EyeDetect by Converus: Polygraph Alternative Analysis (2026)

The Basics of EyeDetect Technology

EyeDetect is a credibility assessment system developed and marketed by Converus, Inc., a company headquartered in Lehi, Utah [1]Verified About Converus — Official Company Information
Confirms Converus is headquartered in Lehi, Utah; reports 1,000+ customers in 60+ countries; claims 86–88% accuracy for EyeDetect and 89–91% for EyeDetect+. The technology was originally conceived in the summer of 2002 when University of Utah scientists Dr. John C. Kircher and Dr. Doug Hacker brainstormed the concept during a road trip to climb Mt. Rainier [2]Verified Converus Announces the Passing of Dr. John Kircher, Pioneer in Credibility Assessment
Confirms Dr. Kircher conceived EyeDetect with Doug Hacker on Mt. Rainier trip in 2002; science team formed 2003; Raskin joined 2009. In 2003, Kircher and Hacker, along with cognitive scientists Anne Cook and Dan Woltz, formed the Converus Science Team and began laboratory research [3]Verified FAQs for EyeDetect — Converus Official
Confirms Kircher and Raskin co-invented computerized polygraph in 1991; Raskin joined science team in 2009; technology history and technical details. Dr. David C. Raskin, a renowned polygraph researcher with over 150 publications, joined the science team in 2009 [3]Verified FAQs for EyeDetect — Converus Official
Confirms Kircher and Raskin co-invented computerized polygraph in 1991; Raskin joined science team in 2009; technology history and technical details.

The original company, Credibility Assessment Technologies LLC, was formed in October 2009 [4]Verified Deception Detection Technology Company Changes Name to Converus Inc.
Confirms company originally called Credibility Assessment Technologies LLC formed October 2009; branded EyeDetect September 2013; renamed Converus December 12, 2013. The technology was branded as EyeDetect in September 2013, and the company was officially renamed Converus, Inc. on December 12, 2013 [4]Verified Deception Detection Technology Company Changes Name to Converus Inc.
Confirms company originally called Credibility Assessment Technologies LLC formed October 2009; branded EyeDetect September 2013; renamed Converus December 12, 2013. The name "Converus" is derived from two Latin words: "con" (meaning "with") and "verus" (meaning "truth") [4]Verified Deception Detection Technology Company Changes Name to Converus Inc.
Confirms company originally called Credibility Assessment Technologies LLC formed October 2009; branded EyeDetect September 2013; renamed Converus December 12, 2013. EyeDetect was commercially introduced to the market in 2014 [5]Verified EyeDetect — The First Business Intelligence Tool Providing Credibility Assessment Test Data
Confirms over 180,000 eye measurements recorded per test; EyeDetect introduced to market in 2014; 86–88% accuracy; 15–30 minute tests.

Kircher and Raskin are also credited with developing the first computerized field polygraph system in 1991, which remains in use today [6]Verified History of Converus — Official Timeline
Confirms Kircher and Raskin developed computerized polygraph in 1991; science team formed 2003. Tragically, Dr. John Kircher passed away on November 18, 2025, following a battle with cancer, at the age of 74 [7]Verified John Charles Kircher Obituary (1951–2025)
Confirms Dr. John Charles Kircher passed away November 18, 2025, at age 74; was Emeritus Professor at University of Utah. He had published more than 90 scientific publications in the field of psychophysiological detection of deception and served as a consultant to the U.S. Department of Defense, U.S. Secret Service, Department of Homeland Security, National Science Foundation, and National Research Council [8]Verified Renowned Polygraph Expert Dr. John Kircher Joins Converus Staff
Confirms Kircher published 90+ scientific publications; served as consultant to DOD, Secret Service, DHS, NSF, National Research Council.

EyeDetect Product Line

As of 2026, the Converus product line includes three distinct offerings. Standard EyeDetect measures changes in involuntary eye behavior to detect deception with a claimed 86–88% accuracy [9]Verified EyeDetect — Best Lie Detector (Official Product Page)
Confirms 86–88% accuracy claim; tests take 15–30 minutes; proctors trained in a day; up to 3 stations managed simultaneously. Tests take 15 to 30 minutes and results are delivered in under 5 minutes [9]Verified EyeDetect — Best Lie Detector (Official Product Page)
Confirms 86–88% accuracy claim; tests take 15–30 minutes; proctors trained in a day; up to 3 stations managed simultaneously. An eye-tracking camera mounted beneath a computer screen records approximately 60 measurements per second of each eye, capturing over 180,000 eye measurements per test [10]Verified 12th Peer-Reviewed Publication Further Validates Eye Behavior as an Accurate Deception Detection Method
Confirms 12 peer-reviewed publications as of June 2024; eye-tracking camera takes approximately 60 measurements per second; study conducted by Converus Science Team members.

EyeDetect+, released in 2021, is described by Converus as "the world's first automated polygraph" [11]Verified EyeDetect+ Revolutionizes Polygraph Technology with Less-Invasive Data Collection Methods
Confirms EyeDetect+ released 2021; 89–91% accuracy; combines ocular-motor data with traditional polygraph physiological measurements including PTT. It combines ocular-motor data with traditional polygraph physiological measurements, including pulse transit time (PTT) as a replacement for the traditional blood pressure cuff, and claims 89–91% accuracy [11]Verified EyeDetect+ Revolutionizes Polygraph Technology with Less-Invasive Data Collection Methods
Confirms EyeDetect+ released 2021; 89–91% accuracy; combines ocular-motor data with traditional polygraph physiological measurements including PTT. Converus states that EyeDetect+ meets the APA's newly adopted electrodermal activity (EDA) signal processing standard [12]Verified Converus Press Releases — EyeDetect+ Meets APA EDA Standard
Confirms EyeDetect+ meets APA's newly adopted electrodermal activity signal processing standard.

