Update: The scientific status of research on polygraph techniques: The case against polygraph tests

This 1999 update by William G. Iacono, published as a pocket-part chapter in Modern Scientific Evidence: The Law and Science of Expert Testimony, critically examines the scientific status of the Comparison Question Test (CQT) and related polygraph screening formats. Drawing on field studies, laboratory research, survey data from scientific organizations, and analysis of the Test for Espionage and Sabotage, the chapter argues that polygraph testing lacks a credible scientific foundation sufficient for forensic or national security use. The work was cited directly by the U.S. Supreme Court in United States v. Scheffer (1998) and remains a landmark reference in the legal and scientific debate over polygraph admissibility.

Critical analytical review synthesizing laboratory studies, field validity research, countermeasure experiments, and large-scale surveys of scientific opinion (SPR members at 91% response rate; APA Division of General Psychology Fellows at 74% response rate), with particular emphasis on the structural methodological flaws of confession-based field study designs.

This 1999 update built on Iacono and Lykken's foundational 1997 critique by incorporating new empirical data and extending analysis of CQT validity in the post-Daubert legal landscape. The review systematically evaluated laboratory mock-crime studies for ecological validity, critiqued field studies for confession-based verification problems, and analyzed screening test accuracy including the Test for Espionage and Sabotage. Iacono integrated professional survey data showing scientific skepticism and reviewed countermeasure research demonstrating CQT vulnerability to manipulation. The analysis concluded that fundamental methodological limitations in polygraph research undermined validity claims for both criminal and security screening applications.

The review provided scientific ammunition for legal challenges to polygraph admissibility and raised serious questions about expanding federal agency screening programs, directly influencing policy debates about polygraph use in national security contexts.

Background & Context

By the late 1990s, polygraph testing occupied a deeply contested position within the American legal and national security landscape. Courts were grappling with post-Daubert standards for scientific evidence admissibility, federal agencies were expanding employee screening programs, and the scientific community remained sharply divided. It was against this backdrop that William G. Iacono — Distinguished McKnight University Professor at the University of Minnesota and one of the foremost psychophysiology researchers of his era — updated his landmark critique of polygraph science for the 1999 Pocket Part of Modern Scientific Evidence: The Law and Science of Expert Testimony.

The 1999 update (appearing at Vol. 1, pp. 174–184) built directly on the foundational 1997 chapter co-authored with David T. Lykken (pp. 582–618 of the original volume). Where the 1997 chapter established the overarching case against polygraph validity, the 1999 update incorporated newly available data, extended the analysis of field studies and screening test procedures, and responded to developments in the pro-polygraph literature. The chapter appeared in a law-and-science reference work explicitly designed to guide judges, attorneys, and expert witnesses navigating forensic evidentiary standards — giving it direct, practical legal weight.

The timing was particularly significant. The U.S. Supreme Court had just ruled in United States v. Scheffer (1998), upholding a blanket military ban on polygraph evidence — a decision that itself cited Iacono and Lykken's work. Meanwhile, federal agencies including the Department of Energy were dramatically expanding counterintelligence scope polygraph screening programs in the wake of nuclear espionage concerns, making the question of polygraph accuracy a matter of urgent national security policy.

Research Design & Methodology

Rather than a single empirical study, the 1999 update functions as a critical analytical review — synthesizing the existing body of empirical literature, evaluating research methodologies, and applying rigorous psychometric and logical criteria to assess the validity claims of polygraph proponents. Iacono drew on both published peer-reviewed studies and his own primary research to construct this evaluation. The analysis focuses principally on the Comparison Question Test (CQT), the dominant format used in forensic and employment screening contexts.

Key methodological elements of the review include:

• Critical appraisal of laboratory studies — evaluating mock-crime analog designs for ecological validity and their capacity to generalize to real-world polygraph conditions
• Critical appraisal of field studies — analyzing the logical and statistical problems inherent in confession-based ground-truth verification methods
• Survey data integration — drawing on the landmark Iacono & Lykken (1997b) surveys of Society for Psychophysiological Research (SPR) members (91% response rate) and American Psychological Association Division of General Psychology Fellows (74% response rate)
• Countermeasure research synthesis — reviewing controlled laboratory studies on the vulnerability of CQT formats to deliberate physiological manipulation
• Screening test analysis — specific evaluation of the Test for Espionage and Sabotage (TES) and Department of Defense Polygraph Institute laboratory data on its accuracy

The review also specifically examined the Patrick & Iacono (1991) Royal Canadian Mounted Police field study — the only published investigation at the time designed to remove the systematic sampling bias that inflates accuracy estimates in conventional field research. This methodological centerpiece allowed Iacono to demonstrate that standard field studies are structurally incapable of yielding unbiased validity estimates.

Results & Key Findings

The headline finding: field studies conducted in high-stakes real-world conditions indicate that approximately 40–43% of innocent subjects fail the CQT polygraph — a false positive rate so large as to render the test fundamentally unsuitable for individual-level diagnostic use. This figure, drawn from studies published in premier peer-reviewed journals including Nature and the Journal of Applied Psychology, stands in stark contrast to the accuracy rates claimed by polygraph proponents.

