Myth: Lie Detectors Are Reliable. Here Is What the Science Actually Shows.

In 1988, the United States government passed a law making it illegal for most private employers to use polygraph tests in hiring decisions. The reason was not bureaucratic overcaution. It was that the scientific evidence for polygraph reliability was, and remains, deeply insufficient. Decades later, the same technology is still being used in criminal investigations, national security screenings, and sex offender monitoring programs, while courts in most jurisdictions refuse to admit polygraph results as evidence precisely because the science does not support them.

This is one of the most consequential gaps between what the public believes and what the research shows. And in a legal system where a failed polygraph can destroy a person's reputation, career, or freedom before they ever set foot in a courtroom, it matters enormously.

The myth

Most people believe that lie detectors work. The name helps. A machine that detects lies sounds authoritative, scientific, and precise. Television and film have reinforced this impression for decades, the polygraph as the final arbiter, the moment of truth, the thing that catches the guilty and clears the innocent.

The reality is considerably less dramatic.

What a polygraph actually measures

A polygraph does not detect lies. It detects physiological arousal, changes in heart rate, blood pressure, respiratory rate, and skin conductance that occur when a person is stressed or anxious. The underlying assumption is that lying causes stress, stress causes arousal, and arousal shows up in the measurements.

This assumption has three serious problems.

First, many things cause physiological arousal that have nothing to do with lying. Anxiety, fear of being disbelieved, anger at being accused, and the general stress of being connected to a machine and interrogated by a stranger all produce the same physiological responses that polygraph examiners interpret as signs of deception. An innocent person who is terrified of failing the test may produce results that look identical to those of a guilty person who is calm and practiced. [1]

Second, some people lie without significant arousal. Individuals with psychopathic traits, people who genuinely believe their own false narrative, and people who have practiced controlling their physiological responses can produce results that appear truthful. The polygraph has no way to distinguish between a person who is telling the truth and a person who is lying calmly. [2]

Third, the scoring of polygraph results is not objective. Examiners exercise significant subjective judgment in interpreting the charts. Two examiners looking at the same results can reach different conclusions. This is not a feature of a reliable scientific instrument. It is a feature of a process that looks scientific while depending heavily on human interpretation. [3]

What the research actually shows

In 2003, the National Academy of Sciences published the most comprehensive review of polygraph research ever conducted, commissioned by the United States Department of Energy. The conclusion was unambiguous. The report found that the scientific evidence on polygraph accuracy was far weaker than commonly believed, that polygraph tests produce erroneous results at rates that would be unacceptable in any other scientific or legal context, and that the technology rests on theoretical foundations that the research does not support. [4]

Estimates of polygraph accuracy in the research literature range widely but cluster around 70 to 90 percent under ideal conditions, and real-world conditions are never ideal. A 70 percent accurate test in a population where most people are telling the truth will generate an enormous number of false positives: innocent people flagged as deceptive. Applied to national security screenings where millions of people are tested, the mathematics produce thousands of incorrectly accused innocents for every genuine spy caught. [4]

The case of Aldrich Ames makes this concrete. Ames was a CIA officer who spied for the Soviet Union for nine years. He passed two polygraph examinations during that period while actively committing espionage. His success at defeating the polygraph was not exceptional. He simply remained calm. [5]

Why it still matters in court and beyond

Federal courts and most state courts do not admit polygraph results as evidence under the Daubert standard, which requires that scientific evidence be based on reliable, tested methodology with known error rates. The polygraph fails this standard. [6]

But polygraphs continue to influence legal outcomes in ways that fall short of formal admission. A failed polygraph can trigger further investigation, damage a suspect's credibility with law enforcement, and lead to coercive interrogation tactics that produce false confessions, the very outcome the test was supposed to prevent. Sex offender supervision programs in many states require regular polygraph testing as a condition of parole, meaning that a person's freedom can depend on a technology the scientific community has largely rejected. [1]

And the damage to innocent people is not hypothetical. Brandon Darby, Jeffrey Deskovic, and numerous others have documented the role polygraph failures played in directing law enforcement attention toward innocent people and away from actual perpetrators. In each case the machine did not detect deception. It detected anxiety, and then someone decided that anxiety meant guilt. [5]

The bottom line

A polygraph is not a lie detector. It is a stress detector administered by a person who interprets the results. The interpretation is subjective, the technology is theoretically flawed, and the scientific consensus is that it does not reliably distinguish truth from deception. Courts know this. The research community knows this. The general public, shaped by decades of television and cultural mythology, largely does not.

In a legal system where the appearance of science carries enormous weight, that gap between perception and reality causes real harm to real people.

If you want to go deeper: the National Academy of Sciences 2003 report, The Polygraph and Lie Detection, is available through the National Academies Press at nap.nationalacademies.org. The American Psychological Association's position on polygraph testing is available at apa.org.

Sources

[1] National Research Council. (2003). The Polygraph and Lie Detection. National Academies Press. nap.nationalacademies.org

[2] Hare, R. D. (1991). The Hare Psychopathy Checklist — Revised. Multi-Health Systems. mhs.com

[3] Iacono, W. G., and Lykken, D. T. (1997). The Validity of the Lie Detector: Two Surveys of Scientific Opinion. Journal of Applied Psychology, 82(3), 426–433. psycnet.apa.org

[4] National Research Council. (2003). The Polygraph and Lie Detection. Committee to Review the Scientific Evidence on the Polygraph. National Academies Press. nap.nationalacademies.org

[5] Weiner, T. (2007). Legacy of Ashes: The History of the CIA. Doubleday. penguinrandomhouse.com

[6] Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). oyez.org/cases/1992/92-102