Introduction
The use of physiological recording instruments to detect ‘lying’ by criminal suspects has been practiced since at least the early 1900s. Today this field is popularly known as ‘lie detection’ and more formally as polygraphy, the detection of deception, or, more recently, forensic psychophysiology.
In spite of its long history and widespread use, polygraphy is one of the most misunderstood and most controversial techniques in the forensic sciences. In this article, a brief history of this field is provided with respect to polygraph instrumentation and testing procedures. The current state of the field regarding testing techniques, the uses of polygraph testing, polygraph examiners, accuracy, legal status and trends in the field are discussed.
History
Polygraph instrumentation
Development in the field can be traced along two lines, one involving changes in instrumentation and the other changes in testing techniques. Historically, the most dramatic attempts at lie detection relied upon ordeals such as placing a hot iron on the tongue of suspects, who presumably would be protected by their innocence or burned by their guilt. (Guilty persons, whose fear of detection would decrease the activity of salivary glands, would be burned because of the absence of moisture to protect the surface of the tongue. Innocent persons would not suffer such a fate.) This approach to lie detection, and similar early attempts, did not, of course, involve the use of scientific recording instruments. It was not until 1895, when Lombroso, an Italian criminologist, and his student, Mosso, used the hydrosphygmograph and the ‘scientific cradle’ that objective measurement of physiological changes became associated with the detection of deception. Following Lombroso, Mun-sterberg, a psychologist, and others noted the effect of lying on breathing, cardiovascular activity, involuntary movements and the galvanic skin response (GSR) – changes in the electrical resistance in the skin. In 1917, Marston reported success at detecting deception with discontinuous measurements of blood pressure. In the 1920s, Larson devised an instrument for making continuous recordings of both blood pressure and breathing. In the 1930s, Keeler refined Larson’s apparatus by adding a device for recording GSR. In the late 1940s, Reid incorporated into the polygraph instrument a device for detecting covert muscular movements.
Although modern polygraphs represent considerable technological improvement over earlier devices, it is true, nevertheless, that the physiological measures incorporated in earlier instruments are essentially those used today. The modern polygraph is a briefcase-sized device that, in most field settings, records changes in electrodermal activity (GSR) by means of two electrodes attached to the fingertips. A standard blood-pressure cuff, which is partially inflated during testing, is used to record changes in relative blood pressure and pulse rate. Finally, changes in breathing activity are recorded by hollow, corrugated-rubber tubes, one placed around the abdomen and one around the upper thorax, which expand and contract during inhalation and exhalation. Activity in each of these physiological systems is monitored by either electronic or mechanical means and is permanently recorded on a paper chart by a pen-and-ink system. Thus, the polygraph, merely a device that enables the simultaneous recording of a number of physiological changes, is a recording device, not really a lie detector.
It is of interest to note that, because of changes in technology, it is possible to convert the polygraph’s analog signals to digital form and thus to make use of computerized instruments. Such devices are widely used today and, while they have altered the way in which physiological data are displayed (e.g. on a computer monitor instead of only on a printed sheet), and to some extent how they are analyzed, the computerized polygraph is still only a recording instrument.
Polygraph testing techniques
Although commonly referred to as a lie detector, neither the polygraph nor any other device is capable of detecting a lie. This is because there is no known physiological response that is unique to lying. Lie detection is in fact an inferential process in which lying is inferred from comparisons of physiological responses to questions that are asked during the testing. It is for this reason that polygraph techniques -that is, the ways in which testing is administered and polygraphic data are interpreted – have shown more dramatic changes than the polygraph instrument itself.
There are two major testing techniques in use today, the Relevant/Irrelevant Technique (RIT) and the Control Question Technique (CQT). A third approach, generically referred to as Concealed Information Testing (CIT), is encountered in the field in special situations. Each of these procedures is in reality a family of related approaches whose specific use is governed by the situation at hand.
The RIT was widely used by early practitioners and, though in lesser favor today, it is still the procedure of choice among some examiners. Like all accepted techniques, it consists essentially of a pretest interview and polygraphic testing. During the interview the examiner discusses with the examinee background information relative to the investigation at hand, and attempts to become familiar with the examinee’s language and personal history in order to ensure that the test questions are properly worded.
