Polygraph Testimony and Juror Judgments: A Comparison of the Guilty Knowledge Test and the Control Question Test1

The present experiment investigated the impact of the Control Question Test (CQT) and the Guilty Knowledge Test (GKT) on the verdicts of mock jurors. Although studies have indicated that polygraph evidence has little influence on jurors' verdicts (Cavoukian & Heselgrave, 1980; Spanos, Myers, Dubreuil, & Pawlak, 1992–1993), no research has previously distinguished between the different types of polygraph tests and their impact on juror verdicts. In the present study, jurors were shown a videotape of a simulated rape-murder trial that contained either CQT polygraph evidence, GKT polygraph evidence, or no polygraph evidence. No differences were found among the 3 conditions for either jury verdicts or individual juror verdicts, and jurors tended to rate both forms of polygraph testimony below other forms of equally suspect evidence, such as eyewitness testimony, in its influence on their decision-making process.     

Trial by polygraph: Scientific and juridical issues in lie detection

The present article examines two methods of polygraph-assisted lie detection: the Control Question Technique (CQT) and the Guilty Knowledge Technique (GKT). It presents the rationale for both, arguing that only the latter is well grounded in psychological theory. It then surveys the empirical support for claims of the polygraph's ability to detect deception, arguing that such support often comes from studies that are methodologically flawed by contamination of various sorts—especially studies of the CQT. The article then explores the legal implications of introducing polygraph test results, as presently gathered, into the criminal courtroom.     

Use of antianxiety drugs as countermeasures in the detection of guilty knowledge.

To evaluate whether antianxiety drugs enable guilty subjects to appear innocent on polygraph tests, we compared the effects of diazepam, meprobamate, and propranolol on the outcome of a guilty knowledge test (GKT). Seventy-five undergraduate students were evenly divided among one innocent and four guilty groups. Subjects in each of the guilty groups received either one of the drugs or a placebo prior to the administration of the GKT and after viewing a videotape that depicted a burglary as seen from the perspective of the burglar. The results showed that drug status had no influence on the outcome of the GKT. Innocent subjects who coincidentally obtained high scores on a recognition memory test covering details of the mock crime tended to obtain higher guilt scores on the GKT.     

A Review of the Office of Technology Assessment Report on Polygraph Validity

This article reviews the assessment of polygraphic lie detection produced by the Office of Technology Assessment (OTA) in November 1983. The review argues that use of the lambda statistic to summarize polygraph accuracy in the OTA report was inappropriate because the studies examined differ widely in base rates of guilt and innocence. Using Lykken's (1981) average accuracy statistic and avoiding overlapping data sets found in the OTA report, the review finds that field studies, analog studies, and guilty knowledge studies produce very similar average accuracy (82% to 88%). Guilty knowledge tests tend to be biased toward false negatives, whereas control question tests tend to be biased toward false positives. The striking similarity of results for field and analog studies using control question tests contradicts the common belief that results of analog testing are weak and not generalizable to field practice. It is argued that analog studies are in fact the only hope for improving the validity of polygraph testing.     

Estimating the validity of the guilty knowledge test from simulated experiments: the external validity of mock crime studies

This experiment was designed to examine the external validity of the standard mock-crime procedure used extensively to evaluate the validity of polygraph tests. The authors manipulated the type of mock-crime procedure (standard vs. a more realistic version) and the time of test (immediate vs. delayed) and examined their effects on the validity of the Guilty Knowledge Test (GKT) and the recall rate of the relevant items. The results indicated that only the type of mock-crime affected the 2 outcome variables. The realistic procedure was associated with a lower recall rate and weaker detection efficiency than the standard procedure. However, these effects were mediated by the type of GKT questions used. Practical implications of these results are discussed.     

