Scientific Lie Detection

Posted by permission of Dr. Thomas O’Connor. Copyright by T. O’Connor.

“A lie told often enough becomes the truth.” Vladimir Lenin

Perhaps the best known method of deception detection is the polygraph technique, often inaccurately referred to as the “liedetector.” However, other methods exist, such as — hypnosis, narcoanalysis(“truth serum”), PSE (psychological stress evaluation), and P300 brainwave “fingerprinting.” Althoughthese other methods will be discussed briefly, most of this lecture focuses onthe polygraph.

HISTORY OF THE POLYGRAPH

Lombroso, the founding father of criminology in 1895, was the first to experiment with a machine measuring blood pressure and pulse to record the honesty of criminals. He called it a hydrosphygmograph. A similar device was used by Harvard psychologist William Marston during World War I in espionage cases, who brought the technique into American court systems. In 1921, John Larson added the item of respiration rate, and by 1939, Leonard Keeler, one of the founding fathers of forensic science, added skin conductance and an amplifier, thus signaling the birth of the “polygraph” as we know it today.

The polygraph is a sound and reliable technique for detectingdeception. The vast majority of studies into the reliability of polygraphtesting estimate it at least at 90% or higher. Numerous research findings, andworks in the field of medicine, have justified the connection betweeninvoluntary (sympathetic nervous system) physiological changes and emotionalstates related to truth-telling or deception. Unfortunately, mostlythrough historical accident, polygraph exams are not legally admissible unlessthere is a stipulated agreement prior to trial. There’s also some questionablevalidity (if something measures what it purports to measure) in the use ofpolygraph for noncriminal purposes, such as preemployment screening, drugtesting, and so forth. Ironically, polygraphs are more commonly found inthese noncriminal, civil law areas. Police departments, for example, makeextensive use of them in their personnel policies, and they are also common withsensitive, security jobs in government and business. Their use in these contextsignores the fact that these were not what the tests were designed for(criminological positivism), and instead, substitutes the Teddy Roosevelt ideathat a public safety employee has nothing to hide anyway (under color of law orpresumption of innocence). Privacy rights and protections againstself-incrimination are much less guaranteed in civil cases or civil matters.

It must be remembered that this is the area that gave us the Fryetest for admissibility (Frye v. United States 1923). It’s worth lookingat the language of Frye in some detail:

Just when a scientific principle or discovery crosses the line between experimental and demonstrable is difficult to define. Some in this twilight zone, the evidential force of the principle must be recognized, and while courts will go a long way in admitting expert testimony deduced from a well-recognized scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field in which it belongs.

The court in Frye decided that the polygraph belongedin the fields of physiology and psychology,but at the time (and to this day still), neither field claims the technique. Aninterdisciplinary field called “psychophysiology” emerged during the1970s, but it consists largely of academics with no practical field experiencein dealing with criminal suspects. On the whole, their research, whilemethodologically sound and statistically sophisticated, tends to involve studiesin the truthfulness of college students (convenience sampling). The field ofpsychology tends to treat the polygraph as another instance of psychometrictests, and one of the most fundamental principles of psychometry is that eachand every subject should get the exact same questions in exactly the same way.All of this flies in the face of polygraph examiners who argue, understandably,that in the hands of a “master”, with experience in dealing withcriminals and flexibility in being able to condition and question suspectsdifferently, the polygraph technique is highly valid and reliable.

As it stands now in 2000 (and there has been numerous,conflicting case law since 1923, but no definitive Supreme Court ruling),polygraph tests in the vast majority of jurisdictions follow a stare decisis(let the precedent stand) pattern in which they are not admissible (per seinadmissibility) at any criminal trial unless there has been a prior stipulationby both prosecution and defense to agree to submit the results of suchexamination. The logical outcome of this is that both sides should agree on theundisputed expertise of a single examiner, or that the prosecution should useits own examiner and the defense should use its examiner, but the power tochoose rests with the defense, a phenomenon referred to as the “friendlyexaminer” syndrome. Studies have not borne out the truth of thissyndrome, but it theoretically refers to the idea that a deceitful suspect whotakes a polygraph voluntarily at the request and arrangement of their ownattorney will stand a better chance of appearing truthful.

