By Nathan J. Gordon, Director, Academy for Scientific Investigative Training, Philadelphia, Pennsylvania
science column written by Sharon Begley for Newsweek titled “Mind Reading Is Now Possible” indicated that functional magnetic resonance imaging (fMRI) would soon be available for law enforcement agencies to detect lies by crime suspects.1 The New York Times ran an article titled “Truth and Justice, by the Blip of a Brain Wave,”2 and the San Francisco Chronicle published “Fib Detector,”3 both making claims about new truth detection techniques.
Since the Vietnam War, U.S.defense related agencies have dedicated significant funds to the development of new lie detection strategies. After the terrorist attacks of September 11, 2001, interest in these strategies has become even stronger, with research funding more available to develop the next generation of lie detectors for use in criminal and terrorist investigations.
Several universities and private companies are trying to develop the new lie detection technologies, using fMRI, electroencephalography (EEG), near-infrared light, and other strategies to access brain function directly.4
Experimental research into lie detection technologies by government and private laboratories will not necessarily translate to an investigative role; many such technologies will never be practical for police use. Despite warnings from researchers, the popular media and other publicly visible parties are attracted to methods or devices that appear to detect deception easily and quickly in an era concerned with terrorist threats and homeland security. Popular television shows continue to depict criminal investigations that are easily resolved through new technology; now, with “mind reading” just over the horizon, the community’s expectations of crime solvers will become even higher.
Police executives need to be familiar with these new developments in truth detection instruments and should be ready to communicate with citizens about truth verification. This article provides police executives with a brief historical overview and highlights current methodologies.
Before 9/11 the whole field of research around deception and credibility assessment was minimal; you could count the research laboratories dedicated to the detection of deception on one hand. Now there are 50 such laboratories in the United States alone undertaking this research. Funding is available, and today it is a race to build better lie detectors.
—Andrew H. Ryan Jr., Ph.D., Past Chair, IACP Psychological Services Committee
Media hype touts the ability of emerging technologies to find terrorists, but for police agencies, the real work of truth finding is in daily criminal investigations. An agency’s choice of equipment built to detect deception can have a direct effect on the administration of justice. Misdirected investigations waste precious budget dollars and manpower as investigators travel in the wrong direction. A successful prosecution may be thwarted by improperly using or interpreting the results of “truth verification” products.
When best-practice standards for detection-of-deception hardware and software are used correctly, truth verification becomes a very effective investigative tool. The general terms used to describe these tests include lie detection, truth verification, detection of deception, polygraph, voice stress, instrumental detection of deception, forensic psychophysiology, and truth tests.
Truth verification instruments are valuable investigative aids used in conjunction with investigations; however, none of today’s instruments are sufficient substitutes for a thorough investigation. Instruments can be used to verify, corroborate, or refute statements; obtain investigative leads; narrow or focus investigations; screen candidates for law enforcement positions; and assist with internal police investigations.
Internally, agencies often use these devices as part of their preemployment hiring process. The best sources of information about applicants are the applicants themselves. Usually applicants will not divulge unflattering or damaging information about themselves when filling out a standard personnel disclosure questionnaire or being interviewed, feeling confident that their “secrets” cannot be discovered. Fortunately, a valid truth verification test can reveal when candidates are holding back detrimental personal information.
Agencies also employ detection-of-deception tests in criminal investigations to exclude suspects or develop information that often leads directly to resolution through a suspect’s confession. In either the background or active case investigation, truth tests can save significant personnel hours and help solve cases that otherwise might go unresolved.
Testing Device History
Instrument-based truth tests have been investigative aids to law enforcement, the military, and the private sector since the first “lie detector” was invented in the 1920s.5
Even before the first instrument was developed, truth verification methodologies were used. In 1895, Cesare Lombroso, considered the father of modern criminology, reported the first instrumental test for detecting deception.6 He placed the hand of a suspect in the sexual assault and murder of a child into a hydrosphygmomanometer, a crude device designed to monitor blood volume changes in the hand by the displacement of water. He then showed the suspect several pictures of children, one of whom was the dead child. Lombroso theorized that if the suspect were innocent, showing the suspect any of the pictures would result in similar minor changes in blood volume, since they were all photos of children the suspect did not know. However, Lombroso reasoned, if the suspect was the killer, viewing a picture of the child he murdered should cause a sudden and significant change in blood volume as the suspect’s sympathetic nervous system responded to the state of fear that the picture induced in him.
