The AFMGQT is a workhorse format across federal screening. This guide breaks down how the Air Force Modified General Question Test structures a lie detector test.
A comprehensive examination of the AFMGQT's origins, question structure, scoring methodology, empirical accuracy data, and its role within federal polygraph practice for military and intelligence applications.
TL;DR — The Short Version
- The AFMGQT is a widely used federal polygraph technique evolved from the Reid Technique (1947) and Backster's Zone Comparison Technique (1960), validated for military and intelligence applications.
- Five question types — sacrifice relevant, primary relevant, secondary relevant, comparison, and irrelevant — work together to differentiate truthful from deceptive examinees.
- The landmark 2008 validation study by Senter, Waller, and Krapohl demonstrated a total accuracy rate of 83.8% and a definitive accuracy of 84.9%, significantly exceeding chance levels.
- Extended analyses using the Empirical Scoring System (ESS) achieved 88.2% accuracy, while the OSS-3 computer algorithm reached 90.2% correct decisions.
- Nelson, Handler, Oelrich, and Cushman (2014) proposed an event-specific format that eliminated outside-issue questions after research failed to support the super-dampening hypothesis.
- The AFMGQT is comparable to the Federal Zone Comparison Test and You-Phase formats, with key differences in question placement and adaptability to both multi-facet and screening contexts.
Who This Guide Is For
- Polygraph examiners seeking to understand AFMGQT methodology and its validated accuracy data
- Military personnel and intelligence professionals preparing for or administering polygraph examinations
- Polygraph students and trainees studying federal examination techniques
- Researchers and academics analyzing polygraph accuracy, scoring systems, and methodology
- Defense attorneys and legal professionals who encounter military polygraph evidence
- Security clearance candidates who may undergo AFMGQT-style examinations
Origins and Evolution of the AFMGQT
Foundational Techniques: Reid and Backster
The Air Force Modified General Question Test (AFMGQT) represents one of the most significant and widely deployed polygraph methodologies within the U.S. federal government. Its development traces back to two foundational polygraph techniques that shaped modern comparison question testing.
The first is the Reid Technique, developed by John E. Reid, a polygraph expert and former Chicago police officer, beginning in the late 1940s [1]Verified A revised questioning technique in lie detection tests
Confirms John E. Reid developed the comparison question technique for polygraph testing in 1947. Reid published his landmark paper 'A revised questioning technique in lie detection tests' in the Journal of Criminal Law and Criminology in 1947 [1]Verified A revised questioning technique in lie detection tests
Confirms John E. Reid developed the comparison question technique for polygraph testing in 1947, establishing the framework for structured polygraph questioning by introducing the concept of comparison (control) questions as a diagnostic tool.
The second foundational method is Cleve Backster's Zone Comparison Technique (ZCT), introduced in 1960 [2]Verified Zone Comparison Technique
Confirms Cleve Backster developed the Zone Comparison Technique in 1960/1963 as the first technique to incorporate numerical scoring. Backster, who had worked with the CIA and later founded his own lie detection school, refined the comparison question approach by organizing questions into distinct 'zones' — the red zone (relevant questions), the green zone (comparison questions), and the black zone (other questions) [2]Verified Zone Comparison Technique
Confirms Cleve Backster developed the Zone Comparison Technique in 1960/1963 as the first technique to incorporate numerical scoring. Critically, Backster's ZCT was the first polygraph technique in general use to incorporate numerical chart analysis, transforming test data analysis from subjective evaluation to an objective, quantifiable scoring process [2]Verified Zone Comparison Technique
Confirms Cleve Backster developed the Zone Comparison Technique in 1960/1963 as the first technique to incorporate numerical scoring.
For a deeper look at how these foundational methods influenced modern practice, see our guide on APA validated polygraph techniques.
Development of the USAF-MGQT
The Modified General Question Technique (MGQT) evolved as a de facto family of polygraph techniques resulting from various modifications of the General Question Technique (Reid, 1947) and the Zone Comparison Technique (Backster, 1963) [3]Verified Short Report: Criterion Validity of the United States Air Force Modified General Question Technique and Iraqi Scorers
Confirms the USAF-MGQT is a modern CQT variant widely used in federal government with 84.9% definitive accuracy. The U.S. Air Force Modified General Question Technique (USAF-MGQT), as formalized by the Department of Defense Polygraph Institute (DoDPI) in 2006, is a modern variant of the Comparison Question Test (CQT) that became widely used due to its efficient structure and its capability to adapt to both multi-facet investigative needs and multi-issue screening contexts [3]Verified Short Report: Criterion Validity of the United States Air Force Modified General Question Technique and Iraqi Scorers
Confirms the USAF-MGQT is a modern CQT variant widely used in federal government with 84.9% definitive accuracy.
