The polygraph has been used by the police in the investigation of serious crimes since at least the early 1900s. Historical development in the field can be traced along two lines, one involving instrumentation and the other testing techniques.
Polygraph Instrumentation
In 1895 Lombroso, an Italian criminologist, used a hydrosphygmograph and the ”scientific cradle” to measure objectively the physiological changes associated with the detection of deception. Shortly after, an American psychologist, Munsterberg, noted the effect of lying on breathing, cardiovascular activity, and the galvanic skin response (GSR)—apparent changes in electrical resistance in the skin. In 1921 Larson devised an instrument for making continuous recordings of both blood pressure and breathing. In 1930, Keeler, generally credited with developing the prototype of the present-day polygraph, added a device for recording GSR.
Modern computerized polygraphs are technical improvements over earlier devices, although the physiological activities collected are essentially the same. The polygraph captures electrodermal activity (EDA) by means of two electrodes attached to the hand. A standard blood pressure cuff is used to record relative blood pressure and pulse rate. Finally, breathing activity is recorded by ”pneumograph” tubes that expand and contract with chest cavity movement. Activity in each of these physiological systems is usually converted from analog to digital form for display on a computer monitor. The ”chart” display can be stored permanently on standard media for on-line or off-line viewing and analysis.
Testing Techniques
There is no known physiological response that is unique to lying. Neither the polygraph nor any other device is capable of detecting a ”lie.” Lie detection is an inferential process in which ”lying” is inferred from comparisons of physiological responses to questions that are asked during polygraph testing. There are three major families of testing procedures in use today: the relevant/irrelevant technique (R/I), the control question technique (CQT), and information recognition testing (IRT).
In its simplest form the R/I technique consists of asking a series of relevant questions, that is, those pertinent to the crime at hand (for example, ”Did you shoot John Doe?”), along with irrelevant questions that are not crime related (for example, ”Are you over eighteen years of age?”). The test questions are asked several times during the testing. An assumption implicit in the R/I technique is that truthful persons will not react differentially to a great degree to relevant and irrelevant questions, whereas people who are lying will. This assumption has been seriously challenged and is the primary reason why CQT is preferred.
The control question technique was first developed by J. Reid in 1950. In the CQT the question list consists of irrelevant, relevant, and ”control” questions, although other types of questions may also be included. The relevant and irrelevant questions are similar to those asked during R/I testing. Control questions deal with matters similar to, but of presumed lesser significance than, the offense under investigation. The examiner frames these questions so that they will be ”probable lies.”
In the CQT, more consistent and greater physiological responses to relevant questions than to control questions indicate lying on the relevant issues. Conversely, consistently greater physiological responses to control than to relevant questions indicate truthfulness in the matter under investigation.
The IRT is limited to situations in which specific details of a criminal offense are known only to the police and the actual perpetrator(s). A single ”test” consists of the asking of a stem question and multiple options. The stem might be: ”Do you know if John Doe was killed with a_?” The options might be the names of various weapons, for example, gun, knife, club, including the actual weapon used. The guilty person, recognizing the correct option, would be expected to show a greater physiological response to that one than to the others, whereas an innocent person would not. Typically, a series of three or more such multiple-choice tests would be carried out, provided that sufficient detailed information about the offense is known.
Uses of Polygraph Testing
In police work specific issue testing is used to investigate whether a particular person was involved in the commission of a known offense. Polygraph testing has been shown to be extremely valuable in this regard and almost every large police agency employs one or more examiners. It is well established that polygraph testing identifies offenders and “clears” innocent suspects; the result is great savings of investigative time and effort.
Polygraph testing for preemployment screening of applicants for police work is widely used but controversial. Such testing substantiates information collected during traditional background investigations and it uncovers information not otherwise available.
Accuracy of Polygraph Testing
Field practitioners maintain that their overall accuracy is about 90%; errors tend to be “false negatives” rather than “false positives.” (A false-positive error is made when a truthful person is found to be “deceptive” during polygraph testing. A false-negative error occurs when a person who lied is reported to be ”truthful.”) A great deal of controversy surrounds these claims because the scientific research is unclear. A recent review of that research by the National Research Council of the National Academy of Sciences concluded that a reasonable estimate of accuracy is between 85% and 90%, though there were caveats to be heeded.
Training and Regulation of Examiners
There are more than twenty private and governmental schools that serve as training facilities for polygraph examiners. The American Polygraph Association (APA), the major professional organization in the United States, accredits training facilities but only those choosing membership.
About forty states now regulate the activity of polygraph examiners. In some of these, mandatory licensure requirements are set out by statute. In others, the kinds of situations in which polygraph testing may be used are proscribed. The lack of uniform regulation is seen by many as a serious problem in the field.
Legal Status of Polygraph Examinations
The commonly held belief that polygraph examination results are not admitted into courtroom proceedings is not true. The initial judicial decision of Frye v. U.S. (293 F. 1013 [1923]) did exclude polygraph evidence. Today, however, polygraph results are admitted by stipulation. At the federal level there is no single standard governing admissibility; some courts have permitted polygraph evidence whereas others have not.
It is very common for prosecutors to use polygraph results to decide whether and which charges to file. It is also common for judges to use polygraph results in sentencing decisions. In addition, defense attorneys rely on polygraph testing to plan their defense and to negotiate pleas. Thus, polygraphy plays an important role in the justice system quite apart from its use by the police.