In 1908, Harvard professor Hugo Münsterberg wrote On the Witness Stand.7 He devoted an entire chapter in the book to perjury, which he believed usurped the judicial system. He suggested instrumentation be attached to witnesses as they gave testimony to monitor respiration changes, cardiovascular changes, muscle tension, skin temperature, and so on, to ensure the veracity of witnesses’ testimony.
In 1914, Vittorio Benussi, an Italian philosopher and psychologist, published research indicating that breathing changes could give accurate determinations of deception.8 Additional research by William Marston, a student of Münsterberg’s at Harvard, was published in 1917, indicating that systolic blood pressure changes could also accurately detect deception.9
Marston and Benussi’s research piqued the curiosity of Berkeley, California, chief of police August Vollmer (1921 IACP president), which led to the development of the polygraph in 1921 by one of Vollmer’s detectives, John Larson. The Berkeley Police Department became the first law enforcement department in the world to use a polygraph as an investigative tool. Larson’s two-pen instrument continuously monitored respiration and cardiovascular activity changes during questioning.10
In the mid-1930s, Leonarde Keeler, who trained at Berkeley under Larson, added an additional physiological parameter to the polygraph to monitor electrodermal (sweating) activity.11 While today’s polygraph still monitors these three basic physiological functions—blood volume/pressure, respiration, and sweating—polygraph instrumentation has greatly evolved. Pneumatic sensors were replaced with electronic versions, and electronic sensors in turn have been replaced by computerized measurement devices.
For the most part, an intimate understanding of polygraph testing was restricted to those trained to administer polygraph tests and the law enforcement community, but over the last decade this information has begun to reach the public. Now several Web sites offer advice to deceptive examinees on how to “beat” the polygraph using deliberate attempts to alter the data, often by causing physiological changes with body movements. Such “countermeasures” attempt to change what should be scored as a deceptive outcome into a truthful or inconclusive outcome. In response, the polygraph profession has added a fourth parameter to monitor body movements associated with countermeasures. By 2012, the American Polygraph Association will require, as part of the minimum instrumentation requirements for all polygraph examinations, a device that monitors physical body movements and thereby identifies deliberate physical distortions of polygraph results. Agencies that administer polygraph examinations but have not already added this parameter should do so as soon as possible.
Several universities and private companies are trying to develop the next generation of lie detection technologies by using functional magnetic resonance imaging, electroencephalography, near-infrared light, and other strategies to directly access brain function.
Four companies manufacture polygraph instruments; all of them now offer devices that measure the new, fourth parameter. The Lafayette Instrument Company in Indiana (www.lafayetteinstrument.com) and the Stoelting Company in Illinois (www.stoeltingco.com) offer both computerized and analog instruments. Axciton Systems, Incorporated, in Texas (www.axciton.com) and Limestone Technologies, Incorporated, in Canada (www.limestonetech.com) offer only computerized polygraph systems.
Psychological and Voice Stress Analysis
In the 1960s the U.S. military researched the development of covert truth testing. Three officers who headed the project retired from the U.S. Army to open their own private corporation, Dektor Counterintelligence and Security, Incorporated. They developed and began marketing the first instrument designed to measure voice changes that they believed to be directly associated with psychological stress and possible deception. Their first Psychological Stress Evaluator (PSE) was sold in 1971 and was used in Vietnam to differentiate between suspected Viet Cong and civilians.12
Since the development of the PSE, other voice stress devices have evolved from other manufacturers. One device, the Computer Voice Stress Analyzer (CVSA), was originally developed in 1988 by the National Institute for Truth Verification and grew out of the Vietnam-era PSE.
