Comparison Questions in Polygraph Testing: Expert CQT Guide

Defining the Comparison Question in Polygraph Science

A comparison question — sometimes called a "control question" in older literature — is a carefully constructed inquiry used during a Comparison Question Technique (CQT) polygraph examination. Its primary purpose is to serve as a physiological benchmark. The examiner uses the body's measurable reactions to comparison questions as a standard against which responses to the actual investigative (relevant) questions are evaluated.

Unlike relevant questions that directly address the matter under investigation — such as "Did you take the missing funds from the cash register?" — comparison questions deal with broader, more general behaviors from the examinee's past. They are intentionally designed to be vague enough to cause virtually everyone a degree of internal discomfort or uncertainty, regardless of whether they are involved in the issue being tested. Research by Horowitz, Kircher, Honts, and Raskin (1997) demonstrated through a mock crime study with 120 participants that comparison questions produce different physiological patterns for innocent versus guilty subjects, providing empirical support for the CQT rationale [1]Verified The Role of Comparison Questions in Physiological Detection of Deception
Mock crime study with 120 participants confirming comparison questions produce different physiological patterns for innocent vs. guilty subjects.

A classic example might be: "During the first 25 years of your life, did you ever take something that did not belong to you?" or "Have you ever lied to someone who trusted you?" These questions cover such a wide scope of human experience that almost every person has engaged in at least some version of the behavior in question — making it highly unlikely anyone can answer "No" with complete physiological calm.

The examiner then measures how the examinee's body responds to the comparison question versus how it responds to the relevant question. The pattern of responses — across multiple polygraph chart collections — reveals whether the examinee is more concerned about the comparison issue or the specific issue under investigation. For more on what results mean, see our guide to No Deception Indicated (NDI) results.

Why the Term 'Comparison' Replaced 'Control'

If you've been researching polygraph testing, you may have encountered both the terms "control question" and "comparison question." These refer to the same concept, but modern polygraph science strongly favors the term "comparison" — and for good reason.

The word "control" in experimental science has a specific meaning: a condition held constant while the independent variable is manipulated. In polygraph testing, the so-called "control" question is not truly a controlled variable in the scientific sense. It is not held constant; it is actively designed to provoke a physiological response. Calling it a "control" question led to confusion in academic circles.

The American Polygraph Association (APA), along with leading researchers like Raymond Nelson, have advocated for the term "comparison question" to more accurately describe its function: it provides a comparison point for physiological reactions, not an experimental control [3]Verified Scientific Basis for Polygraph Testing
Confirms event-specific diagnostic accuracy mean of 89% (CI 83%–95%) and screening accuracy mean of 85% (CI 77%–93%). Nelson's comprehensive 2015 review, published in Polygraph, 44(1), 28–61, confirmed that event-specific diagnostic polygraphs provide a mean accuracy of 89% with a 95% confidence range from 83% to 95% [3]Verified Scientific Basis for Polygraph Testing
Confirms event-specific diagnostic accuracy mean of 89% (CI 83%–95%) and screening accuracy mean of 85% (CI 77%–93%). Since 1947, when the comparison question was first integrated by John Reid into the test format, it has been the central element of polygraph methodology and the most frequently used test technique by polygraph examiners worldwide [4]Verified A Revised Questioning Technique in Lie-Detection Tests
Confirms John E. Reid published the original comparison question technique in 1947 in Journal of Criminal Law and Criminology, Vol. 37(6), 542–547.

The Relevant-Irrelevant Test: The Predecessor

Before comparison questions existed, polygraph examiners relied on the Relevant-Irrelevant Test (RIT), developed in the early 1920s by John Larson and later refined by Leonarde Keeler. This method alternated between relevant questions ("Did you steal the money?") and irrelevant questions ("Is today Tuesday?"). The theory was that guilty examinees would react strongly to relevant questions but not to irrelevant ones, while innocent examinees would show minimal reactions to both.

The problem was significant: innocent examinees who were nervous about the test itself — about the relevant questions, about the consequences of a wrong result — would often show strong reactions to relevant questions simply because those questions were inherently more threatening than asking what day of the week it was. The RIT had no way to distinguish between anxiety caused by guilt and anxiety caused by the accusatory nature of the testing situation. For historical context on early polygraph development, see our guide to polygraph in the 1950s.