Essentially, the RIT consists of the asking of a series of relevant questions interspersed among irrelevant questions. Relevant questions are those pertinent to the crime at hand (e.g. ‘Did you shoot John Doe?’).
Irrelevant questions are not crime-related but usually ask about issues to which the examiner and the examinee know the truth is being told (e.g. ‘Are you over 18 years of age?’). The series of questions may be repeated two or more times during the testing.
In the RIT, simply stated, physiological responses that accompany the relevant questions are compared to each other and to those seen at irrelevant questions. Lying is inferred from the consistency, degree and pattern of the responses.
The CQT was developed in the late 1940s by J.E. Reid. Today there are numerous variations of the CQT, all similar in principle. The CQT, like the RIT, consists of a pretest interview and polygraph testing. During the pretest interview the exact test questions are prepared and reviewed with the examinee. The question list consists of relevant, irrelevant and control questions, although other types of questions (e.g. ‘guilt complex’ questions) may also be asked. The relevant and irrelevant questions are similar to those asked during RIT testing. Control questions deal with matters similar to, but of presumed lesser significance than, the offense under investigation. An example of a control question in a theft case might be: ‘Did you ever steal anything?’ In the CQT, there are two or more repetitions of the question list during an examination.
Simply stated, in the CQT physiological responses that are consistently greater to relevant questions than to control questions indicate lying on the relevant issues. Conversely, consistently greater physiological responses to the control than to relevant questions indicates truthfulness in the matter under investigation.
The CIT procedure is limited to situations in which specific details of a criminal offense are known to the police but are not made available to the examinee. In the CIT, questions consist of a stem question with multiple options.
The options would include the names of various weapons, e.g. gun, knife, club, etc., including the actual one used. Assuming that the examinee had no way to know which option was the correct one in a series of such questions, a greater physiological response to the correct than to the incorrect options would suggest that the examinee recognizes critical information that only the ‘guilty’ person would know; thus, such an outcome would indicate ‘guilt’. Typically, a series of three or more such multiple-choice stem questions are asked, usually after initial testing by either the RIT or CQT. Unfortunately, most crimes do not lend themselves to such testing because valuable details about the offense either are not available or their integrity is compromised.
Uses of Polygraph Testing
In police work there are two major uses of polygraph testing. The first is referred to as specific issue testing. Here polygraph testing is used to investigate whether or not a person is involved in the commission of a particular offense. This type of testing has been shown to be extremely valuable. In addition, it is also well established that polygraph testing often exonerates persons against whom the circumstantial evidence is quite incriminating. Thus, there are great savings of investigative time and effort due to the use of polygraphy.
The second use of polygraphy is in pre-employment screening of applicants for police work. Such testing is employed by over 60% of the large police agencies in the USA and is most commonly used to investigate issues that cannot otherwise be resolved. A number of studies support the effectiveness of polygraphy for this purpose.
Polygraph testing also plays a role in the work of many intelligence agencies. It is used to determine the suitability of applicants for employment, and period-cially to screen employees regarding their conduct in dealing with, for instance, persons involved with foreign governments. Finally, polygraph testing plays a role in the investigation of suspected violations of sensitive intelligence matters.
Polygraph Examiners
There are three general categories of examiners in the USA: police examiners, those employed by nonfederal law enforcement agencies; government examiners, those who work for federal law enforcement and intelligence agencies; and private examiners, those who contract for their services. The data shown in Table 1 reveal that polygraph examiners are overwhelmingly male (95%) and college-educated (70%). (Although there are important differences between categories of examiners in some areas, the lack of space precludes discussion of them.) The great majority of them have served an internship (87%) but only a small majority (51%) work in the field of polygra-phy on a full-time basis.
It can also be seen in Table 1 that 85% of examiners make use of CQT procedures as their primary method. A much smaller percentage use the RIT and other, less familiar procedures.
It is of interest to note that, in spite of the general belief that polygraph results do not play a role in judicial proceedings, 63% of examiners have testified in court regarding polygraph outcomes (Table 1). Nevertheless, only 20% believe that all polygraph examination results should be admissible evidence.