Besides the CVSA, the Truster and Layered Voice Analysis are also found in the marketplace today. All of these devices to a varying degree are employed today by the law enforcement community; however, there is little, if any, viable scientific research to support their accuracy.13
Recent Truth Detection Research
After the terrorist attacks of September 11, 2001, the desire for new lie detection devices increased greatly. One new avenue was thermal imaging, a device that monitors blood flow pooling around the eyes during stress. USA Today reported, “The Defense Department’s Polygraph Institute at Fort Jackson, S.C., is financing at least 20 projects aimed at finding a better lie detector. Another Pentagon office, the Defense Advanced Research Projects Agency, is exploring magnetic resonance imaging (MRI) and other technologies. The FBI and CIA are backing more research.”14
Thermal Imaging: The discovery of thermal imaging as a potential truth detection technology is best explained by Richard Willing:
In 2001, endocrinologist James Levine was performing obesity research by using a thermal-imaging camera that observes how much heat is thrown off when a person chews.
Suddenly, a large screen accidentally fell to the laboratory floor. Levine’s startled test subject yelled. . . . White patches, indicating unusual amounts of heat, had suddenly appeared around the camera’s image of the subject’s eyes.15
Levine wondered whether this sudden heat buildup would occur when subjects felt other forms of stress, such as when attempting deception. The Department of Defense Polygraph Institute (DoDPI—now the Defense Academy for Credibility Assessment [DACA]) was intrigued. They felt that thermal imaging could be performed covertly on unwitting subjects up to 12 feet away, making it potentially useful for checking out travelers seeking to board airliners. The DoDPI tested thermal imaging on 20 soldiers at Fort Jackson. The test properly identified 6 of the 8 subjects (75 percent accuracy) who lied about taking part in a staged crime and 11 of the 12 (92 percent accuracy) who told the truth. One truth-teller was wrongly deemed deceptive. Overall accuracy for the device from this test is estimated at 84 percent.16
fMRI: Research on other physiological functions, such as pupilometry and brain waves, was also undertaken. Most recently, the use of fMRI has been receiving a great deal of attention. Daniel Langleben, a physician and professor at the University of Pennsylvania, and Frank Andrew Kozel, a brain image researcher at the Medical University of South Carolina, received government funding for research into fMRI-based lie detection.
Three major recent research groups are now looking into the possibility of utilizing fMRI imaging to observe differences in brain area activation during truth and deception: Truth Test Technologies, Cephos Corporation, and No Lie MRI, Incorporated. No Lie MRI, associated with Langleben’s work, has already begun offering brain-based lie detector tests under Joel Huizenga, founder of the San Diego–based startup company. The Cephos Corporation of Pepperell, Massachusetts, headed by its president, Steve Laken, is basing its research and development on Kozel’s work. Cephos planned to offer a service similar to No Lie fMRI, using fMRI machines in Charleston, South Carolina, but these tests have not yet begun, as the need for more research became clear.
The author, along with Scott Faro, M.D., vice president of radiology and director of the Functional Brain Imaging Center and Clinical MRI at Temple University School of Medicine in Philadelphia, Pennsylvania; Feroze Mohamed, Ph.D., assistant director of the center; and Steven Platek, Ph.D., formerly with the Department of Psychology, Evolutionary Cognitive Neuroscience Laboratory, at Drexel University in Philadelphia, has formed “Truth Test Technologies” (T3), which has been involved in continuing fMRI research since 2003.
In the February 2006 issue of the journal Radiology, T3 published initial research titled “Brain Mapping of Deception and Truth Telling about an Ecologically Valid Situation: An fMRI and Polygraph Investigation”; a follow-up research paper, “Integrated Zone Comparison Polygraph Accuracy with Scoring Algorithms,” was published in Physiology and Behavior.17 Initial research indicated twice as much brain activity in deceptive subjects as in truthful ones. In this initial research, when the fMRI data were added to the polygraph data, accuracy in detecting truth and deception was above 95 percent. Future research is planned to duplicate the original study; evaluate the use of fMRI testing in multi-issue situations; measure the effects of countermeasures on the fMRI; and measure the effects of justifiable deception on the ability to prevent detection of deception. Of course, a major consideration for this use of fMRI is the cost and the size of the equipment.