John Reid's 1947 Breakthrough

John E. Reid, a Chicago-based examiner, former law student, and member of the staff of the Chicago Police Scientific Crime Detection Laboratory, recognized this fundamental flaw [4]Verified A Revised Questioning Technique in Lie-Detection Tests
Confirms John E. Reid published the original comparison question technique in 1947 in Journal of Criminal Law and Criminology, Vol. 37(6), 542–547. In 1947, he published "A Revised Questioning Technique in Lie-Detection Tests" in the Journal of Criminal Law and Criminology (Volume 37, Issue 6, pages 542–547), introducing the comparison (then called "control") question — a question designed to evoke a physiological response from truthful examinees that would rival or exceed their response to relevant questions [4]Verified A Revised Questioning Technique in Lie-Detection Tests
Confirms John E. Reid published the original comparison question technique in 1947 in Journal of Criminal Law and Criminology, Vol. 37(6), 542–547.

The probable-lie comparison (PLC) question, known as the Reid Examination Technique, is considered one of the most significant developments in polygraph testing [5]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
250-participant mock crime experiment showing equivalent validity for probable-lie and directed-lie CQT variants; confirms Raskin introduced DLCQ in the 1980s. By adding this third question type, Reid created a within-subject comparison that dramatically improved the technique's ability to distinguish between truthful and deceptive individuals.

Reid's reasoning was straightforward: if a truthful person is more worried about lying to a broad moral question ("Have you ever stolen anything?") than about denying a specific allegation they are innocent of, their body will show it. Conversely, a guilty person will be most worried about the specific relevant question — the one that directly threatens them — and will react more strongly to it. The CQT replaced the Relevant/Irrelevant Question Technique and represented a major breakthrough in polygraph methodology [2]Verified Examining Different Types of Comparison Questions in a Field Study of CQT Polygraph Technique
Field study providing evidence for practical effectiveness of both exclusive and non-exclusive comparison questions in real criminal investigations.

Backster's Zone Comparison Refinement

Building on Reid's innovation, Cleve Backster developed the Zone Comparison Technique (ZCT). Historical sources confirm Backster published his "Standardized Polygraph Notepack and Technique Guide: Backster Zone Comparison Technique" in 1963, from the Backster School of Lie Detection in New York [12]Verified Standardized Polygraph Notepack and Technique Guide: Backster Zone Comparison Technique
Confirms Backster's 1963 publication establishing the standardized Zone Comparison Technique from the Backster School of Lie Detection, New York. The 2003 NRC report describes the ZCT as developed by Backster and notes it was named for the three "zones" or blocks of time during the test [6]Verified The Polygraph and Lie Detection
NRC concluded specific-incident polygraph tests discriminate lying from truth telling at rates 'well above chance, though well below perfection'.

Backster introduced strict question ordering, the concept of "zones" for comparison and relevant question pairing, and the Anticlimax Dampening Concept — published in Polygraph journal in 1974 [12]Verified Standardized Polygraph Notepack and Technique Guide: Backster Zone Comparison Technique
Confirms Backster's 1963 publication establishing the standardized Zone Comparison Technique from the Backster School of Lie Detection, New York. He also introduced a quantification system of chart analysis using numerical scoring, making the process more objective and scientific than before. This was the first comparison question test to incorporate a numerical scoring system, using a seven-point scale [8]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
Largest CQT meta-analysis: 138 datasets, AUC of 0.91, confirming CQT performs well above chance with significant information gain over interpersonal deception detection.

Backster's Zone Comparison Technique was adopted by the U.S. Army Military Police School (USAMPS) in 1961 [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles, and his numerical scoring approach has become standard procedure in the polygraph field. Although Reid (1947) provided the first description of a comparison question format, Backster (1963) provided the first highly standardized rationale and structure for the administration and scoring of a comparison question technique [12]Verified Standardized Polygraph Notepack and Technique Guide: Backster Zone Comparison Technique
Confirms Backster's 1963 publication establishing the standardized Zone Comparison Technique from the Backster School of Lie Detection, New York. For more on this era of polygraph development, see our guide to the polygraph industry in the 1990s.