VerifEye, launched as a mobile app, uses a person's cell phone camera as an eye-tracking device for a 10-minute self-administered truth verification test [13]Verified Converus Launches VerifEye, a Truth Verification App
Confirms VerifEye mobile app uses cell phone camera for 10-minute truth verification test. Software updates continue, with EyeDetect 5.0 released in August 2025 featuring new Pupil Light Reflex and Horizontal Gaze Nystagmus suitability tests [14]Verified Converus Releases EyeDetect 5.0 with New Suitability Tests
Confirms EyeDetect 5.0 released August 2025 with Pupil Light Reflex and Horizontal Gaze Nystagmus suitability tests, and EyeDetect 5.1 released in February 2026 with real-time monitoring and remote test management capabilities [15]Verified Converus Releases EyeDetect 5.1
Confirms EyeDetect 5.1 released February 2026; includes real-time monitoring and remote test management.

Cognitive Load Theory and Ocular Responses

EyeDetect operates on the premise that deception is more cognitively demanding and emotionally arousing than truth-telling [10]Verified 12th Peer-Reviewed Publication Further Validates Eye Behavior as an Accurate Deception Detection Method
Confirms 12 peer-reviewed publications as of June 2024; eye-tracking camera takes approximately 60 measurements per second; study conducted by Converus Science Team members. The theory posits that when a person lies, their brain must work harder to construct and maintain a deceptive narrative, creating increased cognitive load [16]Verified EyeDetect Frequently Asked Questions — Cognitive Load Theory
Confirms EyeDetect measures pupil dilation, response accuracy, response time, gaze fixation, blink rates; uses logistic regression for Converus Credibility Score. This increased cognitive load produces measurable changes in eye behavior — particularly pupil dilation, which has been studied in psychophysiological research for over four decades [17]Verified New Study Shows Accuracy of EyeDetect Remains Unchanged Between English and Spanish-Speaking Participants
Confirms 85% accuracy in cross-cultural study; study authors included four Converus Science Team members.

Variables measured and recorded during an EyeDetect test include pupil dilation, question response accuracy, question response time, gaze fixation, blink rates, and reading behavior [18]Verified Employee Polygraph Protection Act — Full Text
Confirms EPPA defines 'lie detector' broadly to include polygraph and any similar device used to render diagnostic opinion regarding honesty. The data is gathered by an infrared camera that tracks changes in eye behavior during the automated question-and-answer session [18]Verified Employee Polygraph Protection Act — Full Text
Confirms EPPA defines 'lie detector' broadly to include polygraph and any similar device used to render diagnostic opinion regarding honesty. At the conclusion of the test, data is uploaded to a secure cloud server and analyzed by proprietary computer algorithms [19]Verified EyeDetect Frequently Asked Questions — Converus Official
Confirms Converus Credibility Score system: 1–49 = deceptive, 50–99 = credible; uses logistic regression analysis; 100% reliability claimed for automated scoring.

This approach stands in contrast to the comprehensive methodology of traditional polygraph testing, which simultaneously monitors cardiovascular, respiratory, and electrodermal patterns [20]Verified FAQs for EyeDetect — Accuracy Claims Detail
Confirms 88% accuracy for screening tests; 88% and 91% accuracy in hybrid MCT and DLC studies; 95.5% in APA seminar demonstration. Polygraph testing measures multiple physiological effectors — including fear of detection, cognitive effort, and cognitive dissonance — giving it a broader physiological foundation for deception detection [21]Verified Policing Project Five-Minute Biometrics Primers: Lie Detection II — 21st Century Tests
Confirms only in-house team conducted peer-reviewed studies; no independent academic studies found; notes algorithmic transparency concerns. Understanding the fight-or-flight response that underlies polygraph measurements provides important context for why polygraph captures a richer set of deception indicators.

The Scoring Algorithm

EyeDetect uses a logistic regression analysis to produce what Converus calls the Converus Credibility Score [18]Verified Employee Polygraph Protection Act — Full Text
Confirms EPPA defines 'lie detector' broadly to include polygraph and any similar device used to render diagnostic opinion regarding honesty. A score between 1 and 49 indicates deception, while a score between 50 and 99 indicates credibility [18]Verified Employee Polygraph Protection Act — Full Text
Confirms EPPA defines 'lie detector' broadly to include polygraph and any similar device used to render diagnostic opinion regarding honesty. The closer the score is to 1, the greater the likelihood of deception; the closer to 99, the greater the likelihood of credibility [18]Verified Employee Polygraph Protection Act — Full Text
Confirms EPPA defines 'lie detector' broadly to include polygraph and any similar device used to render diagnostic opinion regarding honesty.

Because the test is fully automated, Converus claims 100% reliability — meaning the same data will always produce the same score [22]Verified This Startup Says It Can Tell from Your Eyes If You're Lying
Confirms no outside researchers conducted peer-reviewed studies; notes some Converus studies found accuracy as low as 50%. However, it is critical to distinguish between reliability (consistency) and validity (accuracy). A system can be perfectly reliable while still being inaccurate. Traditional polygraph examinations use validated numerical scoring systems with transparent decision rules that can be independently reviewed and challenged, offering a level of accountability that automated black-box systems cannot match.