The AFMGQT uses relevant, probable-lie, sacrifice relevant, and irrelevant questions, and some versions also permit the use of directed-lie comparison questions [4]Verified Terminology Reference for the Science of Psychophysiological Detection of Deception
Confirms AFMGQT uses relevant, probable-lie, sacrifice relevant, and irrelevant questions with optional directed-lie comparison questions. Two closely related variants of the AFMGQT exist, with minor structural differences between them, though there is no evidence to suggest that the performance of one version is superior to the other [5]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms AFMGQT is among 14 validated techniques and accuracy satisfactory for paired testing with ESS scoring.
The definitive validation study was published in 2008 by Stuart Senter, James Waller, and Donald Krapohl in the journal Polygraph [6]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants. This controlled laboratory study provided the first direct empirical evidence for the diagnostic value of the AFMGQT format, finding that it produced total and definitive accuracy rates that significantly exceeded chance levels for both truthful and deceptive participants [6]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants. The Air Force Office of Special Investigations (AFOSI), established in 1948, has been the primary operational user of this technique within Air Force security operations [7]Verified Air Force Office of Special Investigations Fact Sheet
Confirms AFOSI has been the Air Force's major investigative service since 1948 and directs the USAF polygraph program. Learn more about this in our guide on the AFOSI polygraph program.
AFMGQT Question Structure and Types
Overview of the Five Question Categories
The effectiveness of any polygraph technique depends fundamentally on its question structure, and the AFMGQT's design reflects decades of refinement aimed at maximizing diagnostic accuracy. The technique employs five distinct categories of questions, each serving a specific psychophysiological purpose within the examination [4]Verified Terminology Reference for the Science of Psychophysiological Detection of Deception
Confirms AFMGQT uses relevant, probable-lie, sacrifice relevant, and irrelevant questions with optional directed-lie comparison questions.
When properly constructed and administered, these question types work in concert to create conditions under which truthful and deceptive examinees produce distinguishably different patterns of physiological arousal. The question structure is not arbitrary — each type occupies a specific position within the test sequence, and the order has been empirically validated to optimize the comparison between responses to relevant questions and comparison questions [8]Verified The Polygraph and Lie Detection - Appendix A: Polygraph Questioning and Techniques
Confirms the zone comparison test was developed by Backster (1963) with three zones and was the first CQT to use numerical scoring.
Polygraph examiners who administer the AFMGQT must undergo specialized training. The pre-test interview is particularly critical, as it is during this phase that the examiner introduces each question, ensures understanding, and calibrates comparison questions appropriately. For more on how physiological responses are measured, see our guide on cardiovascular arousal in polygraph testing.
1. Sacrifice Relevant Questions
Sacrifice relevant questions are the first relevant-sounding questions presented in the test sequence. Their primary purpose is to absorb the initial orienting response — the natural spike in physiological activity that occurs when an examinee hears a question related to the topic under investigation for the first time during data collection.
Because this orienting response occurs regardless of truthfulness, sacrifice relevant questions are not scored. They serve as a physiological buffer that stabilizes the examinee's baseline before the scored questions begin. Understanding why certain questions are excluded from scoring is essential for appreciating the technique's diagnostic logic. For more on how different question types function, see our guide on stimulation tests vs. acquaintance tests.
2. Primary Relevant Questions
Primary relevant questions are the core diagnostic questions that directly address the examinee's involvement in the specific incident or issue under investigation. They are carefully worded to be clear, unambiguous, and answerable with a 'yes' or 'no' response.
In event-specific AFMGQT configurations, two or three primary relevant questions are typically used [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8. The relevant questions are positioned at specific locations within the question sequence — positions 4, 6, and 8 for the proposed event-specific AFMGQT format, compared to positions 5 and 7 for You-Phase formats and positions 5, 7, and 10 for ZCT formats [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8.
3. Secondary Relevant Questions
Secondary relevant questions address indirect involvement in the matter under investigation. Rather than asking about direct participation, they explore whether the examinee has knowledge of who committed the act, aided or abetted the perpetrator, or has access to evidence linking them to the incident.
Secondary relevant questions expand the diagnostic scope of the examination without diluting the focus on the primary issue. In multi-facet configurations, these questions allow examiners to explore different dimensions of an investigation within a single test session.
4. Comparison Questions (Control Questions)
Comparison questions are the diagnostic counterpart to relevant questions. They are designed to be broad, somewhat vague, and related to general integrity or past behavior — crafted so that most people would feel some uncertainty or discomfort when answering 'no.'
For truthful examinees, comparison questions should produce stronger physiological responses than relevant questions, because the comparison questions touch on areas of genuine concern while the relevant questions do not. For deceptive examinees, the relevant questions should produce stronger responses. This differential pattern is the basis for polygraph diagnosis [10]Verified Comparison of Relevant/Irrelevant and Modified General Question Technique Structures in a Split Counterintelligence-Suitability Phase Polygraph Examination
Confirms that question format substantially influences both diagnostic outcomes and security-relevant information obtained.