Table 1 Personal characteristics and selected practices of polygraph examiners by category
| Examiner category | ||||||||
| Item | Police | Private | Federal | Combined | ||||
| n | (%) | n | (%) | n | (%) | n | (%) | |
| Sex? | ||||||||
| Male | 488 | (96) | 256 | (95) | 149 | (90) | 893 | (95) |
| Female | 21 | (04) | 14 | (05) | 17 | (10) | 52 | (05) |
| Education? | ||||||||
| No college degree | 132 | (40)36? | 50 | (27)30? | 3 | (02)3? | 185 | (30)29? |
| College degree | 139 | (28)38? | 67 | (25)40? | 41 | (25)38? | 247 | (27)38? |
| Graduate work | 95 | (32)26? | 50 | (48)30? | 64 | (73)59? | 209 | (43)33? |
| Internship? | ||||||||
| Yes | 430 | (85) | 231 | (86) | 157 | (96)95? | 818 | (87) |
| Work effort? | ||||||||
| Full time | 225 | (46) | 123 | (48) | 114 | (74) | 462 | (51) |
| Part time | 270 | (54) | 133 | (52) | 40 | (26) | 443 | (49) |
| Primary technique? | ||||||||
| CQT | 438 | (89) | 150 | (80)74? | 131 | (83) | 775 | (85) |
| RIT | 19 | (04) | 19 | (07)9? | 7 | (04) | 45 | (05) |
| Other | 37 | (07) | 35 | (13)17? | 21 | (13) | 93 | (10) |
| Ever testify in court? | ||||||||
| Yes | 295 | (59)58? | 198 | (75)74? | 88 | (54)53? | 581 | (63) |
This latter finding, shown in Table 2, which displays examiners’ views on professional requirements and practices, suggests that most examiners are quite conservative on admissibility, believing that independent review of polygraph results must precede court acceptance. This is true, by the way, even though an overwhelming proportion of examiners (90%) hold the view that CQT polygraph examinations are at least 86% accurate in field settings.
Examiners are almost evenly divided on a college degree as a requirement in the field (47% agree). However, they are very supportive of a required, supervised internship (94%), governmental licensure (87%), required continuing education (97%) and some form of national certification (58%). All of these data are shown in Table 2. It is to be noted that none of these qualifications are now uniform requirements across the USA, although a few are required in some states and in some countries outside the USA.
Training of Examiners
Because of the actual nature of lie detection, the selection, training and regulation of polygraph examiners are critical to the proper development and maintenance of the field, to say nothing, of course, of the value of these items to the quality of the testing that is carried out.
The American Polygraph Association (APA), the major professional organization, currently accredits 13 training facilities. These schools must be at least 10 weeks in length and are required to devote a specified minimum number of hours of instruction to topics such as psychology, physiology and so forth. In addition, a supervised internship may be required after completion of academic instruction. In the most rigorous schools, applicants must undergo a personal interview and pass a polygraph screening examination.
About 40 states in the USA now regulate the activity of polygraph examiners. The lack of uniform regulation, particularly with respect to licensure, training, and educational requirements, is seen by many as a serious problem in the field.
Accuracy of Polygraph Testing
Field practitioners maintain that their accuracy in the field is about 90% and that errors tend to be of the false-negative rather than the false-positive type. (A false-positive error is made when an actually truthful person is found to be ‘deceptive’ during polygraph testing. A false-negative error occurs when a person who actually lied is reported to be ‘truthful’.) A great deal of controversy surrounds these claims because there is no consensus on how to interpret the available research. Another reason is that scientists do not agree on how ‘inconclusive’ outcomes should be considered. Do they represent errors or do they show only that, in some situations, outcomes cannot be evaluated properly and should therefore be ignored?
Table 2 Examiners’ views on selected requirements and practices in polygraphy
| Examiner category | ||||||||
| Item | Police n | (%) | Private n | (%) | Federal n | (%) | Combined n | (%) |
| College degree required? Yes |
183 | (37) | 122 | (47) | 129 | (79) | 434 | (47) |
| Internship required? Yes |
459 | (92) | 244 | (96) | 159 | (98) | 862 | (94) |
| National certification required? Yes |
279 | (56) | 147 | (57) | 106 | (66) | 532 | (58) |
| Licensure required? Yes |
425 | (86) | 241 | (94) | 133 | (82) | 799 | (87) |
| Continuing education required? Yes |
491 | (98) | 243 | (94) | 161 | (99) | 895 | (97) |
| Should polygraph results be admissible evidence? All cases 95 With review 130 Stipulation 214 Not admit 48 |
(19) (26) (44) (10) | 64 66 94 16 | (25) (26) (37) (06) | 26 57 48 26 | (16) (36) (30) (16) | 185 253 356 90 | (20) (28) (39) (10) | |
| How accurate is the CQT in actual cases? 86-95% 96-100% |
238 188 | (49) (38) | 113 116 | (45) (46) | 65 87 | (41) (55) | 416 391 | (46) (44) |
Although it is possible to assess accuracy in controlled settings, such as laboratory mock-crime studies, scientists disagree on whether these results are applicable to actual field settings. There are over 20 such studies now available. In these, the average overall accuracy (excluding inconclusives) is 85%, ranging between 67% and 100%, with about equal false-positive and false-negative rates, each at about 10%. If inconclusive outcomes are counted as errors, the accuracy rate is somewhat lower. These summary statistics are shown in Table 3.