Modern Evolution: The Utah Approach and Federal Standards

From the 1970s onward, academic researchers — particularly David Raskin at the University of Utah — subjected the CQT to rigorous laboratory and field studies [14]Verified Utah Approach to Comparison Question Polygraph Testing
Confirms Raskin began systematic CQT research in 1970 at the University of Utah, producing the Utah CQT and Utah Numerical Scoring System over 30 years. In 1970, Raskin began a systematic study of the probable-lie comparison question polygraph technique, and over 30 years of research produced the Utah CQT and the corresponding Utah Numerical Scoring System [14]Verified Utah Approach to Comparison Question Polygraph Testing
Confirms Raskin began systematic CQT research in 1970 at the University of Utah, producing the Utah CQT and Utah Numerical Scoring System over 30 years.

Raskin also introduced the Directed Lie Control Question (DLCQ) in the 1980s, adding another important variant to the CQT family [5]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
250-participant mock crime experiment showing equivalent validity for probable-lie and directed-lie CQT variants; confirms Raskin introduced DLCQ in the 1980s. The earliest published field study of the directed-lie approach was by Honts and Raskin in 1988 [5]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
250-participant mock crime experiment showing equivalent validity for probable-lie and directed-lie CQT variants; confirms Raskin introduced DLCQ in the 1980s.

The federal government developed its own standardized formats. The Federal Zone Comparison Test (Federal ZCT), which includes relevant questions scored against adjacent comparison questions, is used in specific-issue testing. The Air Force Modified General Question Technique (AFMGQT) is a modified version of the Reid technique that is widely used across the United States Federal Government for both criminal-specific and counterintelligence screening purposes [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles. The AFMGQT incorporated many principles from Backster's ZCT and uses probable-lie comparison questions with numerical evaluation, though some versions permit directed-lie comparison questions [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles. In 1966, the U.S. Army adopted the Reid technique and taught it at the USAMPS polygraph course, and in 1968, USAMPS modified Reid's technique and called it the Army MGQT [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles. These formats are taught at the National Center for Credibility Assessment (NCCA), the successor to the Department of Defense Polygraph Institute. Learn about how polygraph is used in federal agencies with our guide to the ATF polygraph exam.

Understanding the Triad: Relevant, Irrelevant, and Comparison

Every CQT-based polygraph examination uses three distinct question types, each serving a different purpose. Understanding all three is essential to grasping why comparison questions exist and how they function. All polygraph questioning techniques that aim at standardization involve comparisons of physiological responses to relevant questions against physiological responses to comparison questions.

Irrelevant questions are simple, factual questions with known, verifiable answers. Examples include "Are you sitting in a chair right now?" or "Is today Wednesday?" They produce minimal physiological response and serve as neutral anchors in the question sequence, helping the examinee settle into the testing rhythm.

Relevant questions directly address the specific issue under investigation. For an infidelity test, a relevant question might be: "Since we've been married, have you had sexual intercourse with anyone other than me?" For a theft case: "Did you take the missing $5,000 from the company safe?" These are the questions whose physiological responses are compared against comparison question responses to reach a diagnostic opinion.

Comparison questions are broad, deliberately vague questions about general past behavior. They are designed to make virtually everyone slightly uncomfortable when answering "No." They provide the critical physiological baseline. A truthful examinee should show stronger reactions to comparison questions (because they are uncertain or slightly dishonest in their answer), while a deceptive examinee should react more strongly to relevant questions (because those directly threaten them) [2]Verified Examining Different Types of Comparison Questions in a Field Study of CQT Polygraph Technique
Field study providing evidence for practical effectiveness of both exclusive and non-exclusive comparison questions in real criminal investigations.

The interplay between these three question types is what makes the CQT function. Without comparison questions, the test reverts to the problematic relevant-irrelevant format. Understanding the differences between private and court-ordered polygraphs can help you prepare for what to expect.

The Psychological Logic Behind Comparison Questions

To understand how comparison questions work, you need to understand a core psychological principle: relative salience. Your autonomic nervous system — the part of your body that controls heart rate, breathing, sweat gland activity, and blood pressure — responds most strongly to whatever you perceive as the most threatening stimulus in your immediate environment. This principle is the engine that drives the entire CQT methodology.

For truthful examinees, the relevant question is not the biggest threat — they know they did not do it. But the comparison question, which asks about broad, vague behavior from the past, creates genuine uncertainty. When asked "Have you ever lied to get out of trouble?" they must answer "No," even though they almost certainly have at some point. This mild dishonesty or uncertainty triggers a measurable physiological response. Because the comparison question creates more psychological discomfort than the relevant question, it produces the larger physiological reaction. This pattern — comparison reactions exceeding relevant reactions — is the hallmark of a non-deceptive (truthful) result.