Learn more about how these response patterns are identified and evaluated in our guide on deceptive reaction zones in polygraph scoring and non-deceptive response patterns.
5. Irrelevant Questions
Irrelevant questions are neutral, factual questions entirely unrelated to the investigation and known to be truthfully answered. Their purpose is to establish a resting physiological baseline and to provide the examinee with recovery time between emotionally significant questions.
Irrelevant questions help prevent response carryover effects and maintain the integrity of the physiological data. Understanding the role of each question type is essential for appreciating how the AFMGQT produces reliable diagnostic information. For more on factors that can affect test data, see our guide on 5 things that contaminate polygraph exam results.
Event-Specific Modifications and Adaptations
The Shift to Single-Issue Testing
One of the most significant developments in the AFMGQT's history has been its adaptation for event-specific testing. While the original AFMGQT format was designed to address multiple issues within a single examination, the trend in federal polygraph practice has moved increasingly toward single-issue testing — examinations focused on one specific incident, allegation, or security concern.
Nelson, Handler, Oelrich, and Cushman (2014) described the proposed use of the AFMGQT as an event-specific or single-issue exam format, which would allow all scores to be totaled for a single diagnostic decision [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8. This report, produced by the APA Research Committee and accepted by the APA Board of Directors, examined the generalizability of existing scientific knowledge to the AFMGQT when used in an event-specific configuration [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8.
The event-specific AFMGQT targets scenarios such as security breaches, unauthorized disclosures of classified information, sabotage allegations, and other incident-specific investigations common in Air Force operations. The event-specific configuration closely aligns the AFMGQT with the Federal Zone Comparison Test and You-Phase formats used across other federal agencies.
Benefits of the Event-Specific Format
In the event-specific AFMGQT, the test typically employs two or three primary relevant questions, each bracketed by comparison questions. This bracketing ensures that every relevant question has an adjacent comparison question against which it can be evaluated, creating the paired comparison structure fundamental to zone comparison methodology [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8.
By reducing the number of questions and focusing them on a single issue, examiners can collect cleaner physiological data across multiple chart presentations — typically three or more charts are collected per examination. Each chart represents a complete cycle through the question sequence, and consistent response patterns across multiple charts increase confidence in the final diagnosis.
This modification has proven especially valuable in military contexts where the stakes are exceptionally high and the consequences of both false positive and false negative outcomes can have serious national security implications. For additional context on high-stakes polygraph examinations, see our dedicated guide.
The Exclusion of Outside-Issue (Symptomatic) Questions
Why Outside-Issue Questions Were Removed
One of the most consequential changes in the event-specific AFMGQT is the elimination of outside-issue questions, also known as symptomatic questions. In traditional polygraph formats, these questions were included to detect whether an examinee harbored concerns about topics unrelated to the test's primary focus.
The 2014 APA Research Committee report by Nelson, Handler, Oelrich, and Cushman conducted a thorough review of the published evidence on the validity of the super-dampening hypothesis and the effectiveness of outside-issue questions [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8. Their conclusion was clear: empirical evidence regarding the effectiveness of outside-issue questions is confounded and therefore uninformative, and test question formats that do not include outside-issue questions have been shown to produce mean accuracy rates that equal or exceed those of formats that do include them [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8.
Anecdotal reports from polygraph examiners testing outside the U.S. also suggest that symptomatic questions have often proven problematic with other cultures [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8. By removing outside-issue questions, the event-specific AFMGQT became a more streamlined, focused examination format.
The Super-Dampening Hypothesis and Its Rejection
The super-dampening hypothesis, referred to by Backster (2001) as the 'super-dampening concept,' proposed that if an examinee had a major unrelated concern, that concern could suppress their physiological responses to relevant questions, potentially leading to incorrect diagnoses [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8.
The 2014 research review found that this hypothesis failed to meet basic scientific standards. Research on outside-issue questions and the super-dampening hypothesis failed to conclusively demonstrate the hypothesized effects [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8. The physiological response to direct, personally relevant questioning about the specific issue under investigation proved robust enough to maintain diagnostic value even in the presence of extraneous psychological stressors.
This led to a clear recommendation that polygraph formats like the AFMGQT should exclude outside-issue questions because they add complexity without improving accuracy. The rejection reinforced a broader trend toward evidence-based practice in the polygraph field. For more on how memory and recall affect polygraph testing, see our dedicated guide.
Comparison with Other Federal Polygraph Techniques
AFMGQT vs. Federal ZCT vs. You-Phase
The AFMGQT exists within an ecosystem of federal polygraph techniques that share common methodological ancestry in comparison question testing. The most directly comparable techniques are the Federal Zone Comparison Test (FZCT) and the You-Phase format.