Experimental research into lie detection technologies by government and private laboratories will not necessarily translate to an investigative role; many such technologies will never be practical for police use. Despite warnings from researchers, the popular media and other publicly visible parties are attracted to methods or devices that appear to detect deception easily and quickly in an era concerned with terrorist threats and homeland security.
Utility versus Science
With pupil measurements, blood flow around the eyes, voice stress, and classic lie detection in play or in practice, how do law enforcement executives know what to use in their agencies’ efforts to determine truth and advance investigations?
The distinction must be understood between the utility of a device to ascertain information and that device’s ability to make accurate scientific determinations of truth or deception. These two properties must be evaluated to determine the potential value of any truth verification device or system.
Frankly, a lie detection device can have absolutely no scientific ability to determine truth or deception and still be an excellent psychological tool to develop information or extract confessions. For example, if interviewers tell applicants for law enforcement positions that a truth test will be performed after their interviews in order to determine the veracity of the applicants’ statements and self-supplied background, the interviewers will extract more information. This will happen whether or not applicants are actually tested, since they have no idea whether the test has the ability to scientifically and accurately make a determination of truth.
Similarly, a detective or interviewer can attach wires from a photocopier to a suspect’s fingers, as recently portrayed on the HBO television program The Wire:
- The detective tells the suspect that he is attached to a lie detector.
- When the suspect is asked, “Did you shoot John?” he responds, “No.”
- The detective presses the copy button, and a sheet of paper comes out with “LIE!” printed in two-inch-tall letters.
The chances of this detective/interviewer obtaining a confession will increase, even though the copy machine “test” has no ability to make scientific truth or deception determinations. Many would say that this false test worked, in that it succeeded in extracting more information from the suspect during the interview than would have happened without using the device. Similarly, an increased rate of confessions could be expected using such tests after suspects are informed the tests indicate that they lied.
Standard Truth Test
As research continues into new ways to determine truth, the polygraph remains the standard truth test for most law enforcement and government agencies.
The American Polygraph Association (APA), located in Chattanooga, Tennessee, has a compendium of 80 research projects involving 6,380 examinations. Researchers have conducted 12 studies of the validity of field examinations, following 2,174 field examinations, providing an average accuracy of 98 percent. Researchers conducted 11 studies involving the reliability of independent analyses of 1,609 sets of charts from field examinations confirmed by independent evidence, providing an average accuracy of 92 percent.
Ironically, the polygraph possesses the same persuasive powers as all other devices—all tests enable law enforcement agencies to increase information flow and extract confessions. When employed properly by a competent, trained examiner, the polygraph has statistically proven truth/deception accuracy rates of 85 to 95 percent.
Finding a Polygraph Examiner
Where does an agency find properly trained examiners? The APA is the world’s largest accrediting organization. Established in 1966, it has more than 3,200 members, experienced polygraph examiners in the law enforcement community, government, and private business. The APA establishes standards of ethical practices, techniques, instrumentation, and research and provides advanced training and continuing education programs.
To become APA members, examiners must adhere to strict guidelines and educational requirements that include hundreds of hours of coursework at an APA-approved school as well as an internship. APA members must attend annual continuing education and training to maintain their membership. The association has accredited 20 polygraph schools, 15 of which are in the United States. Each APA-accredited school is inspected regularly and rigorously to ensure that its curricula meet the highest standards and offer the most current, fundamental polygraph techniques, and those schools are audited regularly to ensure that their graduates have sufficient knowledge to conduct examinations professionally. Through strict adherence to training and education standards, APA examiners are able to attain accuracy rates exceeding 90 percent.
The association is governed by a board of 11 officers who are elected by the members. Through this board, the APA represents industry segments that include government, law enforcement, private practice, education, and the public at large.
The American Association of Police Polygraphists (AAPP), located in Waynesville, Ohio, certifies operators. The AAPP recognizes 16 polygraph schools in the United States and five schools in other countries but does not accredit polygraph schools.
Past performance and experience are important considerations in selecting polygraph examiners, but these factors should not be considered by themselves. Whether an agency employs its own polygraph examiners or outsources its polygraph needs, it should be choosing examiners certified by the APA and, ideally, the AAPP as well. Agencies should remember that length of experience may not be a sufficient criterion by itself, since quantity does not equal quality.