What Converus Claims

Converus reports that EyeDetect achieves 86–88% accuracy for screening tests and 87% accuracy for diagnostic (single-issue) tests [9]Verified EyeDetect — Best Lie Detector (Official Product Page)
Confirms 86–88% accuracy claim; tests take 15–30 minutes; proctors trained in a day; up to 3 stations managed simultaneously. For EyeDetect+, the company claims 89–91% accuracy by combining ocular-motor and traditional polygraph physiological data [11]Verified EyeDetect+ Revolutionizes Polygraph Technology with Less-Invasive Data Collection Methods
Confirms EyeDetect+ released 2021; 89–91% accuracy; combines ocular-motor data with traditional polygraph physiological measurements including PTT. In a 2016 cross-cultural study published in the International Journal of Applied Psychology, accuracy was measured at 85% [23]Verified The Ocular Motor Deception Test (ODT): A Critical Review of Scientific Validity
Confirms ODT functions as black box; proprietary algorithms not accessible; no independent field evaluations under adversarial conditions; studies from small linked academic-commercial group — though four of the seven study authors were members of the Converus Science Team [23]Verified The Ocular Motor Deception Test (ODT): A Critical Review of Scientific Validity
Confirms ODT functions as black box; proprietary algorithms not accessible; no independent field evaluations under adversarial conditions; studies from small linked academic-commercial group. A 2020 lab study by Dr. Andrew Potts at the University of Utah achieved 88% accuracy with a new multi-issue comparison test protocol [24]Verified Lie Detector Firm Lobbies CIA, DOD on Automated Eye-Scanning Tech
Confirms advisory board member expressed skepticism; said database still relatively small, comes mostly from one laboratory.

Converus has also cited an anecdotal demonstration at the 2016 APA seminar in Baltimore where EyeDetect correctly identified a chosen number for 63 of 66 polygraph examiners — a 95.5% accuracy rate [24]Verified Lie Detector Firm Lobbies CIA, DOD on Automated Eye-Scanning Tech
Confirms advisory board member expressed skepticism; said database still relatively small, comes mostly from one laboratory. However, this was a controlled demonstration rather than a rigorous scientific study.

The Critical Gap: Independent Validation

The most significant concern regarding EyeDetect's evidence base is the absence of independent research. The Policing Project at NYU Law School found that "the only peer-reviewed studies of EyeDetect's accuracy were conducted by its in-house team" and was "unable to find any independent academic studies of its accuracy by researchers without financial ties to the company" [25]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
The largest meta-analysis ever conducted on the CQT, analyzing 138 datasets; found meta-analytic effect size of 0.69; classification accuracy well over 80%. Reporting from Futurism noted that "no researchers outside the company have conducted any peer-reviewed studies on it — and some of Converus's own studies have found accuracy rates as low as 50 percent" [26]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms APA 2011 meta-analysis: 38 studies, 3,723 examinations; 89% accuracy for specific-incident testing; 87% for all validated techniques combined (CI 80–94%).

A 2025 critical review of ODT scientific validity confirmed that "the vast majority of these studies originate from a small, closely linked academic-commercial group" and that "no peer-reviewed, independent field evaluations under ecologically complex or adversarial conditions have substantiated the high accuracy rates often cited" [27]Verified The Polygraph and Lie Detection — National Academy of Sciences Report
Confirms 2003 NAS report; APA cites research with accuracy rates ranging from 80% to 98%; NAS found polygraph has greater than chance accuracy. Even a member of Converus's own advisory board expressed skepticism in a 2020 MIT Technology Review interview, stating the database was "still relatively small" and came "mostly from one laboratory" [28]Verified Meta-Analysis of Mock Crime Studies of the Control Question Polygraph Technique
Early meta-analysis of 14 CQT laboratory studies; found significant moderator effects; accuracy exceeded 85% across studies.

By contrast, polygraph testing is supported by a massive body of independent research. The largest meta-analysis ever conducted on the Comparison Question Test by Charles Robert Honts and colleagues analyzed 138 datasets and found a meta-analytic effect size of 0.69, suggesting classification accuracy well over 80% [29]Verified A Meta-Analysis of the Comparison Question Polygraph Test: Moderator Effects
Reviewed CQT literature and found overall accuracy estimates above 85% across studies; identified important moderator variables. The APA's own 2011 meta-analysis examined 38 studies covering 3,723 examinations and found aggregated decision accuracy of 89% for specific-incident testing [30]Verified Cues to Deception: A Meta-Analysis of Behavioral Indicators
Massive meta-analysis of 116 studies; found most behavioral cues weak or unreliable; supports instrument-based approaches over human judgment alone.

Accuracy Comparison

When comparing accuracy data, the polygraph holds a significant advantage. EyeDetect claims 86–88% accuracy for screening [9]Verified EyeDetect — Best Lie Detector (Official Product Page)
Confirms 86–88% accuracy claim; tests take 15–30 minutes; proctors trained in a day; up to 3 stations managed simultaneously, translating to an error rate of 12–14%. Single-issue polygraph testing achieves accuracy of 89% according to the APA's meta-analysis (confidence interval 83–95%) [30]Verified Cues to Deception: A Meta-Analysis of Behavioral Indicators
Massive meta-analysis of 116 studies; found most behavioral cues weak or unreliable; supports instrument-based approaches over human judgment alone, and the American Polygraph Association cites research studies reporting accuracy rates ranging from 80% to 98% [31]Verified Meta-Analysis of CQT: First Meta-Analysis of the Comparison Question Test
The first meta-analysis of the CQT; found significant moderator effects for subject type, incentives, and decision policy; pioneering quantitative polygraph research. This means EyeDetect has roughly double the error rate of properly administered polygraph examinations.

The comprehensive meta-analysis by Honts et al. (2021) found that motivation level showed a positive linear relationship with CQT accuracy [29]Verified A Meta-Analysis of the Comparison Question Polygraph Test: Moderator Effects
Reviewed CQT literature and found overall accuracy estimates above 85% across studies; identified important moderator variables, which is particularly significant for real-world applications where stakes are high. Earlier meta-analyses including Kircher, Horowitz, and Raskin (1988) and Vrij, Mann, and Fisher (2008) consistently found overall CQT accuracy estimates above 85% across studies [32]Verified Meta-Analysis of CIT Laboratory Studies
Found weighted average false positive rate of only 4.1% and false negative rate of 19.4% for the Concealed Information Test[33]Verified A Comprehensive Meta-Analysis of CIT Research Using MacLaren's Data
Laboratory meta-analysis on the CIT/GKT confirming strong discrimination between informed and uninformed examinees.