All three share the fundamental diagnostic logic of comparison question testing: using the differential between responses to relevant and comparison questions to determine truthfulness. The primary differences lie in question placement, sequence, and specific scoring conventions.
The AFMGQT positions relevant questions at positions 4, 6, and 8, while You-Phase formats place them at positions 5 and 7, and the ZCT at positions 5, 7, and 10 [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8. The AFMGQT is distinguished by its capability to adapt easily to both multi-facet investigative needs and multi-issue screening contexts [3]Verified Short Report: Criterion Validity of the United States Air Force Modified General Question Technique and Iraqi Scorers
Confirms the USAF-MGQT is a modern CQT variant widely used in federal government with 84.9% definitive accuracy, whereas the You-Phase is regarded primarily as a technique for specific-issue testing [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8.
Validation evidence for the AFMGQT can be generalized to structurally similar techniques like the LEPET and Utah MGQT when scored with the same validated test data analysis (TDA) methods [5]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms AFMGQT is among 14 validated techniques and accuracy satisfactory for paired testing with ESS scoring. Agencies like the CIA, FBI, and DEA each use their preferred variants of comparison question testing, but the underlying principles remain consistent across all validated federal formats. For more on specific agency testing procedures, see our CIA polygraph examiner career guide and CBP polygraph examiner career guide.
Scoring Methods and Reliability
Three-Position vs. Seven-Position Scoring
The AFMGQT can be scored using multiple validated test data analysis (TDA) models. The two primary manual scoring approaches are the traditional seven-position scale and the newer three-position scale used in the Empirical Scoring System (ESS) [11]Verified Criterion validity of the Empirical Scoring System and the Objective Scoring System, version 3 with the USAF Modified General Question Technique
Confirms manual ESS achieved 88.2%, automated ESS 89.7%, and OSS-3 reached 90.2% correct decisions with USAF-MGQT.
The seven-position scale assigns scores from -3 to +3 across three physiological channels for each relevant-comparison question pair. The three-position ESS model simplifies this to -1, 0, or +1, incorporating statistically optimal cutscores and two-stage decision rules [11]Verified Criterion validity of the Empirical Scoring System and the Objective Scoring System, version 3 with the USAF Modified General Question Technique
Confirms manual ESS achieved 88.2%, automated ESS 89.7%, and OSS-3 reached 90.2% correct decisions with USAF-MGQT. Krapohl (1998) provided early comparative data between these two scales [11]Verified Criterion validity of the Empirical Scoring System and the Objective Scoring System, version 3 with the USAF Modified General Question Technique
Confirms manual ESS achieved 88.2%, automated ESS 89.7%, and OSS-3 reached 90.2% correct decisions with USAF-MGQT.
Research on the USAF-MGQT has found that criterion accuracy of three-position scores was significantly greater than chance, with no significant differences in test sensitivity to deception compared to results from the seven-position and ESS TDA models [12]Verified Criterion Validity of the United States Air Force Modified General Question Technique and Three Position Scoring
Confirms three-position scoring accuracy was significantly greater than chance but specificity was weaker than seven-position and ESS models. However, test specificity — the ability to correctly identify truthful examinees — was significantly weaker for the three-position model compared to the other TDA models, and truthful case inconclusive results were significantly higher [12]Verified Criterion Validity of the United States Air Force Modified General Question Technique and Three Position Scoring
Confirms three-position scoring accuracy was significantly greater than chance but specificity was weaker than seven-position and ESS models. For a deeper exploration of the three-position scale, see our complete examiner guide to 3-position scoring.
The Empirical Scoring System (ESS)
The Empirical Scoring System (ESS) is an evidence-based normative system for test data analysis that provides procedural descriptions for all aspects of the scoring model, including physiological features, mathematical transformations, decision rules, and cutscores based on normative data [13]Verified Using the Empirical Scoring System
Describes the ESS procedures including physiological features, mathematical transformations, decision rules, and normative cutscores.
Blalock (2011) demonstrated that the ESS achieved outstanding accuracy with the USAF-MGQT: manual ESS achieved 88.2% unweighted accuracy, automated ESS achieved 89.7%, and the OSS-3 computer algorithm reached 90.2% correct decisions, with Pearson correlations between scoring models exceeding r =.93 [11]Verified Criterion validity of the Empirical Scoring System and the Objective Scoring System, version 3 with the USAF Modified General Question Technique
Confirms manual ESS achieved 88.2%, automated ESS 89.7%, and OSS-3 reached 90.2% correct decisions with USAF-MGQT. These results confirmed that modern scoring approaches can significantly enhance the diagnostic power of the AFMGQT.
Bootstrap analysis of examiner trainee scores with the ESS resulted in a mean accuracy rate of 90.1% (95% CI = 83.8% to 95.8%), excluding just 3.3% inconclusives [12]Verified Criterion Validity of the United States Air Force Modified General Question Technique and Three Position Scoring
Confirms three-position scoring accuracy was significantly greater than chance but specificity was weaker than seven-position and ESS models. This level of performance demonstrates that even inexperienced examiners can achieve high accuracy when using empirically validated scoring methods.
