Polygraphs and lie detectors

"The limits of the polygraph"
Faigman, David L, Issues in Science in Technology, Fall 2003;

http://findarticles.com/p/articles/mi_qa3622/is_200310/ai_n9343906/print

Polygraph test, lie detectors are given more credit than they deserve in television and even by technicians, and politicians. CSI and similar shows lead us, the public, to believe in the validity and accuracy of many forensics tools. The truth is that these highly toted tools are actually a lot less reliable than purported. A review article in the Fall 2003 issue of Issues in Science and Technology states that many courts at all levels are precluding many identification techniques on the bases that they are not found rigorous scientific evidence. With exception to the scientifically seasoned DNA analysis, “Fingerprinting, fiber analysis, hair analysis, ballistics, bite marks, etc…” have all come under questions for lack of scientific foundation.

In the same issue of Issues in Science and Technology, author David Faigman, looks at the use of polygraphs in law enforcement and national security. As a decisive tool, the polygraph is limited to questions that demand a straightforward answer. For example, “Did you see the victim on Sunday?” This questions is black and white and thus there is no confusion, the polygraph is very accurate in these situations However for screening, such as detecting a spy or a member of a terrorist organization, questions may no revolve around a specific event. And so the questions are generic. “Did you ever reveal classified information to and unauthorized personnel?” It may be unclear to an examinee whether a particular activities “justifies” a yes answer, so he may feel like he’s lying even though he’s providing a truthful response.

Allegations of espionage by Wen Ho Lee, a nuclear scientist at Department of Energy’s Los Alamos National laboratory, cause many of his colleagues to be subsequently subjected to polygraph test. This sparked Congress to ask the National Research Council to study the polygraph testing’s ability. To distinguish accurately between lying and truth-telling across variety of setting and examinee, even in the face of countermeasures.
1st they examined the scientific bases of underlying the physiological measures of the polygraph. Despite its ability to detect psychological states associated with deception, similar results were obtained for people merely anxious about being tested.

The committee collected all available published and unpublished data on the accuracy of polygraph tests. They found the quality of studies to be low in general. Furthermore they found testing methods to be inherently flawed. Laboratory studies suffered from lack of realism: less variation in test implementation, in characteristics of examinees and in the nature and context. Generally, the limited scopes of these experiments lead to overestimates in the accuracy of polygraph tests. Field studies are limited by the difficulty in determining the truth against which test results could be compared. Moreover, they were not conducted in a double-blind control study. Examiners had prior knowledge of the case and often the cases outcome would be affected by the examination. Biases inherent to these studies made it difficult to obtain the objective truth.

Hypothetical studies were conducted using computer for security screening applications and crimination investigation. Polygraph tests were conducted in two different modes: Suspicious (more sensitive) and friendly(less sensitive). In both security screening and criminal investigation, suspicious mode yields more positive results, more guilty/spies but it also created more false positives (innocent who failed the test). In friendly, few innocent people were wrongfully accused but the large majority of guilty/spies would pass the test.

Despite the inaccuracy of polygraph tests and high chance of errors, it is still used as a investigative tool on grounds that it produces accurate results, and the it deters people on the premise that they believe that deception may be revealed by the test.

Faigman also discusses the potential issue of polygraph tests use in post-conviction sex-offender maintenance. In 30 states, as part of their probation program sex offenders must submit them to periodic polygraph tests. Examinees are often asked questions surrounding “sexually deviant” or “high risk” behavior, such as the use of alcohol or drugs, sexually activity with a consenting adult, or “masturbation to deviant fantasy,” rather than on the detection of actual sex crimes or other violations of the terms of parole. This testing is based on presumption that legal but “undesirable behaviors are indicators of illegal activity. Faigman says that there is no evidence that a “failed” polygraph test is an actual indicator of concealed sex crimes and that there is no scientific bases for using the polygraph as a management of treatment tool for sexual offenders.

The polygraph test is not very accurate that much is clear. It often works on the threat that it will reveal the truth. False positives and false negative are serious risks which make polygraphs unreliable. From a scientific standpoint polygraph holds little ground. They should as Faigman says be used with great caution, if at all in courts.

However, the current stance of the government is can we afford not to be cautious or not use the polygraphs. Even if a number of innocent are found guilty is not worth it to find that 1 person who might cause massive destruction or is not worth it to be more cautious to a former sex offender. In such cases, most people I feel would rather on the side of caution. While the prosecution, courts, and the government continue to use the polygraph test, new more accurate technology which is base on rigorous scientific testing must be developed.