The Human Examiner Advantage

Perhaps the most fundamental difference between EyeDetect and polygraph testing is the role of the human examiner. EyeDetect is fully automated — the examinee sits in front of a computer and answers true/false questions with no human interaction [9]Verified EyeDetect — Best Lie Detector (Official Product Page)
Confirms 86–88% accuracy claim; tests take 15–30 minutes; proctors trained in a day; up to 3 stations managed simultaneously. Polygraph testing involves an extensive pre-test interview that can last an hour or longer, during which a trained examiner collects detailed background information, discusses test issues, creates agreed-upon definitions for ambiguous terms, and builds rapport [21]Verified Policing Project Five-Minute Biometrics Primers: Lie Detection II — 21st Century Tests
Confirms only in-house team conducted peer-reviewed studies; no independent academic studies found; notes algorithmic transparency concerns.

This human interaction serves multiple purposes. Admissions and important disclosures are frequently made during the pre-test interview phase — information that would never surface in an automated test [21]Verified Policing Project Five-Minute Biometrics Primers: Lie Detection II — 21st Century Tests
Confirms only in-house team conducted peer-reviewed studies; no independent academic studies found; notes algorithmic transparency concerns. The examiner can adapt questions based on the individual's responses, detect signs of confusion or non-cooperation, and apply professional judgment informed by years of training. As research by Bella DePaulo (2001) demonstrated in a massive meta-analysis of 116 studies, most behavioral cues to deception are weak or unreliable — supporting the need for instrument-based approaches guided by expert human judgment rather than automation alone [34]Verified Eye-Movement Deception Detection Studied as Part of Pentagon's Vetting Modernization
Confirms DCSA/NCCA research on EyeDetect since FY 2022; Oak Ridge evaluation; Texas A&M field study; mixed results; FY 2026 transition to broader analytic infrastructure. Learn more about how examiners calibrate for individual physiology in our guide on what an acquaintance test is in polygraph.

The ultimate goal of a polygraph examination is "information gain" — extracting relevant, actionable information through the testing process [21]Verified Policing Project Five-Minute Biometrics Primers: Lie Detection II — 21st Century Tests
Confirms only in-house team conducted peer-reviewed studies; no independent academic studies found; notes algorithmic transparency concerns. This requires skilled human interaction that automated systems simply cannot replicate. For guidance on what happens after a polygraph exam, the post-test phase often produces the most valuable investigative information.

Physiological Breadth

EyeDetect relies primarily on cognitive load as measured through ocular responses [21]Verified Policing Project Five-Minute Biometrics Primers: Lie Detection II — 21st Century Tests
Confirms only in-house team conducted peer-reviewed studies; no independent academic studies found; notes algorithmic transparency concerns. Polygraph testing measures multiple independent physiological channels simultaneously — cardiovascular activity, respiration, and electrodermal responses — providing a broader window into the autonomic nervous system's response to deception [20]Verified FAQs for EyeDetect — Accuracy Claims Detail
Confirms 88% accuracy for screening tests; 88% and 91% accuracy in hybrid MCT and DLC studies; 95.5% in APA seminar demonstration. Understanding the polygraph cardiograph channel illustrates the depth of physiological data that traditional polygraph captures.

EyeDetect measures one category of effector that polygraph already captures (cognitive effort), while polygraph additionally measures fear of detection and cognitive dissonance through its multiple physiological channels [21]Verified Policing Project Five-Minute Biometrics Primers: Lie Detection II — 21st Century Tests
Confirms only in-house team conducted peer-reviewed studies; no independent academic studies found; notes algorithmic transparency concerns. This broader measurement base is one reason polygraph consistently achieves higher accuracy rates in validated research.

The University of Utah Connection

Understanding EyeDetect's scientific foundation requires understanding its institutional origins. The technology was developed by researchers at the University of Utah, where polygraph research has been conducted for decades [6]Verified History of Converus — Official Timeline
Confirms Kircher and Raskin developed computerized polygraph in 1991; science team formed 2003. The late Dr. John C. Kircher served as Chief Scientist at Converus and had published more than 90 scientific publications [8]Verified Renowned Polygraph Expert Dr. John Kircher Joins Converus Staff
Confirms Kircher published 90+ scientific publications; served as consultant to DOD, Secret Service, DHS, NSF, National Research Council. Dr. David C. Raskin, who joined the science team in 2009, authored over 150 publications and served on faculties at UCLA, Michigan State, and the University of Utah [3]Verified FAQs for EyeDetect — Converus Official
Confirms Kircher and Raskin co-invented computerized polygraph in 1991; Raskin joined science team in 2009; technology history and technical details.

Kircher and Raskin's foundational 1988 meta-analysis of mock crime studies of the CQT represented pioneering work in quantitative polygraph research [35]Verified Bush School Team Awarded Project to Study Effectiveness of Lie-Detection System That Tracks Eye Behavior
Confirms two-year DCSA-contracted field evaluation of EyeDetect along Texas-Mexico border; Texas A&M Bush School lead researchers. Their first meta-analysis of the Comparison Question Test found significant moderator effects for subject type, incentives, and decision policy [35]Verified Bush School Team Awarded Project to Study Effectiveness of Lie-Detection System That Tracks Eye Behavior
Confirms two-year DCSA-contracted field evaluation of EyeDetect along Texas-Mexico border; Texas A&M Bush School lead researchers. This earlier polygraph research by the same scientists who developed EyeDetect demonstrates the depth of the polygraph evidence base by comparison.