Key findings across the review include:

• ~43% of innocent individuals fail the CQT in high-quality real-life criminal field studies, when tests are scored according to standard physiological criteria
• The sole methodologically sound field study (Patrick & Iacono, 1991) found CQT accuracy for innocent suspects at only 57% — barely above chance
• DoDPI laboratory studies of the Test for Espionage and Sabotage found a 12.5% false positive rate (14 of 115 innocent subjects) — but Iacono argued this figure is a dramatic underestimate for real-world screening conditions
• With less than 30 minutes of countermeasure instruction, 50% or more of guilty individuals can defeat the CQT in laboratory settings, with trained examiners unable to detect countermeasure use
• Only about one-third of surveyed SPR and APA-DGP scientists believed the CQT was scientifically sound
• Only about one-quarter of those scientists believed polygraph results should be admissible as court evidence
• Only 20% considered the CQT a standardized test; only 10% considered it objective

Iacono also demonstrated a critical structural bias: conventional field studies systematically overestimate accuracy because cases are selected only when the original examiner's deceptive call was followed by a confession. This design eliminates false positives and false negatives from the sample by construction, inflating hit rates to near-perfect levels that are entirely artifactual. The CQT's theoretical foundations — requiring innocent subjects to respond more strongly to control questions than to relevant questions about serious crimes — were shown to be psychologically implausible given that innocent people clearly perceive relevant questions as carrying far higher personal stakes.

Discussion & Significance

The chapter's significance extends well beyond academic debate. By appearing in a legal reference series explicitly targeted at courts and attorneys, Iacono and Lykken's critique directly shaped how the Daubert evidentiary framework was applied to polygraph evidence. The work was cited directly by the U.S. Supreme Court majority in United States v. Scheffer (1998), which upheld the military's blanket exclusion of polygraph evidence, and the argument that CQT field accuracy is "little better than could be obtained by the toss of a coin" was quoted in that decision. This gave the scientific critique concrete legal standing at the highest judicial level.

The 1999 update also advanced a particularly important argument about the structural impossibility of adequate field validation. Because polygraph examiners know that failing large numbers of screened employees would be institutionally catastrophic, real-world national security screening tests are not scored according to objective criteria — examiners make subjective post-interview adjustments to keep failure rates manageably low. This means the same test effectively operates under two different scoring regimes: a rigorous one that produces the alarming false positive rates seen in criminal field studies, and a subjective one that produces near-zero failure rates in employment screening — neither of which can be considered scientifically interpretable.

The survey findings carry their own independent weight. The breadth of scientific consensus against the CQT — measured across two distinct professional organizations with different sampling frames — demonstrates that skepticism toward polygraph validity is not the idiosyncratic view of a few critics, but the dominant position of the relevant scientific community. This is precisely the kind of evidence courts weigh when assessing whether a technique has achieved general scientific acceptance, a key criterion under both the older Frye standard and the more flexible Daubert framework.

Limitations & Considerations

Iacono explicitly acknowledges the core epistemological limitation that runs through the entire analysis: the absence of methodologically sound field validity data makes it impossible to provide a reliable error rate estimate for real-world CQT use. This is not a finding that favors one side of the debate — it means proponents cannot establish high accuracy any more than critics can definitively quantify low accuracy. The indeterminacy itself is the problem.

• Laboratory-to-field generalizability: Mock crime studies use volunteer subjects with minimal stakes, making results inherently difficult to apply to adversarial real-world examinations where careers, liberty, and reputations are at risk
• Sampling bias in field studies: Confession-based ground truth selection systematically excludes cases where examiners erred, producing inflated accuracy estimates that cannot be corrected without access to unbiased case samples
• Screening context novelty: The TES and directed lie formats analyzed were relatively new, with limited empirical data at the time of the 1999 update
• Survey limitations: Scientific opinion, while informative for general acceptance analysis, is not itself a direct measure of test validity — experts can be wrong collectively
• Chapter format constraints: As a legal reference update (pp. 174–184), the 1999 chapter is concise by necessity, with fuller treatments available in the original 1997 chapter and related journal publications

Practical Applications

For legal practitioners, this chapter remains a critical reference when challenging the admissibility of polygraph evidence under Daubert or assessing whether a technique meets the general acceptance standard. Its direct citation by the Supreme Court gives it unusual persuasive authority. Defense attorneys and civil liberties advocates working against mandatory polygraph screening programs — particularly in federal employment or national security contexts — will find Iacono's analysis of the structural biases embedded in screening-test scoring especially relevant.

For consumers and examinees, the practical implication is sobering: in real-world high-stakes conditions, the physiological arousal that an innocent person naturally experiences when confronted with serious accusations may be indistinguishable from a guilty person's response. The fact that 40% or more of innocent subjects fail in rigorous field conditions means that passing a polygraph test does not reliably establish truthfulness, and failing one does not reliably establish deception. Anyone subject to a polygraph examination — whether in a criminal investigation, pre-employment screening, or security clearance context — should understand that the test's outcome reflects physiology, not truth.