Today, after a proper pretest interview and correctly using proper instrumentation, trained polygraph examiners score and interpret their results using one of several acceptable manual methods of data analysis, along with six computerized polygraph algorithms to perform quality control of the manual scoring decisions made by the polygraph examiner. The Lafayette Computerized Polygraph System offers five algorithms: PolyScore, Objective Scoring System (OSS), Identifi, Quest, and the Academy for Scientific Investigative Training’s PolySuite, also available from the Academy (www.polygraph-training.com); Stoelting offers its own algorithm, called CPS; Axciton offers its own algorithm, called White Star; and Limestone offers the OSS.
Law enforcement executives should also know that there are approximately 14 major polygraph testing formats used in the United States today by APA- and AAPP-trained polygraph examiners. Some formats date from the polygraph’s origin in the 1920s. The Zone Comparison Test (ZCT) format originated in the work of Cleve Backster, a famed polygraph instructor and innovator; this format has shown to produce the highest overall accuracy rates. The ZCT format can be constructed for single-issue, multifaceted, or multi-issue examinations.
Single-issue and multifaceted tests are most likely used in conducting criminal examinations, but they can also be used for preemployment and other purposes. Suspects, targets, and subjects with presumed direct or tangential roles in criminal cases receive these examinations.
A single-issue test focuses on one relevant aspect about a crime that is posed in several different ways. For example, a suspect might be asked all of the following questions: “Did you shoot John?” “Regarding John, did you shoot him?” “Yesterday, did you shoot John?”
The multifaceted test contains relevant questions dealing with more than one aspect of the same crime. For example: “Did you plan with anyone to have John shot?” “Did you shoot John?” “Do you know for sure who shot John?”
A multi-issue test contains relevant questions about totally different issues. For example, a multi-issue preemployment examination could contain relevant questions dealing with an applicant’s past honesty, criminal activity, and drug involvement.
Single-issue tests provide the greatest opportunity for accurate truth verification determinations. Scoring multifaceted and multi-issue tests is more difficult because they contain queries relating to more than one relevant issue, creating a less focused examination. When suspects/applicants react in a questionable manner to queries in these types of tests, a “hurdle approach” is often used: a more accurate single-issue examination tests the specific area of possible deception.
Law enforcement agencies should have a published policy on the use of any deception detection equipment and a standard operating procedure clearly delineating who should be tested and when such tests should be conducted. This is important because some states regulate or prohibit testing of victims in certain crimes, such as rape. In some jurisdictions, verification of victim statements for any crimes is not permissible under state law. There also are federal laws that affect polygraph testing. The Americans with Disabilities Act (ADA) affects preemployment testing and limits medical information that job applicants need to provide prior to receiving a job offer. The federal Employee Polygraph Protection Act has no effect on the legitimate use of polygraphs by law enforcement agencies in criminal investigations, but if law enforcement examiners or their agencies release polygraph information to an employer concerning an employee, each violation could bring a $10,000 fine to the employer.
A model policy on polygraph use is available from the IACP. To obtain a copy, readers can simply call the IACP Policy Center at 1-800-THE-IACP.
From the beginning of its existence, the human species has had deviant members. Similarly, there has always been a search for truth. The law enforcement community, as the protector of society, often spearheads that search. When used properly, truth tests can help agencies hire the best applicants, save hundreds of investigative work hours by narrowing lists of suspects under investigation, and obtain a confession in a scenario where there is little leverage. Law enforcement executives who have a basic understanding of truth tests, including their strengths and weaknesses, will be in the best position to make sound decisions concerning how best to use these systems and how best to allocate training dollars and equipment purchases. Police leaders should remember that all truth tests are not created equal—that is the truth. ■
Nathan J. Gordon is founder and director of the Academy for Scientific Investigative Training. In 2007 and 2008, he was elected vice president private of the APA and president of the Pennsylvania Polygraph Examiners’ Association. In nearly 40 years of criminal, retail, and industrial investigation, Gordon has personally conducted over 11,000 polygraph examinations, resolving many hundreds of cases, including numerous homicides, by obtaining confessions. A coauthor of Effective Interviewing and Interrogation Techniques, he has published numerous research papers and articles on forensic psychophysiology, investigation, and security. Gordon is the innovator of the Integrated Zone of Comparison Polygraph Technique, the Horizontal Scoring System, and the Academy for Scientific Investigative Training algorithm for manual chart analysis.