A quantitative review of the guilty knowledge test

The guilty knowledge polygraph test (GKT; D. T. Lykken, 1959, 1960) is a psychophysiological method of identifying suspects with concealed information about a crime. A meta-analysis of 50 treatment groups drawn from 22 laboratory simulation studies (total N = 1,247) was conducted to provide a comprehensive estimate of GKT accuracy under controlled conditions. Electrodermal measures correctly identified 76% of participants with concealed knowledge and 83% of those without information. Informed participants were detected at rates significantly in excess of chance, with a mean weighted effect size of .57. Enactment of mock crimes increased the hit rate to 82%. The rates of false-positive error among noninformed treatment groups did not significantly exceed chance. Applications and research directions are discussed.     

Trial by polygraph

A brief review of the early and current applications of lie detectors to detecting deception is presented. Research findings on the accuracy of polygraphs in real life situations are discussed. Additionally, the circumstances of polygraph testing (i.e., friendly or adversarial), methods of “beating” the polygraph, and the acceptance of polygraph results in the courtroom are reviewed.     

The influence of FMRI lie detection evidence on juror decision-making

In the current study, we report on an experiment examining whether functional magnetic resonance imaging (fMRI) lie detection evidence would influence potential jurors' assessment of guilt in a criminal trial. Potential jurors (N = 330) read a vignette summarizing a trial, with some versions of the vignette including lie detection evidence indicating that the defendant was lying about having committed the crime. Lie detector evidence was based on evidence from the polygraph, fMRI (functional brain imaging), or thermal facial imaging. Results showed that fMRI lie detection evidence led to more guilty verdicts than lie detection evidence based on polygraph evidence, thermal facial imaging, or a control condition that did not include lie detection evidence. However, when the validity of the fMRI lie detection evidence was called into question on cross-examination, guilty verdicts were reduced to the level of the control condition. These results provide important information about the influence of lie detection evidence in legal settings. Copyright © 2011 John Wiley & Sons, Ltd.     

Orienting to guilty knowledge

The Guilty Knowledge Test (GKT; Lykken, 1959) assesses whether suspects conceal information about a crime ( “guilty knowledge”). Previous studies have demonstrated larger physiological reactions to guilty knowledge compared to unknown information. In three experiments, we investigated whether guilty knowledge also demands attention. During an alleged polygraph examination, participants were presented with to‐be‐detected pictures ( “guilty knowledge”), nonsignificant, familiar pictures ( “mere knowledge”), and previously unseen pictures ( “neutral information”) for 250 ms in a modified dot probe task. In all three experiments, probe responses were slower on guilty knowledge trials as compared to the neutral trials. Results are discussed in terms of an information‐processing view on orienting to guilty knowledge.     

New perspectives in assessing deception: The evolution of the truth machine

In recent years a growing interest has arisen in the development of tools for the detection of deception. Since William M. Marston's first publication (1917) on the use of the polygraph as a lie detector, the application of this tool, commonly known as the truth machine, has evolved. Modern technologies are now trying to push the issue further, investigating brain activity during deception using functional Magnetic Resonance Imaging (fMRI). The aim of this paper is to summarise the evolution of research from the original use of the polygraph to the use of new technologies in detecting deception, in order to provide an overview of the recent developments on the use of measurements of deception, and promote new research in this highly important domain of applied cognitive psychology.     

GKT范式在确定犯罪嫌疑人中的应用：一项ERP测谎研究

刑事犯罪已成为当今各国面临的普遍性社会问题,因此急需可靠有效的测谎技术以适应当前打击犯罪的严峻形势。本研究采用犯罪知识测试范式（Guilty knowledge test, GKT）,以犯罪嫌疑人为对象,以真实犯罪现场为素材,探究真实环境下与测谎探测相关的脑电指标（P300、CNV）的有效性和可靠性。行为数据显示,目标刺激的反应正确率显著小于探测和无关刺激,探测刺激的反应时显著小于目标刺激和无关刺激。脑电数据显示,相较于目标刺激和无关刺激,探测刺激诱发的P300波幅最大且P300的检出率为92.5%,而在CNV指标上目标刺激诱发的波幅最大, 无关刺激次之,探测刺激最小。结果表明,与CNV相比,P300能够有效检测被试所隐藏的犯罪细节信息,且检出率较高,稳定性较好,是一种较为可靠的测谎指标。     