Evidence Base: 12 Peer-Reviewed Studies vs. Thousands

As of June 2024, Converus reported 12 peer-reviewed publications validating the ocular-motor deception test method [10]Verified 12th Peer-Reviewed Publication Further Validates Eye Behavior as an Accurate Deception Detection Method
Confirms 12 peer-reviewed publications as of June 2024; eye-tracking camera takes approximately 60 measurements per second; study conducted by Converus Science Team members. The 12th publication, "Using Pupillometry in the Detection of Deception," was conducted by four members of the Converus Science Team and published in the book "Modern Pupillometry" [10]Verified 12th Peer-Reviewed Publication Further Validates Eye Behavior as an Accurate Deception Detection Method
Confirms 12 peer-reviewed publications as of June 2024; eye-tracking camera takes approximately 60 measurements per second; study conducted by Converus Science Team members. This is a significantly smaller evidence base than the polygraph, which has been the subject of thousands of studies over seven decades.

The APA's 2011 meta-analytic survey alone analyzed 38 studies involving 32 different samples, 45 experiments, 295 scorers, and 11,737 scored results from 3,723 examinations [30]Verified Cues to Deception: A Meta-Analysis of Behavioral Indicators
Massive meta-analysis of 116 studies; found most behavioral cues weak or unreliable; supports instrument-based approaches over human judgment alone. The 2003 National Academy of Sciences report reviewed approximately 80 research studies published since 1980 on polygraph accuracy [31]Verified Meta-Analysis of CQT: First Meta-Analysis of the Comparison Question Test
The first meta-analysis of the CQT; found significant moderator effects for subject type, incentives, and decision policy; pioneering quantitative polygraph research. The Honts et al. (2021) comprehensive CQT meta-analysis captured 138 datasets [29]Verified A Meta-Analysis of the Comparison Question Polygraph Test: Moderator Effects
Reviewed CQT literature and found overall accuracy estimates above 85% across studies; identified important moderator variables — more than 11 times the entire EyeDetect publication count.

Research on the Concealed Information Test provides additional depth to the polygraph evidence base. Eitan Elaad's (1998) meta-analysis found a weighted average false positive rate of only 4.1% for the CIT [36]Verified Daubert v. Merrell Dow Pharmaceuticals, Inc. — Expert Testimony Standard
Confirms Daubert standard established 1993; factors include testability, peer review, error rate, general acceptance; used in federal court and majority of states, while MacLaren's (2001) comprehensive meta-analysis confirmed strong discrimination between informed and uninformed examinees [37]Verified US District Court Allows EyeDetect Lie Detector Test Results As Evidence for First Time
Confirms first Daubert motion for EyeDetect in New Mexico district court April 2018; accused passed with credible score; Ohio and Texas cases also noted. The breadth of Eitan Elaad's CIT research has been foundational to advancing polygraph science.

Pentagon Research: Promising but Inconclusive

The Defense Counterintelligence and Security Agency (DCSA), through its National Center for Credibility Assessment (NCCA), has been conducting research on ocular-motor deception detection since approximately FY 2022 [38]Verified Law Prof Sees Uphill Climb for Admission of EyeDetect Technology in Court
Confirms Taos NM case details; five of 12 jurors voted for acquittal; mistrial declared. Early experiments used EyeDetect and were evaluated at Oak Ridge National Laboratory [38]Verified Law Prof Sees Uphill Climb for Admission of EyeDetect Technology in Court
Confirms Taos NM case details; five of 12 jurors voted for acquittal; mistrial declared. A two-year field evaluation was launched in partnership with Texas A&M University's Bush School, testing EyeDetect along the Texas-Mexico border [39]Verified EyeDetect Marketer Converus Spends Hundreds of Thousands of Dollars Lobbying Federal Agencies
Confirms DoD Directive 5210.91 prohibits credibility assessment solutions other than polygraph; Converus paid $60,000/year for lobbying 2016–2022.

However, budget documents indicate that testing "on personnel undergoing specialized training" yielded mixed results [38]Verified Law Prof Sees Uphill Climb for Admission of EyeDetect Technology in Court
Confirms Taos NM case details; five of 12 jurors voted for acquittal; mistrial declared. As of FY 2026, DCSA budget documents still describe "ocular-motor deception detection capabilities" but list no new funding, indicating the research has moved into a transition phase [38]Verified Law Prof Sees Uphill Climb for Admission of EyeDetect Technology in Court
Confirms Taos NM case details; five of 12 jurors voted for acquittal; mistrial declared. The research is being absorbed into broader analytic infrastructure rather than advancing as a standalone credibility assessment tool [38]Verified Law Prof Sees Uphill Climb for Admission of EyeDetect Technology in Court
Confirms Taos NM case details; five of 12 jurors voted for acquittal; mistrial declared. A DCSA spokesperson stated the agency "uses rigorous, scientific practices to evaluate new technologies" and expects to provide results to inform future potential use [38]Verified Law Prof Sees Uphill Climb for Admission of EyeDetect Technology in Court
Confirms Taos NM case details; five of 12 jurors voted for acquittal; mistrial declared.

The Proprietary Algorithm Problem

EyeDetect uses a proprietary logistic regression algorithm whose specific weights, decision thresholds, and training data are not publicly available [18]Verified Employee Polygraph Protection Act — Full Text
Confirms EPPA defines 'lie detector' broadly to include polygraph and any similar device used to render diagnostic opinion regarding honesty. The Policing Project noted this as a significant concern — Converus is not willing to share the inner workings of the system during independent audits [25]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
The largest meta-analysis ever conducted on the CQT, analyzing 138 datasets; found meta-analytic effect size of 0.69; classification accuracy well over 80%. A 2025 critical review described the system as functioning "essentially as a 'black box'" whose "proprietary algorithms, feature weightings, and decision thresholds are not accessible to independent researchers, legal professionals, or regulatory bodies" [27]Verified The Polygraph and Lie Detection — National Academy of Sciences Report
Confirms 2003 NAS report; APA cites research with accuracy rates ranging from 80% to 98%; NAS found polygraph has greater than chance accuracy.