The law is even more complicated than per se inadmissibility.Because of stare decisis and the Fifth Amendment, the state cannotrequire a criminal suspect to submit to a polygraph, even under some pretrialstipulated agreement. If there is a state-sponsored examination, it must bevoluntarily taken, and then if the results are negative, they are generallyinadmissible because the test was given under coercive circumstances. The factof the matter is that only positive test results ever get seen at trial. Forthis reason, lawmakers in about half the states (including North Carolinaespecially) have decided to NOT allow polygraph evidence to be admitted incriminal cases at all, for any reason, stipulated agreement or otherwise. Asmall handful of other states (most notably New Mexico) have gone the oppositedirection, permitting polygraph results in criminal cases, pursuant tostipulation and over objection, especially if the test results were unfavorableto the defendant.

HOW LIE DETECTION WORKS

Lie detection methods have been used for years by policeinterrogators. Physiologically, when a suspect lies about their involvement incrime, it’s fairly easy to notice a flushed face, throbbing of the carotidartery, dryness of the mouth, sundry other clues. Psychologically, verbally, andnonverbally, there are other clues and cues. The assumptionbehind all lie detection methods is that there’s a natural interaction betweenmind and body, and depending upon the individual and their level of involvement,deceptive suspects will utilize certain mental, emotional, and physical defensemechanisms that are dependent upon the amount of stress they’re under or whatdanger they perceive themselves to be into. Now, that’s a big assumption,and the phrase “defense mechanisms” might be better called the”psychological set” to rule out any idea that the technique ispsychoanalytically grounded, which it is not. Polygraph exams are believed tooffer individual, rather than class, evidence because through the years,developmentally, a person develops set ways of reacting to reacting to stressfulor threatening situations. During a polygraph, an examiner is always payingattention to these fundamental clues and cues, developing a sense of thesuspect’s values, beliefs, motives, and attitudes.

The machine part of a polygraph examination is designed topay attention to the actions of the nervous system, particularly the autonomicnervous system, and then certain sympathetic members of the autonomic systemwhich alert the body to stress or threatening situations. The machine hascomponents that measure the following:

• respiration (pneumograph — pneumatic tubes, assisted by beaded chains, are fastened around the chest and abdomen of the person)

• electrodermal skin response (galvanometer — two electrodes are affixed to two fingers on the same hand, and an imperceptible amount of electricity is run through them)

• blood volume and pulse rate (cardiospymograph — a blood pressure cuff, of the type used by physicians, is fastened around the upper arm)

The machine is not just operated. The examiner should be aperson of ability, experience, education, intelligence, and integrity who usesthe machine in a predetermined manner. There are three (3) phases of the testprocedure: (1) a pretest interview; (2) chart recording; and (3) diagnosis.Beforehand, the examiner is provided with all relevant information regarding thecase, such as the criminal charges against the person and the statement offacts. They then spend some time alone preparing a pool of test questions thatare neither too broad nor too specific. Anything calling for an opinion orbelief that can change with time or motivation is ruled out as a possible testquestion, as is anything vague. The pool of questions should focus on a singleincident, the facts, and narrowly defined issues of disputed action, notintention.

During the pretest interview, the examiner will condition thesubject by clarifying the purpose of the test, reassuring them about itsobjectivity, and/or defining terms that will be used. Also, a controlquestion will be developed and selected. A control question is unrelatedto any legal issue, but it addresses a related behavior. For example, with acrime of violence, a control question might be “Have you ever lost yourquestion or done things you regret?” Relevant questionsare those that have a direct bearing on the case, and irrelevantquestions have no bearing whatsoever, but can only be answeringtruthfully (“Are you sitting in a four-legged chair right now?”).

Generally, a series of 9-10 prepared questions are asked,allowing about 10 seconds following an irrelevant question and 15-20 secondsfollowing a relevant or control question. It’s also standard to run through allquestions a minimum of three times before a diagnosis is attempted.

