Can They Fool the Polygraph?

0 Comments | Insight on the News, July 2, 2001 | by Diana Ray

The FBI is requiring employees to undergo increased polygraph testing as a result of security lapses, but cool liars are able to evade detection by the simple machines.

Polygraph testing did not keep Aldrich H. Ames from selling CIA secrets to the Russians before his arrest in 1994. He passed the tests with flying colors while employed by the agency. And testing didn't prevent FBI counterspy Robert Philip Hanssen from reported treason -- because, during his entire 25 years with the bureau, he never was required to take one. Recently indicted, Hanssen is charged with selling secrets to the Soviet Union and then Russia during a period of 16 years, right up to his arrest in February.

After the uncovering of Hanssen, former FBI director William Webster was tasked to investigate FBI security procedures. In the interim, the bureau announced, it would require 500 employees with access to confidential data to undergo polygraph tests during a 60-day period, including Louis Freeh, the bureau director. That was to begin late in March. Although the 60-day deadline wasn't met, only Freeh remained to be polygraphed as Insight went to press, according to FBI spokesman Paul Bresson. Freeh announced he would retire in June after 27 years with the FBI.

Ames wrote a letter last year from the federal penitentiary at which he was imprisoned deriding the polygraph test. "Like most junk science that just won't die because of the usefulness or profit their practitioners enjoy, the polygraph stays with us. Its most obvious use is as a coercive aid to interrogators. It depends upon the overall coerciveness of the setting -- you'll be fired, you won't get the job, you'll be prosecuted, you'll go to prison -- and the credulous fear the device inspires"

The Ames letter was sent to Steven Aftergood at the Federation of American Scientists in response to an article Aftergood wrote for Science magazine about polygraph testing. "It was a rather unexpected piece of correspondence" Aftergood tells Insight.

Bill Franklin, director of the Virginia School of Polygraph, says it's fear of the test that normally causes the guilty to confess. That didn't work with Ames. Bresson of the FBI says, "It's just one tool we use, though at times it has proven to be a more valuable tool than others." Sometimes it has proved critical. "People confess to crimes on polygraphs and we've been able to detect prior drug use during the prescreening process. But, like any forensic tool, it's only a tool."

If a test subject sometimes can deceive a test when guilty, isn't it likely that a truthful response also may be read incorrectly? "Probably some truth to that," Bresson says. "We don't deny that." Which usually is the point at which opponents insist polygraphs are unreliable and invasive. They also claim anyone can learn to fool the test from information available in most libraries and on the Internet. Ames, of course, had other sources.

Drew C. Richardson, a polygraph expert with the FBI, has testified that the test may hold little more diagnostic value than "tea-leaf reading." But the FBI declined to testify about the matter at an April hearing before the Senate Judiciary Committee on the grounds that its own panel had not yet concluded the broad investigation under Webster.

The tests are based on the premise that people who lie will show physiological reactions that can be measured to show deception. The term polygraph means "many writings" -- a reference to the manner in which selected physiological activities are simultaneously and variously recorded. Rubber tubes that record respiratory activity are placed over the test subject's chest and abdominal area. Two small metal plates are attached to fingers to record any sweating, and a blood-pressure cuff or similar device is put onto the subject's upper arm to record cardiovascular activity.

The examinee is asked to sit quietly and very still because the sensitive machine records every movement. The entire environment must be controlled carefully, lest elements other than the questions color response. That's why Franklin doesn't even allow others in the room when he is administering a test and does not do demonstration tests for television or radio talk shows.

According to the American Polygraph Association (ARA), a typical test includes a pretest in which the polygraph examiner completes required paperwork and talks comfortably with the examinee about what is happening. During this period, the examiner will discuss the questions to be asked and familiarize the subject with the testing procedure.

Then the examiner will administer the test, watching a number of polygraph charts, asking questions and recording physiological responses. The results will be analyzed and the examiner may offer the subject an opportunity to explain physiological responses. That was the case when Ames took his polygraphs. According to reports, he was able to explain away unresolved physiological responses. And critics say there is a great disparity from operator to operator in the manner questions are asked and the way results are interpreted. The examiner merely forms an opinion by comparing the physiological reactions to each set of questions.