For deceptive examinees, the relevant question is by far the most threatening stimulus. It directly references the act they committed and the consequences they face if detected. Meanwhile, the comparison question about past lying or stealing is only mildly concerning. This pattern — relevant reactions exceeding comparison reactions — is the hallmark of a deceptive result [2]Verified Examining Different Types of Comparison Questions in a Field Study of CQT Polygraph Technique
Field study providing evidence for practical effectiveness of both exclusive and non-exclusive comparison questions in real criminal investigations.

Research by Horowitz, Kircher, Honts, and Raskin (1997) provided empirical support for this mechanism through a mock crime study with 120 participants, demonstrating that comparison questions produced different physiological patterns for innocent versus guilty subjects [1]Verified The Role of Comparison Questions in Physiological Detection of Deception
Mock crime study with 120 participants confirming comparison questions produce different physiological patterns for innocent vs. guilty subjects. MacNeill et al. (2014) further investigated cognitive and emotional reactions to comparison questions, finding that when participants generated personally relevant comparison questions, the CQT rationale was better supported [10]Verified Cognitive and Emotional Reactions to Questions in the Comparison Question Test
Investigated cognitive and emotional reactions to CQT questions; found personally relevant comparison questions better support the CQT rationale.

The Scoring Process

The examiner evaluates these relative differences across multiple chart collections (typically three to five repetitions of the question sequence). Each comparison-relevant question pair is scored numerically — usually on a 3- or 7-position scale — based on the magnitude and consistency of the physiological differences [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles. These scores are then summed to produce a total score, which is compared against validated cutoff thresholds to reach a diagnostic opinion of non-deceptive (NDI), deceptive (DI), or inconclusive (INC).

Honts (1999) demonstrated that between-chart discussion of comparison questions significantly improved CQT accuracy [7]Verified The Discussion of Comparison Questions Between List Repetitions Is Associated with Increased Test Accuracy
Confirms between-chart discussion of comparison questions significantly improves CQT accuracy, a procedural refinement that became standard practice in modern CQT administration. Krapohl (2020) further refined scoring methodology by establishing an optimal minimum difference threshold of 30% for scoring electrodermal responses against the stronger of two comparison questions [11]Verified Electrodermal Response Ratios: Scoring Against the Stronger of Two Comparison Questions
Established optimal 30% minimum difference threshold for scoring electrodermal responses, with point-biserial correlation of 0.680 with ground truth.

Modern examinations also use automated scoring algorithms — such as the Objective Scoring System (OSS), the Empirical Scoring System (ESS), or the Computerized Polygraph System (CPS) — as complementary analysis methods. These algorithms apply statistical models to the raw physiological data, providing an objective probability estimate that is compared with the examiner's manual scoring for quality assurance. For more on how statistical analysis works in polygraph testing, see our guide on AI lie detection vs polygraph.

Two Approaches to Comparison Question Design

The polygraph field uses two fundamentally different approaches to comparison question construction, each with its own psychological rationale and advantages.

Probable-Lie Comparisons (PLCs): The examiner constructs a question that covers a broad category of behavior — stealing, lying, cheating — and the examinee is expected to answer "No." The examiner suspects (with high probability) that the answer is not fully truthful, which is precisely the point. Reid's original Probable Lie Control Question (PLCQ), later labelled as the Non-Exclusive Control Question (NECQ), was later modified by Backster into the Exclusive Control Question (ECQ) [5]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
250-participant mock crime experiment showing equivalent validity for probable-lie and directed-lie CQT variants; confirms Raskin introduced DLCQ in the 1980s. Example: "Before the age of 30, did you ever steal anything from an employer?"

Directed-Lie Comparisons (DLCs): The examiner explicitly instructs the examinee to answer "No" to a question they know is a lie. For example: "Have you ever told even one lie in your entire life?" The examiner tells the examinee beforehand that they should answer "No" and should think about a specific time they actually did lie. Raskin introduced the Directed Lie Control Question in the 1980s to address concerns about standardization and cultural bias [5]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
250-participant mock crime experiment showing equivalent validity for probable-lie and directed-lie CQT variants; confirms Raskin introduced DLCQ in the 1980s.