Diagnosis is made by verifying other clues and cues with thechart. A truthful subject’s chart will show emotional attention was paidtoward the control questions and deflected away from the relevant questions. Adeceptive subject’s chart will show emotional attention directed toward relevantquestions and away from control questions.

HYPNOSIS AS A METHOD OF LIE DETECTION

Only when an accused person or eyewitness suffers from aclear case of amnesia will the court consider the use of hypnosis. Defensecounsel usually calls for it (in the case of suspects), and is under anobligation to have the results or psychiatric testimony corroborated by anindependent third party. Hypnotically induced testimony is inadmissible incourt, and hypnotically refreshed or recollected memories are highlycontroversial.

In a 5-4 decision, the Supreme Court in Rock v. Arkansas(1987), ruled that there should be no per se inadmissible approaches tohypnotically refreshed testimony, that these constitute arbitrary restrictionsand prohibit a person from being able to testify in their own defense. Moststates now follow a case-by-case approach, and controversy is the norm in thescientific literature.

It is far more common to see hypnosis used in the pre-indictment state of criminal justice, where the police have yet to zero in on a suspect. Eyewitnesses are often involved here. The most popular methods involve past-memory regression and memory enhancement. The first method is usually designed to “unblock” something that is preventing the eyewitness from remembering, and the second method is designed to better “see” some detail, such as a license plate number. The second method is more controversial than the first, and is reminiscent of the psychic technique of “clairvoyance,” which ironically was involved back in 1845 when hypnosis was first used in a court of law. Hypnosis, when used to relax, or “unblock” the subject’s ability to “watch” the event as it happened has the potential to be of great value to criminal investigation, and is consistent with Dr. Martin Reiser’s retrieval theory of hypnosis. The retrieval theory of hypnosis stands in stark opposition, however, to Dr. Martin Orne’sconstruction theory of hypnosis, which states that there are all sorts of distracting situational factors to events which are stored in memory.

NARCO-HYPNOSIS AND NARCOANALYSIS (TRUTH SERUM)

Controversy and disagreement reign over the use of variousdrugs — scopolamine, sodium amytal, and sodium pentothal — to reliably producetruthfulness. All such drugs inhibit control of the nervous system andreduce inhibitions. All courts refuse to admit truth serum evidence, the leadingcase coming from the New Jersey Supreme Court, State v. Pitts (1989),which disallowed the results of a sodium amytal interview, ruling that it is nota valid scientific technique. The problems with drug-induced memory recall illustrate some of the problems with hypnosis in general, and some of those problems are as follows:

• hypermnesia or confabulation, where the subject fills in gaps with false material

• hypnotic recall, which is a term meaning something felt or thought at the time of undergoing the drug or hypnotic treatment gets “retroactively” integrated with the original material of memory

• memory hardening, where false confidence becomes attached to a bad memory simply because of the extensiveness of the drug or hypnotic procedures carried out

PSYCHOLOGICAL STRESS EVALUATION

PSE, sometimes called voice stress analysis, is based on theuse of a certain machine developed in the late 1960s that presumably detects”guilt-revealing,” laryngeal microtremors which exist in the voice andare associated with stress and lying. One of the assumptions that PSE makes is in regard to the “startle reaction” which scientists haven’t yet agreed upon as being a reflex or an emotion (Ekman & Rosenberg 1998). Most research has produced negative ormixed findings of a relationship between microtremors and deception. Like thepolygraph, the key to success is the preparation of prepared test questions. Ahandful of states, like Arkansas and Louisiana, have licensing requirements forPSE evaluators, but it’s generally conceded that PSE is not admissible in civilor criminal cases and only useful in investigative settings.

