avoidance options in the case of the polygraph. One parallel between the two social diseases is that the probability of being exposed is relatively low for each individual, so that most people tend to ignore the danger until they or their close associates become victims.