Honts and Reavy (2015) conducted a mock crime experiment with 250 paid participants contrasting probable-lie and directed-lie CQT variants. They found substantial main effects of guilt in both computer and human scoring, with no significant differences between the two test types [5]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
250-participant mock crime experiment showing equivalent validity for probable-lie and directed-lie CQT variants; confirms Raskin introduced DLCQ in the 1980s. Shaw's (2012) empirical review similarly found little to no meaningful difference in effect size or detection accuracy between directed-lie and probable-lie comparison questions [9]Verified The Empirical Basis for the Use of Directed Lie Comparison Questions in Diagnostic and Screening Polygraphs
Found little to no meaningful difference in effect size or detection accuracy between directed-lie and probable-lie comparison questions. Driscoll, Honts, and Jones (1987) showed that DLC produced fewer false positives while maintaining detection of deception [15]Verified Directed Lie Comparison Questions in Polygraph Testing
Laboratory comparison found DLC produced fewer false positives while maintaining detection of deception, supporting DLC adoption as standard practice.

PLCs are considered more natural and may produce stronger responses in some populations, but they require more skill from the examiner in question construction and pre-test interview technique. DLCs are more standardized and easier to teach, making them particularly useful in screening contexts. The Directed Lie Screening Test (DLST) and the Test for Espionage and Sabotage (TES) employ directed-lie comparisons, while the traditional Zone Comparison Test uses probable-lie comparisons [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles. Some modern formats, like the AFMGQT, permit the use of either type [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles. Research by Honts, Handler, Shaw, and Gougler (2015) suggests that modern testing formats with directed lies may also be more robust against deliberate manipulation attempts [16]Verified Directed Lie Comparison Questions and Countermeasure Resistance
Research suggests modern testing formats with directed lies may be more robust against deliberate manipulation attempts.

The Art and Science of Question Construction

Crafting an effective comparison question requires both skill and training. The question must be broad enough to create uncertainty for virtually every examinee, yet specific enough to feel personally relevant. A qualified polygraph examiner will tailor comparison questions during the pre-test interview to maximize their effectiveness for each individual.

Key principles of comparison question construction include ensuring the question covers a wide scope of behavior that most people have engaged in at some point, separating the question from the relevant issue by time, place, or category (in exclusive comparison question formats), and using language that makes it psychologically difficult to answer "No" with complete honesty.

Ginton's (2017) field study comparing different types of comparison questions in real criminal investigations provided evidence for the practical effectiveness of both exclusive and non-exclusive comparison questions in operational settings [2]Verified Examining Different Types of Comparison Questions in a Field Study of CQT Polygraph Technique
Field study providing evidence for practical effectiveness of both exclusive and non-exclusive comparison questions in real criminal investigations. Amsel's (1999) Israeli field study comparing exclusive versus non-exclusive comparison question formats found both formats produced comparable accuracy rates in real criminal investigations [17]Verified Exclusive or Nonexclusive Comparison Questions: A Comparative Field Study
Israeli field study found both exclusive and non-exclusive comparison question formats produced comparable accuracy rates in real criminal investigations.

Matte (2011) examined the effect of habituation on comparison questions across successive charts, finding that 62.9% of confirmed deceptive cases showed higher scores in subsequent charts, supporting differential habituation theory [18]Verified Effect of Habituation to Least Threatening Zone Questions on the Most Threatening Zone Comparison Questions
Found 62.9% of confirmed deceptive cases showed higher scores in subsequent charts, supporting differential habituation theory. This research underscores the importance of proper comparison question formulation and management throughout the examination process.

All questions — including comparison questions — are thoroughly reviewed with the examinee during the pre-test interview before any chart collection begins. This transparency is a hallmark of ethical polygraph practice. Learn more about the dynamics of polygraph testing and the limits of admission.

Numerical Scoring Systems

The comparison of physiological responses between comparison and relevant questions is quantified through numerical scoring systems. The two most common systems are the 3-position scale (used in many federal formats) and the 7-position scale (used in the Utah system and others) [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles.

In both systems, the examiner assigns a numerical value to each comparison-relevant question pair based on the relative magnitude of physiological responses. Positive scores indicate greater responses to comparison questions (suggesting truthfulness), while negative scores indicate greater responses to relevant questions (suggesting deception). Zero indicates no discernible difference.

The Utah Numerical Scoring System, developed by Bell, Raskin, Honts, and Kircher (1999), resulted from over 30 years of scientific research and provides some of the highest rates of criterion accuracy and interrater reliability of any polygraph examination protocol [14]Verified Utah Approach to Comparison Question Polygraph Testing
Confirms Raskin began systematic CQT research in 1970 at the University of Utah, producing the Utah CQT and Utah Numerical Scoring System over 30 years.