Other methods of PSE involve simple psycholinguistics, spectrographic voice recognition, or the identification of verbal and nonverbal behaviors which signal deception. The following is a short list of those indicators:

• language that deflects away from self; increased pitch; overgeneralizations

• speech hesitations and pauses; thinking through what to say; no apparent spontaneity

• increase in the number of shrugs; leg and foot movements; eye blinking, nervous stroking

• hyperventilation, breath holding, sighing, flushed appearance

• reduced use of hand gestures; reduced use of visual or sensory images

• other unusual nonverbal behavior; other inconsistent verbal behavior

BRAIN WAVE FINGERPRINTING

Neuroscientist Lawrence Farwell, who runs a Brain WaveInstitute in Fairfield, Iowa patented this technique in 1995, which hasattracted the attention of the FBI and CIA as a better way to detect moles. Iowa judges are also fond ofadmitting the technique, even though Iowa is a state where the polygraph isoutlawed. The basic principle is that different regions of the brain light upwhen people tell the truth or lie, and further, that different regions areactivated depending upon the type of lie. Dr. Farwell’s research, however, looksat a specific type of electrical brain wave, called P300, which activates when aperson sees a familiar object. For example, if a murder suspect is claiming analibi, then their P300 wave won’t activate when they are shown the murderweapon. The technology is promising in that the research indicates the brainstores visual images. In the P300 test, a subject wears a headband of electrodesand faces a computer screen. In similar tests, a subject wears a helmet ofelectrodes, and experts try to make interpretations from a record of what areasof the brain “light up” or receive intensive blood flow. The technique doesn’t have anything to do with emotions, or whether a person is sweating or not; it simply detects scientifically whether certain information is stored in the brain.

Since 1995, brain fingerprinting has been extensively tested by the FBI. For example, FBI agents who would only know certain key words or phrases associated with a crime were tested, and the results had 100% accuracy in distinguishing agents who knew about the crime from those who didn’t. Sometime around 2003, brain fingerprinting became admissible in court for use in identifying or exonerating individuals in the U.S. A 2004 case, involving the exoneration of a convicted killer in Oklahoma was a critical test case for the technique.

REFERENCES
Burke, T. (1999). “Brain Fingerprinting: Latest Tool for Law Enforcement.” Law & Order 47(6): 28-31.

Ekman, P. & Rosenberg, E. (Eds.) (1998). What the Face Reveals. NY: Oxford Univ. Press.

Hollien, H., L. Geison & J. Hicks. (1987). “VoiceStress Evaluators and Lie Detection” Journal of Forensic Sciences32(2):405-18.

Hollien, H. (1990). The Acoustics of Crime: The New Science of Forensic Phonetics. NY: Springer.

MacDonald, J. (1955). “Truth Serum” Journal of Criminal Law &Criminology 46:259-69.

Matte, J. (1996). Forensic Psychophysiology Using the Polygraph. NY: JAM Publications.

Moenssens, A., J. Starrs, C. Henderson & F. Inbau. (1995). ScientificEvidence in Civil and Criminal Cases. Westbury, NY: Foundation Press.

Moore, M., Petrie, C. & Braga, A. (Eds.) (2003). The Polygraph and Lie Detection. NY: National Academies Press.

Nachshon, I. et al. (1985). “Validity of the Psychological StressEvaluation” Journal of Police Science and Administration 13:275-35.

Nardini, W. (1987). “The Polygraph Technique: An Overview” Journalof Police Science and Administration 15:239-49.

Orlansky, J. (1962). An Assessment of Lie Detection Capability. WashingtonD.C.: Institute for Defense Analysis.

Orne, M. (1984). “Hypnotically Induced Testimony” in E. Loftus (ed.) EyewitnessTestimony. NY: Free.

Reid, J. & F. Inbau. (1977). Truth and Deception: The Polygraph Technique.Baltimore: Williams & Wilkins.

Tanner, D. & Tanner, M. (2004). Forensic Aspects of Speech Patterns: Voice Prints, Speaker Profiling, Lie and Intoxication Detection. NY: Lawyers and Judges Publishing Co.

Vrij, A. (2000). Detecting Lies and Deceit: The Psychology of Lying and the Implications for Professional Practice. NY: Wiley.

Weinrich, H. (2006). The Linguistics of Lying. Seattle: Univ. of Washington Press.

Online Resources