In the field setting it is very difficult to assess the accuracy of polygraph testing. The primary reason for this is that ‘ground truth’ criteria (measures that establish with certainty who is and is not actually telling the truth) are often lacking. This problem, of course, is one that must be confronted in almost all assessments of test outcomes in real life, and it is a difficult one to deal with
Table 3 Average ‘accuracy’ of control question technique in three categories of scientific studies
Laboratory: controlled ‘mock crime’ studies (n=22)
Average overall accuracy = 85% Accuracy range across studies = 67-100% Average error: false positive =12% false negative = 10%
Field: blind review, ‘proficiency’ studies (n=11)
Average overall accuracy = 88% Accuracy range = 64-100%
Average error: false positive = 12% false negative = 06%
Field: accuracy of field decisions in ‘independent criterion’ studies (n=4)
Average overall accuracy = 94% Accuracy range = 87-100%
Average error = Not specified
Research on the accuracy of the CQT in real life has been approached in two different ways. The first of these has been referred to as blind-review or dependent criterion studies. (These are very similar in nature to what is done in proficiency studies in other forensic science research.) The earliest example of this approach with the CQT was reported in 1971. In that study examinations were carried out in a number of independent, actual cases. In each, some examinees ‘passed’ (that is, they were reported to be truthful) and some were found to be ‘deceptive’ to the issue under investigation. The deceptive person in each investigation was questioned after the polygraph testing and each made a full confession, exonerating the other suspects in the case. Thus, in each case the polygraphic data of each suspect were confirmed as being either truthful or deceptive by the confession of the ‘guilty’ person.
In these cases then, the examinees’ truthfulness or deception had been ‘confession verified’. For each, complete polygraphic examination data (records) were available. From these, the records of 20 truthful persons and 20 deceptive persons were selected and each was independently evaluated by ten examiners, none of whom was involved in the actual testing. These examiners were asked to determine which were the records of truthful persons and which of deceptive persons. Overall, the ten examiners produced an average accuracy of 88%.
More than ten generally recognized studies similar to this one have been reported in the literature. While the findings in these studies have varied somewhat, similar to laboratory-based studies, the average ‘accuracy’ has been 88%, as shown in Table 3. In some studies, false-positive rates have been as high as 50%, whereas false-negative rates generally were much lower; on average, these rates are 12% and 6%, respectively. The results from such research are controversial and there is no consensus on what they reveal about accuracy in real-life cases.
Another approach to estimating the accuracy of polygraph testing has involved attempts to establish ground truth by use of criteria independent of the polygraph examinations themselves. For example, in one of the most recent of these studies, researchers attempted to verify the outcome in cases in which polygraph examinations had been carried out by police. They reportedly were able to verify the polygraph examiners’ outcome in 89 instances; in these, the examiners’ decisions were correct 94% of the time. On actually truthful persons the examiners’ decisions were correct 90% of the time; on deceptive persons, decisions were 100% correct. Similar results have been shown in several other attempts to confirm examiners’ decisions in the field by use of criterion measures independent of the examination process. Across these studies the average overall accuracy, as shown in Table 3, has been 94%, ranging between 87% and 100%.
There is also considerable disagreement about which of these independent-criterion studies is of value and, as with the blind-review studies, how best to interpret these findings. One concern, as an example, has to do with how accurately the independent criterion assesses examinees’ actual truthfulness or deception. Thus, in spite of the available scientific data, the controversy over the accuracy of polygraphy continues.