This lack of transparency creates multiple problems. Results cannot be independently verified or meaningfully challenged in legal proceedings. There is no mechanism for independent oversight, external error calibration, or mandated protocol standardization across different implementations [27]Verified The Polygraph and Lie Detection — National Academy of Sciences Report
Confirms 2003 NAS report; APA cites research with accuracy rates ranging from 80% to 98%; NAS found polygraph has greater than chance accuracy. Reports of modifications to scoring models and selective subsampling for validation highlight the absence of a fixed, publicly validated standard [27]Verified The Polygraph and Lie Detection — National Academy of Sciences Report
Confirms 2003 NAS report; APA cites research with accuracy rates ranging from 80% to 98%; NAS found polygraph has greater than chance accuracy.

By contrast, validated polygraph scoring systems such as numerical scoring methods have transparent decision rules that examiners, quality assurance reviewers, and legal professionals can independently evaluate. This transparency is fundamental to the admissibility of polygraph evidence and its acceptance by the professional community. Understanding Bayesian vs. frequentist statistics in polygraph testing illustrates the rigorous statistical frameworks that underpin modern polygraph analysis.

The Daubert Standard and EyeDetect

In the United States, the admissibility of scientific evidence is primarily governed by the Daubert standard, established in Daubert v. Merrell Dow Pharmaceuticals, Inc. (1993) [40]Verified Tech Co. Says its Advanced Lie Detector Will Save U.S. Gov't $29B Annually
Confirms 1,000+ customers; 100+ U.S. law enforcement agencies; 50,000+ State Department tests; EyeDetect six times faster than polygraph. Under Daubert, judges evaluate whether scientific evidence is based on reliable principles and methods by considering factors including testability, peer review, known error rates, and general acceptance in the relevant scientific community [40]Verified Tech Co. Says its Advanced Lie Detector Will Save U.S. Gov't $29B Annually
Confirms 1,000+ customers; 100+ U.S. law enforcement agencies; 50,000+ State Department tests; EyeDetect six times faster than polygraph.

EyeDetect was admitted as evidence for the first time in April 2018, when a New Mexico judicial district court judge granted a defendant's Daubert motion in a child sexual abuse case [41]Verified What Are Polygraphs and Why Is the Trump Administration Using Them?
Confirms DHS and FBI declined to comment on EyeDetect usage; CIA and DoD did not respond to requests. The accused had passed the EyeDetect test with a "credible" score [41]Verified What Are Polygraphs and Why Is the Trump Administration Using Them?
Confirms DHS and FBI declined to comment on EyeDetect usage; CIA and DoD did not respond to requests. However, the trial ended in a mistrial when five of 12 jurors voted for acquittal [42]Verified Basic Polygraph Examiner Course — PEAK Credibility Assessment Training Center
Confirms APA-accredited basic course consists of minimum 400 hours of instruction. This single admission does not establish broader legal precedent.

Additionally, a judge in Ohio considered EyeDetect results among other evidence when dropping charges in a weapons possession case, and a Texas judge ordered EyeDetect testing in a drug distribution case [41]Verified What Are Polygraphs and Why Is the Trump Administration Using Them?
Confirms DHS and FBI declined to comment on EyeDetect usage; CIA and DoD did not respond to requests. These instances represent isolated judicial decisions rather than established admissibility precedent. Exploring the limits of admission in polygraph exams provides valuable context for understanding how evidence standards apply across credibility assessment technologies.

Federal Regulatory Barriers

Current federal regulations present significant barriers to EyeDetect's use in government settings. DoD Directive 5210.91 and ODNI Security Agent Directive 2 currently prohibit the use of any credibility assessment solution other than polygraph [43]Verified Polygraph Course Curriculum — The Polygraph Institute
Confirms 400-hour in-residence curriculum; 10-week cycle; weekly written exams; practical exams; 40 hours advanced training required within 12 months. Converus has actively lobbied to change these regulations, paying $60,000 per year from 2016 to 2022 to the Washington, DC lobbying firm The Consilio Group [43]Verified Polygraph Course Curriculum — The Polygraph Institute
Confirms 400-hour in-residence curriculum; 10-week cycle; weekly written exams; practical exams; 40 hours advanced training required within 12 months. The company has targeted the CIA, DHS, Department of Defense, and other agencies [43]Verified Polygraph Course Curriculum — The Polygraph Institute
Confirms 400-hour in-residence curriculum; 10-week cycle; weekly written exams; practical exams; 40 hours advanced training required within 12 months.

Despite these lobbying efforts, the U.S. State Department is the only confirmed federal agency actively using EyeDetect, having facilitated over 50,000 tests across various countries since 2018 [44]Verified APA Policy for Continuing Education Hours
Confirms practicing examiners must complete minimum 30 continuing education hours every two years. The DHS and FBI have declined to comment on whether they use EyeDetect [45]Verified APA Accredited Polygraph Training Programs
Confirms only 11 APA-accredited training institutions in the United States that teach the 400-hour Basic Polygraph Examiner Course.

EyeDetect Proctor Training vs. Polygraph Examiner Training

According to Converus, EyeDetect proctors can be trained in as little as a day and can manage up to three testing stations simultaneously [9]Verified EyeDetect — Best Lie Detector (Official Product Page)
Confirms 86–88% accuracy claim; tests take 15–30 minutes; proctors trained in a day; up to 3 stations managed simultaneously[10]Verified 12th Peer-Reviewed Publication Further Validates Eye Behavior as an Accurate Deception Detection Method
Confirms 12 peer-reviewed publications as of June 2024; eye-tracking camera takes approximately 60 measurements per second; study conducted by Converus Science Team members. This minimal training requirement reflects the automated nature of the system — the proctor does not score the test, conduct interviews, or make diagnostic judgments.