Automated Scoring: The OSS-3 Algorithm
The Objective Scoring System, version 3 (OSS-3) is a computer algorithm that analyzes polygraph data without using integer scores or integer cutscores, providing an open-source, objective, and scientifically defensible method for analyzing polygraph data [14]Verified Extended Analysis of Senter, Waller and Krapohl's USAF MGQT Examination Data with the Empirical Scoring System and the Objective Scoring System, version 3
Confirms no significant differences between laboratory and field USAF-MGQT scores, supporting operational generalizability. Developed by Raymond Nelson with contributions from Donald Krapohl and Mark Handler, OSS-3 uses ratio transformation and bootstrapping techniques to reduce variability in scoring.
In the extended analysis of Senter, Waller, and Krapohl's USAF-MGQT data, multi-variate analysis found no significant differences between total and subtotal scores of the laboratory sample and those from a sample of field investigation cases using the same technique [14]Verified Extended Analysis of Senter, Waller and Krapohl's USAF MGQT Examination Data with the Empirical Scoring System and the Objective Scoring System, version 3
Confirms no significant differences between laboratory and field USAF-MGQT scores, supporting operational generalizability. This finding is crucial because it suggests that laboratory validation results generalize to real-world operational conditions.
Accuracy and Empirical Findings
The 2008 Validation Study
The definitive validation of the AFMGQT was published by Senter, Waller, and Krapohl (2008) in the journal Polygraph [6]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants. This controlled laboratory study — conducted at the Department of Defense Polygraph Institute at Fort Jackson, SC — used a sample of approximately 69 participants and tested the diagnostic value of the AFMGQT format [6]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants.
The study's key findings were impressive. The AFMGQT achieved a total accuracy rate of 83.8% for all decisions, and a definitive accuracy rate of 84.9% when excluding inconclusive results [6]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants. Decision accuracy was significantly above chance levels for both truthful participants (91.7%) and deceptive participants (75.8%), with only 1.5% of cases producing no opinion decisions [6]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants. These results provided strong empirical support for the technique's operational use.
Senter, Dollins, and Krapohl also published a separate 2008 study on the effectiveness of the AFMGQT format that further validated the military's standard examination technique [15]Verified Effectiveness of the Air Force Modified General Question Test Format
Validated the USAF-MGQT format with total accuracy of 83.8% and definitive accuracy of 84.9%.
Extended Analyses and Replication Studies
Subsequent studies expanded the evidence base considerably. Nelson and Blalock (2016) conducted an extended analysis of the original Senter, Waller, and Krapohl data using the ESS and OSS-3, finding that field investigation cases produced subtotal scores of greater absolute value than the laboratory sample, suggesting that real-world conditions may actually enhance test performance [14]Verified Extended Analysis of Senter, Waller and Krapohl's USAF MGQT Examination Data with the Empirical Scoring System and the Objective Scoring System, version 3
Confirms no significant differences between laboratory and field USAF-MGQT scores, supporting operational generalizability.
Nelson, Handler, Morgan, and O'Burke (2012) examined the criterion validity of the USAF-MGQT with Iraqi polygraph examiners, reporting a mean blind-scoring criterion accuracy level of 84.9% excluding inconclusive results [14]Verified Extended Analysis of Senter, Waller and Krapohl's USAF MGQT Examination Data with the Empirical Scoring System and the Objective Scoring System, version 3
Confirms no significant differences between laboratory and field USAF-MGQT scores, supporting operational generalizability. Automated scoring models (ESS and OSS-3) achieved accuracy levels of 89.5% and 90.2%, respectively [14]Verified Extended Analysis of Senter, Waller and Krapohl's USAF MGQT Examination Data with the Empirical Scoring System and the Objective Scoring System, version 3
Confirms no significant differences between laboratory and field USAF-MGQT scores, supporting operational generalizability.
A comprehensive meta-analysis by Honts, Handler, Shaw, and Gougler (2021) — the largest meta-analysis ever conducted on the Comparison Question Test, analyzing 138 datasets — found that motivation level showed a positive linear relationship with accuracy [16]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
The largest CQT meta-analysis (138 datasets) finding motivation level positively correlated with accuracy. This finding has particular relevance for the AFMGQT, as military and intelligence examinees typically face high-stakes consequences that increase motivation.