Relative Accuracy of Polygraphy
In spite of the lack of scientific consensus about the accuracy of polygraph testing, it is seldom that the issue is placed in proper perspective. One important question to be asked is not how accurate polygraph testing is in the abstract, but rather how accurate it is relative to other types of evidence and other processes used to accomplish similar objectives. When considered in this light, the evidence is quite different from that usually presented. CQT polygraph testing in controlled conditions, for instance, shows an accuracy that equals or exceeds that of other common means of investigation. In one study it was shown that polygraph testing produced an accuracy that was comparable to results obtained by document examiners and fingerprint analysts, and greatly exceeded that of eyewitnesses. Equally important, polygraph testing was shown to have greater utility, that is, it was useful in more situations than other forms of similar evidence considered.
A comparison of the accuracy of polygraph testing with other common forensic techniques shows similar results. For example, in recent years, in the USA at least, proficiency testing of widely accepted and extensively used crime laboratory techniques has become an accepted method of assessing performance. Generally, in this testing, crime laboratory personnel are asked to analyze in the blind, samples of physical evidence which have been specially prepared for study. Assessments of these samples by practicing crime laboratory specialists are then returned to a central site where they are matched against known standards. In this way, the proportion of agreements and disagreements of the experts’ decisions with the standards can be determined and the proficiency with which certain kinds of physical evidence are analyzed can be specified.
In one of the earliest proficiency studies, carried out in the early 1980s, the results shown in Table 4 were reported. It can be seen in that table that the proportion of laboratories submitting ‘unacceptable responses’ (returns that did not match known standards) varied widely for different types of evidence, from 1.6% for examination of certain fluids to 71.2% for examination of certain blood samples. In other words, common forensic techniques were shown to have widely varying accuracy rates.
The results in more recent proficiency studies, reported in 1995, have shown some improvement in specific areas but there has not been dramatic change overall. For example, Table 5 shows representative findings pertaining to three forensic tests, from which it can be seen that the accuracy with which certain samples of common evidence (blood stains, automobile paint and questioned documents) were correctly identified varied between 64% and 89%; false-positive rates in these tests ranged between 7% and 49%. Although not shown in Table 5, the accuracy range for identification of specific drug samples was between 47% and 100%, and between 85% and 98% for samples of blood and other body fluids.
What such studies demonstrate is that all forensic testing is fallible, that accuracy statistics vary considerably, depending on the technique evaluated, and that polygraphy, when considered in relation to other commonly used forensic techniques, yields comparable and, in some cases, superior accuracy.
As a final point, crime laboratory proficiency studies have reinforced the same lesson that has been made clear in the studies dealing with polygraphy. When tests are carried out by poorly trained examiners, when improper procedures are applied, or when other significant violations of accepted standards are present, the outcomes are less accurate than would otherwise be the case.
Table 4 Percentage of crime laboratories submitting results of ‘unacceptable proficiency’ in various forensic analyses: ordered by percentage of unacceptable responses
| Sample number | Sample type | No. of laboratories | No. of | Laboratories submitting |
| responding with | ‘unacceptable’ | ‘unacceptable’ responses | ||
| data | responses | (%)a | ||
| 13b | Physical fluids (b) | 129 | 2 | 01.6 |
| 6 | Drugs | 181 | 3 | 01.7 |
| 12 | Fibers | 120 | 2 | 01.7 |
| 13a | Physical fluids (a) | 129 | 3 | 02.3 |
| 3 | Blood | 158 | 6 | 03.8 |
| 4 | Glass | 129 | 6 | 04.8 $$$4.7 |
| 7 | Firearms | 132 | 7 | 05.3 |
| 20 | Questioned documents (a) | 74 | 4 | 05.4 |
| 1 | Drugs | 205 | 6 | 07.8 $$$2.9 |
| 21 | Firearms | 88 | 12 | 13.6 |
| 15 | Drugs | 143 | 26 | 18.2 |
| Questioned documents (b) | 14 | 18.9 $$$100.0 | ||
| 5 | Paint | 121 | 24 | 20.5 $$$19.8 |
| 19 | Wood | 65 | 14 | 21.5 |
| 17 | Metal | 68 | 15 | 22.1 |
| 18b | Hair (b) | 90 | 25 | 27.8 |
| 2 | Firearms | 124 | 35 | 28.2 |
| 14 | Arson | 118 | 34 | 28.8 |
| 9 | Glass | 112 | 35 | 31.3 |
| 16 | Paint | 103 | 35 | 34.0 |