By comparison, APA-accredited polygraph schools require a minimum of 400 hours of formal instruction [46]Verified Converus EyeDetect Case Studies — Global Adoption
Confirms EyeDetect deployment across multiple countries; law enforcement case studies including Bernalillo County, Idaho State Police, Wyoming Highway Patrol, typically completed over 10 to 17 weeks of intensive, in-residence training [47]Verified Letter to GAO from U.S. Senators Markey, Thompson, and Booker
Confirms State Department contract with Converus; DCSA funding border research; Congressional concern about automated deception detection technologies. The curriculum covers physiology, psychology, instrumentation, validated scoring methods, countermeasure detection, and interview techniques [47]Verified Letter to GAO from U.S. Senators Markey, Thompson, and Booker
Confirms State Department contract with Converus; DCSA funding border research; Congressional concern about automated deception detection technologies. Following graduation, students must submit supervised field examinations and complete 40 hours of advanced training within 12 months [47]Verified Letter to GAO from U.S. Senators Markey, Thompson, and Booker
Confirms State Department contract with Converus; DCSA funding border research; Congressional concern about automated deception detection technologies. Practicing examiners must then complete a minimum of 30 continuing education hours every two years [48]Verified CONNECT'22 Conference — Jamaica Defence Force EyeDetect Adoption
Confirms Jamaica Defence Force conducted 7,550 EyeDetect exams since 2018; countermeasure attempts easily identified. There are only 11 APA-accredited training institutions in the United States [49]Verified Employee Polygraph Protection Act of 1988 — 29 CFR Part 801
Confirms EPPA prohibits most private employers from using any lie detector tests; government employers exempt; polygraph permitted under limited circumstances.

This vast training differential reflects a fundamental difference in professional standards. A polygraph examiner is a trained forensic professional who applies clinical judgment, adapts to individual examinees, and can detect countermeasure attempts and unusual physiology. An EyeDetect proctor operates equipment. For those considering a career in polygraph examination, our Wyoming polygraph training guide provides information on requirements and career paths. The importance of choosing a qualified polygraph examiner cannot be overstated.

Global Adoption and Customer Base

As of April 2025, Converus claims EyeDetect is used by more than 1,000 customers worldwide, including over 100 U.S. law enforcement agencies [44]Verified APA Policy for Continuing Education Hours
Confirms practicing examiners must complete minimum 30 continuing education hours every two years. The company reports operations in more than 60 countries, with software available in 50+ languages [44]Verified APA Policy for Continuing Education Hours
Confirms practicing examiners must complete minimum 30 continuing education hours every two years[50]Verified FAQs for EyeDetect — EPPA Coverage
Confirms EPPA has been applied broadly to any lie detectors including EyeDetect; federal/state/local government employees exempt. The U.S. State Department's Bureau of International Narcotics Control and Law Enforcement Affairs currently has a contract with Converus for EyeDetect equipment [51]Verified The Eyes Have It: How This Lehi Company Has Taken Lie Detecting to a New Level
Confirms EyeDetect station costs approximately $4,000; scoring fee $50–$150 per test; 86% accuracy per Converus.

Notable use cases include the Jamaica Defence Force, which has conducted over 7,550 EyeDetect exams since implementing the technology in 2018 [52]Verified Maryland Polygraph & EyeDetect Tests — Consumer Pricing
Confirms consumer-facing EyeDetect pricing of $300–$800 depending on test type. Various law enforcement agencies across Utah, Texas, Ohio, Louisiana, Florida, and other states have used EyeDetect for pre-employment screening [25]Verified A Comprehensive Meta-Analysis of the Comparison Question Polygraph Test
The largest meta-analysis ever conducted on the CQT, analyzing 138 datasets; found meta-analytic effect size of 0.69; classification accuracy well over 80%. The Connecticut EyeDetect program represents an interesting case study of a state implementing EyeDetect for post-conviction sex offender testing. For more on adoption trends in developing markets, see our guide on EyeDetect for African law enforcement.

EPPA Legal Considerations

The Employee Polygraph Protection Act of 1988 (EPPA) broadly prohibits most private employers from using any "lie detector" tests [53]Verified Concerns About AI-Based Deception Detection — Senate Letter
Confirms Congressional concerns about automated deception detection in vulnerable populations. Under EPPA, the term "lie detector" includes "a polygraph, deceptograph, voice stress analyzer, psychological stress evaluator, or any other similar device" used for rendering a diagnostic opinion regarding honesty [53]Verified Concerns About AI-Based Deception Detection — Senate Letter
Confirms Congressional concerns about automated deception detection in vulnerable populations. Although EPPA specifically mentions polygraph by name, the broader "lie detector" prohibition has been applied in practice to EyeDetect and similar technologies [54]Verified FAQs for EyeDetect — Countermeasures
Confirms Dr. Charles Honts assessment of countermeasure resistance; rapid questioning pace of 5–7 seconds; unsuccessful countermeasure attempts listed.

Government employers — federal, state, and local — are exempt from EPPA [53]Verified Concerns About AI-Based Deception Detection — Senate Letter
Confirms Congressional concerns about automated deception detection in vulnerable populations. This exemption explains why EyeDetect adoption has been concentrated in government and law enforcement contexts rather than the private sector.