1Sharon Begley, “Mind Reading Is Now Possible,” Newsweek, January 21, 2008.
2B. J. Feder, “Truth and Justice, by the Blip of a Brain Wave,” New York Times, October 9, 2001.
3C. T. Hall, “Fib Detector,” San Francisco Chronicle, November 26, 2001.
4Paul Root Wolpe, Kenneth R. Foster, and Daniel D. Langleben, “Emerging Neurotechnologies for Lie-Detection: Promises and Perils,” American Journal of Bioethics 5, no. 2 (March–April 2005): 39.
5The first workable polygraph is attributed to John Larson (1892–1983) in 1921. Its use spread relatively quickly in policing circles, and since then it has been improved a number of times by such means as the introduction of computer scoring. See Charles R. Swanson et al., eds., Criminal Investigation, 10th ed. (New York: McGraw-Hill, forthcoming), 170.
6Despite the unscientific nature of his theories, Lombroso (1835–1909) was highly influential in introducing the importance of scientific studies of the criminal mind. Among his books are L’Uomo Delinquente [The criminal man] (1876) and Le Crime, Causes et Remèdes [Crime, its causes and remedies] (1899). See http://www.cerebromente.org.br/n01/frenolog/lombroso.htm (accessed July 3, 2008).
7Hugo Münsterberg, On the Witness Stand: Essays on Psychology and Crime (New York: McClure, 1908), http://psychclassics.yorku.ca/Munster/Witness/index.htm (accessed July 7, 2008).
8Vittorio Benussi, “Gesetze der inadäquaten Gestalterfassung,” Archiv für die gesamte Psychologie 32 (1914): 50–57; “Die Atmungssymptom der Lüge,” Archiv für die gesamte Psychologie 31 (1914): 513–542; “Versuche zur Bestimmung der Gestaltzeit,” Bericht über den 6. Kongress für experimentelle Psychologie Göttingen (Leipzig: Barth, 1914).
9William M. Marston, “Systolic Blood Pressure Symptoms of Deception,” Journal of Experimental Psychology 2 (April 1917): 117–163. 10See William G. Bailey, ed., “Polygraph,” in The Encyclopedia of Police Science (New York: Garland, 1989), 507.
12Swanson et al., Criminal Investigation, 171.
13See Harry Hollien and James D. Harnsberger, “Voice Stress Analyzer Instrumentation Evaluation,” Final Report for Department of Defense Counterintelligence Field Activity Contract FA 4814-04-0011, March 17, 2006, http://www.clas.ufl.edu/users/jharns/Research%20Projects/UF_Report_03_17_2006.pdf; and Kelly R. Damphousse, “Voice Stress Analysis: Only 15 Percent of Lies about Drug Use Detected in Field Test,” NIJ Journal, no. 259 (March 2008), http://www.ojp.usdoj.gov/nij/journals/259/voice-stress-analysis.htm (both accessed August 1, 2008).
14Richard Willing, “Terrorism Lends Urgency to Hunt for Better Lie Detector,” USA Today, November 4, 2003, http://www.usatoday.com/tech/news/techpolicy/2003-11-04-lie-detect-tech_x.htm (accessed July 7, 2008).
17Feroze B. Mohamed et al., “Brain Mapping of Deception and Truth Telling about an Ecologically Valid Situation: Functional MR Imaging and Polygraph Investigation—Initial Experience,” Radiology 238 (June 2006): 679–688, http://radiology.rsnajnls.org/cgi/content/full/238/2/679 (accessed July 7, 2008); Nathan J. Gordon et al., “Integrated Zone Comparison Polygraph Technique Accuracy with Scoring Algorithms,” Physiology & Behavior 87 (February 2006): 251–254.