Alice-in-Wonderland terminological usage in,and communicational concerns about,that peculiarly American flight of technological fancy

Alice-in-Wonderland (AW) terminological usage employs basic terms in a systematically misleading and taxonomically anarchic way. The so-called “control” question “test” (CQT) polygraph procedure, which enjoys a controversial but nevertheless scientific status in North America, involves such AW terminological usage. The basic concepts of “test,” “control,” and “quantification” are loosely employed. There is loose talk about the “detection ofdeception,” which generates the paradox that in the CQT it is in the innocent that deception should be detected. Moreover, deception is not really assessed either in the CQT or in the more scientifically-based Guilty Knowledge Test (GKT). Finally, there is loose practice in evaluating the CQT, which should not be primarily assessed in terms of its overall accuracy, but in terms of its specific effects in improving accuracy through the provision of physiological information to the examiner. Such an assessment has not been carried out even in laboratory analogues of polygraphy. The treatment of the CQT exemplifies most psychophysiologists’ reluctance to treat basic definitional issues seriously, and also means that opponents of the CQT are unable to communicate clearly about it with non-psychophysiologist professionals and with the lay community.     

刺激强度与个体差异对心理生理测谎的影响

探讨不同说谎经验的被试在完成不同刺激强度的任务时,其生理反应和说谎分数的变化。结果表明,在测谎过程中,任务刺激强度不仅影响测试结果的有效性,而且与被试的生理反应和说谎分数成正相关;不同类型被试的说谎分数没有显著差异,测谎应对任何人群都普遍有效。     

Alice-in-Wonderland terminological usage in, and communicational concerns about, that peculiarly American flight of technological fancy. The CQT polygraph

Alice-in-Wonderland (AW) terminological usage employs basic terms in a systematically misleading and taxonomically anarchic way. The so-called "control" question "test" (CQT) polygraph procedure, which enjoys a controversial but nevertheless scientific status in North America, involves such AW terminological usage. The basic concepts of "test," "control," and "quantification" are loosely employed. There is loose talk about the "detection of deception," which generates the paradox that in the CQT it is in the innocent that deception should be detected. Moreover, deception is not really assessed either in the CQT or in the more scientically-based Guilty Knowledge Test (GKT). Finally, there is loose practice in evaluating the CQT, which should not be primarily assessed in terms of its overall accuracy, but in terms of its specific effects in improving accuracy through the provision of physiological information to the examiner. Such as assessment has not been carried out even in laboratory analogues of polygraphy. The treatment of the CQT exemplifies most psychophysiologists' reluctance to treat basic definitional issues seriously, and also means that opponents of the CQT are unable to communicate clearly about it with non-psychophysiologist professionals and with the lay community.     

The validity of psychophysiological detection of information with the Guilty Knowledge Test: a meta-analytic review

The authors performed a meta-analysis based on 169 conditions, gathered from 80 laboratory studies, to estimate the validity of the Guilty Knowledge Test (GKT) with the electrodermal measure. The overall average effect size was 1.55, but there were considerable variations among studies. In particular, mock-crime studies produced the highest average effect size (2.09). Three additional moderators were identified: Motivational instructions, deceptive ("no") verbal responses, and the use of at least 5 questions were associated with enhanced validity. Finally, a set of 10 studies that best approximated applications of the GKT under optimal conditions produced an average effect size of 3.12. The authors discuss factors that might limit the generalizability of these results and recommend further research of the GKT in realistic setups.     