Krapohl's (2020) research on electrodermal response ratios established that the optimal minimum difference for scoring electrodermal responses against the stronger of two comparison questions was 30%, with a point-biserial correlation of 0.680 between scores and ground truth [11]Verified Electrodermal Response Ratios: Scoring Against the Stronger of Two Comparison Questions
Established optimal 30% minimum difference threshold for scoring electrodermal responses, with point-biserial correlation of 0.680 with ground truth. This kind of precision in scoring methodology continues to refine the accuracy of CQT examinations. Quality assurance through peer review further ensures consistent scoring standards across the profession.

Major Meta-Analyses and Reviews

The CQT has been the subject of extensive scientific research over the past five decades. Multiple large-scale reviews have examined its accuracy.

Nelson's (2015) comprehensive review of published scientific literature, appearing in Polygraph, 44(1), 28–61, found that event-specific diagnostic polygraphs provide a mean accuracy of 89% with a 95% confidence range from 83% to 95% [3]Verified Scientific Basis for Polygraph Testing
Confirms event-specific diagnostic accuracy mean of 89% (CI 83%–95%) and screening accuracy mean of 85% (CI 77%–93%). Multi-issue screening polygraphs were shown to provide a mean accuracy of 85% with a 95% confidence range of 77% to 93% [3]Verified Scientific Basis for Polygraph Testing
Confirms event-specific diagnostic accuracy mean of 89% (CI 83%–95%) and screening accuracy mean of 85% (CI 77%–93%).

The largest meta-analysis of the CQT to date was conducted by Honts, Thurber, and Handler (2021), published in Applied Cognitive Psychology, 35(2), 411–427 [8]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
Largest CQT meta-analysis: 138 datasets, AUC of 0.91, confirming CQT performs well above chance with significant information gain over interpersonal deception detection. They analyzed 138 independent datasets using broad inclusion criteria [8]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
Largest CQT meta-analysis: 138 datasets, AUC of 0.91, confirming CQT performs well above chance with significant information gain over interpersonal deception detection. The meta-analytic effect size including inconclusive outcomes was 0.69 [0.66, 0.79], which converts to a Cohen's d of 1.92 and an AUC of 0.91 [8]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
Largest CQT meta-analysis: 138 datasets, AUC of 0.91, confirming CQT performs well above chance with significant information gain over interpersonal deception detection. Information Gain analysis showed that CQT outcomes provide a significant information increase over interpersonal deception detection across almost the complete range of base rates [8]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
Largest CQT meta-analysis: 138 datasets, AUC of 0.91, confirming CQT performs well above chance with significant information gain over interpersonal deception detection. A CQT deceptive outcome was found to be approximately 6 times more informative than a layperson's judgment that a person is lying [8]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
Largest CQT meta-analysis: 138 datasets, AUC of 0.91, confirming CQT performs well above chance with significant information gain over interpersonal deception detection.

The APA's 2011 meta-analytic survey of criterion accuracy examined 38 studies covering 3,723 examinations, finding an overall 87% decision accuracy with confidence intervals from 80% to 94% across all validated polygraph techniques [3]Verified Scientific Basis for Polygraph Testing
Confirms event-specific diagnostic accuracy mean of 89% (CI 83%–95%) and screening accuracy mean of 85% (CI 77%–93%).

The 2003 National Research Council report, the most authoritative independent governmental review, concluded that specific-incident polygraph tests can discriminate lying from truth telling at rates "well above chance" in populations of examinees untrained in countermeasures [6]Verified The Polygraph and Lie Detection
NRC concluded specific-incident polygraph tests discriminate lying from truth telling at rates 'well above chance, though well below perfection'. An analysis of seven field studies involving specific incidents revealed a median accuracy of 89% [6]Verified The Polygraph and Lie Detection
NRC concluded specific-incident polygraph tests discriminate lying from truth telling at rates 'well above chance, though well below perfection'. For deeper context on intelligence community polygraph practices, see our article on the Aldrich Ames case.