Earlier MGQT Validation Research
The empirical foundation for the MGQT family extends beyond the 2008 AFMGQT study. Podlesny (1993) validated an expanded-issue MGQT in a simulated distributed-crime-roles context, finding that the technique successfully differentiated guilty participants from innocent participants at above-chance accuracy levels [17]Verified Validity of an Expanded-Issue (Modified General Question) Polygraph Technique in a Simulated Distributed-Crime-Roles Context
Confirms the expanded-issue MGQT differentiated guilty from innocent participants with 84.7% and 94.7% accuracy respectively. Excluding inconclusives, decisions based on total numerical scores were 84.7% correct for the guilty group and 94.7% correct for the innocent group [17]Verified Validity of an Expanded-Issue (Modified General Question) Polygraph Technique in a Simulated Distributed-Crime-Roles Context
Confirms the expanded-issue MGQT differentiated guilty from innocent participants with 84.7% and 94.7% accuracy respectively.
Senter (2003) conducted a systematic exploration of decision rules for the MGQT format, identifying optimal cutscores and two-stage decision rules that maximized accuracy while minimizing inconclusive rates [18]Verified Modified General Question Test Decision Rule Exploration
Identified optimal cutscores and two-stage decision rules that maximized accuracy while minimizing inconclusive rates for the MGQT. Ansley (1998) also examined the validity of the MGQT format, contributing to the growing body of evidence supporting this technique family [19]Verified The validity of the modified general question test (MGQT)
Foundational research on the validity of the MGQT format.
Weaver and Garwood (1985) compared the relevant/irrelevant and MGQT technique structures in a split counterintelligence-suitability phase polygraph examination, demonstrating that question format substantially influences both diagnostic outcomes and the nature of security-relevant information obtained [20]Verified Comparison of Relevant/Irrelevant and Modified General Question Technique Structures
Demonstrates that question format substantially influences diagnostic outcomes in counterintelligence polygraph testing.
Strengths and Practical Applications
Strengths of the AFMGQT
The AFMGQT offers several distinct advantages that make it well-suited for federal and military applications. Its efficient structure is based on generally accepted valid principles for CQT test construction [3]Verified Short Report: Criterion Validity of the United States Air Force Modified General Question Technique and Iraqi Scorers
Confirms the USAF-MGQT is a modern CQT variant widely used in federal government with 84.9% definitive accuracy. The technique is remarkably versatile, capable of adapting to multi-facet investigative needs, multi-issue screening contexts, and event-specific diagnostic testing [3]Verified Short Report: Criterion Validity of the United States Air Force Modified General Question Technique and Iraqi Scorers
Confirms the USAF-MGQT is a modern CQT variant widely used in federal government with 84.9% definitive accuracy.
The AFMGQT has achieved accuracy rates that satisfy APA requirements for paired testing when scored with the Empirical Scoring System [5]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms AFMGQT is among 14 validated techniques and accuracy satisfactory for paired testing with ESS scoring. Published and replicated studies indicate that the AFMGQT scored with the seven-position TDA model produces mean accuracy rates over 80% with mean inconclusive rates lower than 20% [5]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms AFMGQT is among 14 validated techniques and accuracy satisfactory for paired testing with ESS scoring. When scored with modern automated systems like the OSS-3, accuracy rates approaching and exceeding 90% have been demonstrated [11]Verified Criterion validity of the Empirical Scoring System and the Objective Scoring System, version 3 with the USAF Modified General Question Technique
Confirms manual ESS achieved 88.2%, automated ESS 89.7%, and OSS-3 reached 90.2% correct decisions with USAF-MGQT.
The technique's structural similarity to the LEPET and Utah MGQT means that validation evidence for the AFMGQT is generalizable to these related techniques when scored with the same TDA methods [5]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms AFMGQT is among 14 validated techniques and accuracy satisfactory for paired testing with ESS scoring, expanding its utility across different operational contexts.
Military and Intelligence Applications
The AFMGQT serves a critical role in Air Force security operations administered by AFOSI, which directs the USAF polygraph program [7]Verified Air Force Office of Special Investigations Fact Sheet
Confirms AFOSI has been the Air Force's major investigative service since 1948 and directs the USAF polygraph program. AFOSI polygraph examiners conduct examinations supporting criminal investigations, counterintelligence operations, and force protection issues across military installations worldwide [7]Verified Air Force Office of Special Investigations Fact Sheet
Confirms AFOSI has been the Air Force's major investigative service since 1948 and directs the USAF polygraph program.
Experienced AFOSI agents selected for polygraph duties attend a 14-week Department of Defense course to acquire the specialized skills required for this work [7]Verified Air Force Office of Special Investigations Fact Sheet
Confirms AFOSI has been the Air Force's major investigative service since 1948 and directs the USAF polygraph program. The AFMGQT is deployed in scenarios ranging from espionage investigations to unauthorized disclosure of classified materials, personnel reliability assessments, and security breach investigations.
For those preparing for federal polygraph examinations, understanding that different agencies use different question formats — but that all validated formats rely on the same fundamental comparison question logic — can help demystify the process. Our guides on the ATF polygraph exam and calming techniques for polygraph examinations provide practical preparation advice.