Attitudinal Consequences of the Preemployment Polygraph Examination1

An important factor concerning the use of the polygraph in employment settings has been overlooked: How does the polygraph experience affect a new employee's work-related attitudes? In two experiments, subjects were exposed to different hiring scenarios-one included a polygraph examination, the other did not—and then responded to a questionnaire of work-related attitudes. Results indicated that preemployment polygraph examinations may induce negative work-related attitudes. These findings suggest that the polygraph experience acts as a symbolic communication from the employer to the prospective employee, imparting information that may induce feelings of distrust and dislike. Moreover, the polygraph experience may be interpreted by prospective employees as evidence of high levels of employee theft, thus establishing a new (higher) norm of peer theft behavior.     

Detecting concealed information with reaction times: Validity and comparison with the polygraph

The Concealed Information Test is used to assess recognition of information that the examinee cannot or does not want to reveal. Physiological measures recorded with the polygraph have shown to be highly valid measures of concealed information. Research suggests that a reaction‐time (RT) based test may also successfully reveal concealed information. Due to its simplicity, an RT‐based test has great advantages for applied testing. In the present study, we assessed the validity of the RT‐based test for concealed information detection, and compared its discriminative power with the polygraph. Thirty two participants in a feigned amnesia study were promised a financial reward when successfully concealing autobiographical information. Participants performed an RT‐based test, and a polygraph test. The data support the validity of the RT‐based test for concealed information detection, and indicate its discriminative power is similar to the polygraph. Our data confirm the potential of the RT‐based test. Copyright © 2009 John Wiley & Sons, Ltd.     

P300和CNV在GKT的延时反应范式中测谎效果的分离

在杀人游戏情境中,设置无反馈和有反馈两个实验组,采用GKT（guilt knowledge test）的延时反应范式,考察P300和CNV两种ERP成分的测谎效果。结果发现：刺激呈现后大约450ms诱发了一个正性ERP成分（P300）,大约800ms诱发了一个负性ERP成分（CNV）。在有、无反馈两种实验范式下,P300和CNV两种测谎指标的变化情况发生了分离——有无反馈不影响P300峰值大小;而在有反馈条件下,“杀手”对于犯罪细节词语（探测刺激）的CNV明显增大。使用Bootstrap波幅差异分析法,分别计算P300和CNV在有、无反馈情况下的检出率,结果发现：两个指标的检出率在有、无反馈两种实验范式下的变化情况也发生了分离。这些结果表明：P300和CNV两个指标具有不同的测谎功能,这可能和它们所反映的不同欺骗心理过程有关。     

Effects of questions' repetition and variation on the efficiency of the guilty knowledge test: a reexamination

The effect of question repetition and variation on the efficiency of the Guilty Knowledge Test (GKT), based on electrodermal and respiration measures, was examined in a between-subjects experiment with 3 conditions. Each participant was presented with a sequence of 12 biographical questions. In Condition 1, a single question was repeated 12 times; in Condition 4, each of 4 different questions was repeated 3 times; and in Condition 12, 12 different questions were used. A monotonic relationship between the number of different questions used and detection efficiency was observed only with the electrodermal measure (the areas under the receiver operating characteristic curves, obtained with this measure in Conditions 1, 4, and 12 were .68, .81, and .99, respectively). These results demonstrate that a GKT based on multiple questions is superior to the use of many repetitions of a single or a few questions, and it can reach an almost perfect detection efficiency.     

Science and the CQT polygraph

Despite substantial contrary scientific evidence, polygraph tests to detect deception continue to be employed in the United States. It is argued that polygraph tests (in particular, the Control Question Technique) lack construct validity. Polygraph tests do not assess deceptiveness, but rather are situations designed to elicit and assess fear. Correct decisions may be obtained when subjects fear detection; however, because there is no way to determine the cause of a subject’s fear or anxiety, validation is impossible. The paper also considers other physiological detection paradigms and concludes that because of the nature of honesty their use to detect deception is unlikely.