Professional Standards and Institutional Support

The CQT is supported by established professional organizations and standards bodies. The American Polygraph Association (APA) recognizes the CQT as a validated testing methodology and maintains standards of practice for its administration [3]Verified Scientific Basis for Polygraph Testing
Confirms event-specific diagnostic accuracy mean of 89% (CI 83%–95%) and screening accuracy mean of 85% (CI 77%–93%). The National Center for Credibility Assessment (NCCA), the U.S. Department of Defense's polygraph training institution, teaches CQT formats including the Federal ZCT and AFMGQT [13]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT is widely used across the U.S. Federal Government for criminal and counterintelligence purposes; adopted Backster's ZCT principles.

ASTM International's Committee E52 on Forensic Psychophysiology, formed in 1996, has developed a series of standards covering all aspects of polygraphy, from research methodology to ethics and instrumentation [19]Verified ASTM Committee E52 on Forensic Psychophysiology Standards
Confirms ASTM Committee E52 was formed in 1996 and has developed professional standards covering research, instrumentation, training, and ethics in polygraphy. These standards — including E2229 on interpretation of polygraph data and E2062 on examination standards of practice — are widely cited in the profession and recognized in courts [19]Verified ASTM Committee E52 on Forensic Psychophysiology Standards
Confirms ASTM Committee E52 was formed in 1996 and has developed professional standards covering research, instrumentation, training, and ethics in polygraphy. ASTM E52 standards address test formats, question types, testing locations, and other procedural issues relevant to CQT administration [19]Verified ASTM Committee E52 on Forensic Psychophysiology Standards
Confirms ASTM Committee E52 was formed in 1996 and has developed professional standards covering research, instrumentation, training, and ethics in polygraphy.

For more on the importance of polygraph in national security, see our dedicated guide.

Addressing Myths and Misunderstandings

Misconception 1: Comparison questions are trick questions designed to trap you. In reality, comparison questions are fully reviewed with the examinee before testing begins. There are no surprises. The examiner explains every question and ensures the examinee understands and agrees to the wording.

Misconception 2: You must lie to the comparison question for the test to work. While the probable-lie format relies on examinees being less than fully truthful, the directed-lie format explicitly instructs the examinee to answer dishonestly. Both approaches have been shown to produce equivalent accuracy [5]Verified The Comparison Question Polygraph Test: A Contrast of Methods and Scoring
250-participant mock crime experiment showing equivalent validity for probable-lie and directed-lie CQT variants; confirms Raskin introduced DLCQ in the 1980s. The critical factor is that the comparison question generates a measurable physiological response.

Misconception 3: Nervous people will always fail because they react to everything. The CQT is specifically designed to account for general nervousness. Because the test measures relative differences between comparison and relevant question responses — not absolute response levels — a generally anxious person's reactions to comparison questions should still be proportionally larger than their reactions to relevant questions if they are truthful. This is precisely the advantage the CQT has over the older Relevant-Irrelevant Test.

Misconception 4: The examiner decides the result subjectively. Modern CQT examinations use standardized numerical scoring systems with validated cutoff thresholds, often supplemented by automated computer scoring algorithms [3]Verified Scientific Basis for Polygraph Testing
Confirms event-specific diagnostic accuracy mean of 89% (CI 83%–95%) and screening accuracy mean of 85% (CI 77%–93%). The process is designed to minimize subjective judgment. Learn more about how false allegations can be addressed through objective testing procedures.

Preparing for the CQT Experience

If you are scheduled for a CQT polygraph examination, understanding the comparison question process can significantly reduce unnecessary anxiety. Here is what to expect:

During the pre-test interview (typically 30–60 minutes), the examiner will explain the entire testing procedure, discuss all questions to be asked, and develop the comparison questions in collaboration with you. This is the time to ask questions and express any concerns. Nothing asked during the test should come as a surprise.

During chart collection (typically 20–40 minutes for data recording), the question sequence will be repeated three to five times. Each repetition is called a "chart." Between charts, the examiner may discuss the comparison questions with you to maintain their psychological salience — a practice that Honts (1999) found significantly improves test accuracy [7]Verified The Discussion of Comparison Questions Between List Repetitions Is Associated with Increased Test Accuracy
Confirms between-chart discussion of comparison questions significantly improves CQT accuracy.

You will typically encounter two or three comparison questions per chart, interspersed with relevant questions and irrelevant questions. The entire examination, including pre-test interview, chart collection, and post-test discussion, usually takes 90 to 120 minutes.

For specific scenarios, see our guides on marriage and infidelity polygraph tests or how polygraph testing saved a North Carolina couple's marriage.