Ethical and Methodological Considerations
Evidence-Based Practice
The evolution of the AFMGQT exemplifies the polygraph field's commitment to evidence-based practice. The removal of outside-issue questions, the adoption of empirically validated scoring systems, and the ongoing refinement of decision rules all reflect a discipline that takes scientific rigor seriously [9]Verified APA Research Committee Report: Proposed Usage for an Event-specific AFMGQT Test Format
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8.
The APA's meta-analytic survey of criterion accuracy established clear boundary requirements for technique validation: techniques must demonstrate accuracy rates significantly greater than chance across multiple replicated studies [5]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms AFMGQT is among 14 validated techniques and accuracy satisfactory for paired testing with ESS scoring. The AFMGQT meets these requirements when scored with validated TDA methods.
Iacono and Ben-Shakhar (2019) provided an updated review of CQT research, noting the importance of continued methodological improvement in polygraph studies [21]Verified Current Status of Forensic Lie Detection with the Comparison Question Technique: An Update
Updated review of CQT research reaffirming NRC (2003) conclusions and stimulating debate on polygraph accuracy estimation. The polygraph community has responded to these critiques by developing increasingly rigorous validation protocols, as demonstrated by the progression from the original 2008 validation study to the multiple replication and extension studies that followed.
For more on how the profession maintains quality standards, see our guides on European Polygraph Association training standards and how a polygraph works.
Ongoing Research and Future Directions
Research on the AFMGQT continues to evolve. The Senter et al. (2008) validation study itself recommended that further research be conducted to explore the effectiveness of different variants of the technique, including different scenarios, question types, and question configurations [6]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants. Additional and continuous research is recommended to expand the body of knowledge pertaining to the variety of uses afforded by the AFMGQT [6]Verified Air Force Modified General Question Test Validation Study
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants.
Areas of active investigation include optimizing decision rules and cutscores for manually scoring the USAF-MGQT, continued comparison of ESS and seven-position models, and exploring the technique's performance across diverse populations and operational contexts [12]Verified Criterion Validity of the United States Air Force Modified General Question Technique and Three Position Scoring
Confirms three-position scoring accuracy was significantly greater than chance but specificity was weaker than seven-position and ESS models. The Honts et al. (2021) comprehensive meta-analysis also identified areas where further research could refine our understanding of comparison question test accuracy [16]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
The largest CQT meta-analysis (138 datasets) finding motivation level positively correlated with accuracy.
Pros
- Validated accuracy of 83.8% total and 84.9% definitive in the landmark 2008 study, with automated scoring reaching 90.2%
- Highly versatile — adapts to multi-facet investigations, multi-issue screening, and event-specific diagnostic testing
- Evidence-based design with outside-issue questions removed after research showed they added no diagnostic value
- Structurally similar to LEPET and Utah MGQT, allowing generalization of validation evidence across techniques
- Multiple validated scoring methods available including manual seven-position, three-position ESS, and automated OSS-3
- Strong institutional support from the Department of Defense and Air Force Office of Special Investigations
Cons
- Three-position scoring model shows significantly weaker specificity for correctly identifying truthful examinees compared to seven-position and ESS models
- The 2008 validation study used a laboratory sample of approximately 69 participants; larger field studies would strengthen the evidence base
- Real-world base rates of deception likely differ from the approximately 50% used in laboratory studies
- Inconclusive rates for human scorers (24.1%) are significantly higher than automated scoring models
Frequently Asked Questions
What is the AFMGQT and who uses it?
The Air Force Modified General Question Test (AFMGQT) is a polygraph examination technique widely used across the U.S. federal government, particularly by the Air Force Office of Special Investigations (AFOSI). It is a modern variant of the Comparison Question Test that evolved from the Reid Technique (1947) and Backster's Zone Comparison Technique (1960). The technique is used for criminal investigations, counterintelligence operations, personnel reliability assessments, and security breach investigations.
How accurate is the AFMGQT?
The landmark 2008 validation study by Senter, Waller, and Krapohl found a total accuracy rate of 83.8% and a definitive accuracy rate of 84.9% (excluding inconclusive results). Decision accuracy was 91.7% for truthful participants and 75.8% for deceptive participants. When scored with the Empirical Scoring System (ESS), accuracy reached 88.2%, and the automated OSS-3 algorithm achieved 90.2% correct decisions.
What types of questions are used in the AFMGQT?
The AFMGQT uses five question categories: sacrifice relevant questions (absorb the initial orienting response and are not scored), primary relevant questions (directly address the investigation issue), secondary relevant questions (address indirect involvement), comparison questions (broad integrity questions that serve as the diagnostic benchmark), and irrelevant questions (neutral questions that establish a physiological baseline). Some versions also permit directed-lie comparison questions.
What is the difference between the AFMGQT and the Federal Zone Comparison Test?