| 11 | Soil | 93 | 33 | 35.5 |
| 18e | Hair (e) | 90 | 32 | 35.6 |
| 18 | Hair (a) | 90 | 45 | 50.0 |
| 10 | Paint | 111 | 57 | 51.4 |
| 18c | Hair (c) | 90 | 49 | 54.4 |
| 18d | Hair (d) | 90 | 61 | 67.8 |
| 8 | Blood | 132 | 94 | 71.2 |
Table 5 Correct and incorrect results (%) in three representative forensic tests in recent (1995) crime laboratory proficiency studies
| Evidence type | Correct | Wrong | Inconclusive | False negative | False positive |
| Blood stains | 89 | 6 | 5 | 2.2 | 11.3 |
| Automobile paint | 74 | 23 | 2 | 6.3 | 49.1 |
| Questioned documents | 64 | 3 | 32 | 2.3 | 7.4 |
Legal Status of Polygraph-Testing Results
Although polygraphy is widely used informally on a daily basis in the US justice system, the admissibility of polygraph testing results in court is in a state of flux. This is partly due to two recent decisions by the US Supreme Court, Daubert v. Dow Merrell Pharmaceutical Company and US v. Scheffer. The former decision, the Daubert case in 1993, overturned the 1923 Frye v. US decision, which established the ‘general acceptance’ rule for all scientific evidence, and gave judges greater latitude in deciding which forms of scientific evidence to admit at trial. The more recent 1998 Scheffer decision dealt directly with the issue of admissibility of polygraph testing results in military trials. In this case the court ruled that there was still not ‘scientific consensus’ regarding polygra-phy, and for that reason a presidential directive prohibiting admissibility in military trials was legally permissible.
Both the majority and minority views in the Scheffer case raised some important issues regarding polygraphy. In the former instance, the majority hinted that, in spite of the current scientific controversy about accuracy, when the research data are clearer the court may reconsider admissibility. In the latter instance, it was noted that the government argued against admissibility largely based on a claim of unacceptable accuracy. Yet, it is the government that is one of the major consumers of polygraph examinations, even in the most highly sensitive and complex cases involving national security. This inconsistency in the government’s position suggests that the issue of accuracy is not at the heart of opposition to admissibility of polygraphy. Other issues raised in this case, as well as in other cases in which admissibility has been considered, may be weighted more heavily. One of these, for example, is the belief that the admission of polygraph examination results will produce unnecessary collateral litigation; there will be simply too much court time spent on determination of admissibility in individual cases. Another is that the inclusion of such results in court intrudes on the ‘trier of fact’, the jury. However, several scientific studies, as well as juror polls, show that jurors are not overwhelmed by polygraph examination results.
Moreover, proper court treatment of polygraph testing can ensure that it is given the weight it deserves in a particular case. This is consistent with the role of the court in dealing with other types of scientific evidence.
In spite of the general exclusion of polygraph evidence, there appears to be a growing tendency in federal and state courts for cautious acceptance. In most of these cases, admissibility is predicated on the establishment of rather stringent control over the qualifications and experience of the examiner and the process used in the administration of the examination. One reason for the growing acceptance is that polygraph examinations play an increasingly important role in the justice system. For instance, almost all major police agencies in the USA employ one or more polygraph examiners; they carry out examinations in almost every major criminal investigation and are quite successful, even in those cases in which other means of investigation are either not possible or are quite limited. Prosecutors and defense attorneys routinely make use of polygraph examinations and judges regularly use them, if not formally, then informally, to help decide issues of interest in pretrial, trial and posttrial stages. The evidence suggests that these uses will continue to grow.
Worldwide Use of Polygraphy
The USA has between 3000 and 4000 active examiners. Growth in the field is primarily outside the USA. There are 56 countries in which polygraph testing is now being used; 18 of these, including Canada, Israel, Japan, Romania and Singapore, have very active polygraph testing programs. Training schools have recently been held in Russia, South Africa, Turkey, Mexico, Venezuela and Singapore. In addition, personnel from a wide variety of countries, including China, have been trained in the USA in the past decade or so. These developments will certainly lead to more widespread acceptance and use of polygraphy.