Both techniques share the fundamental logic of comparison question testing, but they differ in question placement. In the event-specific AFMGQT, relevant questions are positioned at locations 4, 6, and 8 within the question sequence, while ZCT formats place them at positions 5, 7, and 10. The AFMGQT also does not include symptomatic (outside-issue) questions in its event-specific format, and it can adapt to both multi-facet and screening contexts, while the You-Phase ZCT is primarily used for specific-issue testing.
What is the super-dampening hypothesis and why was it rejected?
The super-dampening hypothesis, attributed to Cleve Backster, proposed that an examinee's preoccupation with an unrelated outside concern could suppress their physiological responses to relevant questions, leading to incorrect diagnoses. A 2014 APA Research Committee review by Nelson, Handler, Oelrich, and Cushman found that research on this hypothesis failed to conclusively demonstrate the hypothesized effects. Formats without outside-issue questions produced accuracy rates equal to or better than those with them, leading to the recommendation that the AFMGQT exclude these questions.
What scoring systems are used with the AFMGQT?
Three main scoring approaches are validated for use with the AFMGQT: the traditional seven-position scale (scoring from -3 to +3), the Empirical Scoring System (ESS) which uses a simplified three-position model with optimal cutscores, and the OSS-3 automated computer algorithm. Research shows that automated scoring tends to outperform manual scoring, with the OSS-3 achieving up to 90.2% accuracy.
How does the event-specific AFMGQT differ from the standard format?
The event-specific AFMGQT, proposed by Nelson, Handler, Oelrich, and Cushman in 2014, focuses on a single incident or allegation rather than multiple issues. It eliminates outside-issue (symptomatic) questions, uses two or three primary relevant questions bracketed by comparison questions, and allows all scores to be totaled for a single diagnostic decision. This format produces cleaner physiological data and aligns the AFMGQT with other validated event-specific federal techniques.
Is the AFMGQT validated for pre-employment screening?
The AFMGQT is used in both multi-facet event-specific contexts and multi-issue screening contexts across the federal government. However, validation evidence for event-specific diagnostic variants cannot automatically be generalized to multi-issue screening variants, as these are scored and interpreted differently (independent vs. non-independent criterion variance). The APA meta-analysis notes that screening test accuracy research remains more limited than diagnostic testing research.
Sources & References
Confirms John E. Reid developed the comparison question technique for polygraph testing in 1947
Confirms Cleve Backster developed the Zone Comparison Technique in 1960/1963 as the first technique to incorporate numerical scoring
Confirms the USAF-MGQT is a modern CQT variant widely used in federal government with 84.9% definitive accuracy
Confirms AFMGQT uses relevant, probable-lie, sacrifice relevant, and irrelevant questions with optional directed-lie comparison questions
Confirms AFMGQT is among 14 validated techniques and accuracy satisfactory for paired testing with ESS scoring
Confirms the AFMGQT achieved 83.8% total accuracy and 84.9% definitive accuracy with 91.7% accuracy for truthful and 75.8% for deceptive participants
Confirms AFOSI has been the Air Force's major investigative service since 1948 and directs the USAF polygraph program
Confirms the zone comparison test was developed by Backster (1963) with three zones and was the first CQT to use numerical scoring
Confirms findings on outside-issue questions, super-dampening hypothesis rejection, and proposed event-specific AFMGQT format with question positions 4, 6, and 8
Confirms that question format substantially influences both diagnostic outcomes and security-relevant information obtained
Confirms manual ESS achieved 88.2%, automated ESS 89.7%, and OSS-3 reached 90.2% correct decisions with USAF-MGQT
Confirms three-position scoring accuracy was significantly greater than chance but specificity was weaker than seven-position and ESS models
Describes the ESS procedures including physiological features, mathematical transformations, decision rules, and normative cutscores
Confirms no significant differences between laboratory and field USAF-MGQT scores, supporting operational generalizability
Validated the USAF-MGQT format with total accuracy of 83.8% and definitive accuracy of 84.9%
The largest CQT meta-analysis (138 datasets) finding motivation level positively correlated with accuracy
Confirms the expanded-issue MGQT differentiated guilty from innocent participants with 84.7% and 94.7% accuracy respectively
Identified optimal cutscores and two-stage decision rules that maximized accuracy while minimizing inconclusive rates for the MGQT
Foundational research on the validity of the MGQT format
Demonstrates that question format substantially influences diagnostic outcomes in counterintelligence polygraph testing
Updated review of CQT research reaffirming NRC (2003) conclusions and stimulating debate on polygraph accuracy estimation
The first meta-analysis of the CQT, finding significant moderator effects for subject type, incentives, and decision policy
A free course on the Air Force Modified General Question Technique — multi-issue and event-specific applications, sequencing, and scoring.
Start the Free Course →This technique is part of a professional examiner's repertoire, so to book a real exam, find a lie detector test near you and check pricing.