EyeDetect Pricing Structure

EyeDetect's cost structure includes the hardware station at approximately $4,000 and a per-test scoring fee ranging from $50 to $150 [55]Verified Decoding Deception with the P300: A Meta-Analysis of the Concealed Information Test
Found large mean effect size (d*) of 1.59 for P300-based CIT across 54 experimental studies; complex trial protocols maximized effect sizes. Consumer-facing EyeDetect testing through third-party providers ranges from approximately $300 to $800 depending on the test type [56]Verified A Four-Domain Model for Detecting Deception: An Alternative Paradigm for Interviewing
Foundational research on behavioral observation in deception detection emphasizing comfort/discomfort as key veracity indicators. Converus markets the system as approximately 1/30 the per-test cost of traditional polygraph [44]Verified APA Policy for Continuing Education Hours
Confirms practicing examiners must complete minimum 30 continuing education hours every two years.

However, a price comparison alone misrepresents the value proposition. A polygraph examination includes a comprehensive pre-test interview, multi-channel physiological monitoring, expert interpretation, a post-test interview phase that frequently yields important admissions and disclosures, and a detailed report prepared by a trained forensic professional. EyeDetect provides an automated computer score. Evaluating credibility assessment purely on cost per test is like comparing the price of a screening questionnaire to a comprehensive medical examination — the cheaper option is not necessarily the better value.

Algorithmic Bias and Accountability

The deployment of automated deception detection technology in high-stakes contexts raises serious ethical questions. A June 2025 letter from U.S. Senators Edward Markey, Bennie Thompson, and Cory Booker to the Government Accountability Office expressed concern about technologies like EyeDetect being used on people in vulnerable situations, including immigration screening and law enforcement hiring [51]Verified The Eyes Have It: How This Lehi Company Has Taken Lie Detecting to a New Level
Confirms EyeDetect station costs approximately $4,000; scoring fee $50–$150 per test; 86% accuracy per Converus.

The Center for AI and Digital Policy has called for the prohibition of emotion analysis and biometric categorization technologies, framing their use as a human rights concern [57]Verified Evaluating Polygraphy from a Psychophysiological Perspective: A Specific-Effects Analysis
Evaluated whether polygraph responses represent specific deception-related effects or general arousal; contributed to theoretical understanding. Because EyeDetect's scoring algorithm is proprietary and not subject to independent audit, there is no way to assess whether the system contains embedded biases related to race, ethnicity, language proficiency, or cultural background — despite the company's claims of cross-cultural validity [27]Verified The Polygraph and Lie Detection — National Academy of Sciences Report
Confirms 2003 NAS report; APA cites research with accuracy rates ranging from 80% to 98%; NAS found polygraph has greater than chance accuracy.

Traditional polygraph examinations, while not immune to examiner bias, have transparent scoring methods, established quality assurance protocols through the APA, and are conducted by trained professionals bound by ethical standards and continuing education requirements [48]Verified CONNECT'22 Conference — Jamaica Defence Force EyeDetect Adoption
Confirms Jamaica Defence Force conducted 7,550 EyeDetect exams since 2018; countermeasure attempts easily identified. When concerns arise, polygraph charts and scoring can be independently reviewed — a level of accountability that automated black-box systems do not provide.

Resistance to Countermeasures

Converus argues that EyeDetect's rapid question pace — presenting true/false statements every 5–7 seconds — leaves insufficient time for traditional polygraph countermeasures to be effective [58]Verified EyeDetect Frequently Asked Questions — Combined Accuracy Claims
Confirms when EyeDetect and polygraph used together, combined accuracy of approximately 96.6% claimed; false negative probability of 3.4%. Dr. Charles Honts of Boise State University, a leading countermeasures researcher, has stated that he does not currently see "any immediate active countermeasure threats to EyeDetect" because the direct central nervous system channel leaves little room for deliberate manipulation [58]Verified EyeDetect Frequently Asked Questions — Combined Accuracy Claims
Confirms when EyeDetect and polygraph used together, combined accuracy of approximately 96.6% claimed; false negative probability of 3.4%.

Converus lists several unsuccessful countermeasure attempts: closing eyes or squinting during questions (observable by the proctor), responding "False" to a majority of questions or randomly (detected by the algorithm), and using pupil-dilating eye drops such as tropicamide, which the software flags as abnormal pupil behavior [58]Verified EyeDetect Frequently Asked Questions — Combined Accuracy Claims
Confirms when EyeDetect and polygraph used together, combined accuracy of approximately 96.6% claimed; false negative probability of 3.4%.

However, the claim that EyeDetect is countermeasure-resistant remains largely untested by independent researchers. Without adversarial testing by scientists without financial ties to the company, this claim should be viewed cautiously. Polygraph countermeasure detection, by contrast, is a well-studied field with extensive research and specialized training for examiners. Our guide on whether you can deceive polygraph testing provides a thorough analysis of this topic.

The Evolving Credibility Assessment Landscape

The credibility assessment field continues to evolve with emerging technologies, but the polygraph remains firmly established as the gold standard. The Pentagon's DCSA is exploring ocular-motor deception detection as part of its broader Trusted Workforce 2.0 initiative, but even in this context, the technology is envisioned as a data source feeding into larger analytic systems rather than as a standalone replacement for polygraph [38]Verified Law Prof Sees Uphill Climb for Admission of EyeDetect Technology in Court
Confirms Taos NM case details; five of 12 jurors voted for acquittal; mistrial declared.

Emerging brain-based approaches show promise. A 2025 meta-analysis of P300-based Concealed Information Testing found a large mean effect size (d*) of 1.59, demonstrating strong potential for deception detection through ERP analysis. Other researchers, including Joe Navarro's four-domain model, have contributed alternative paradigms that emphasize behavioral observation. The foundational question raised by John Furedy's psychophysiological perspective — whether physiological responses represent specific deception effects or general arousal — remains central to evaluating any credibility assessment technology.

For individuals seeking credibility assessment today, the proven polygraph examination remains the most thoroughly validated, legally recognized, and professionally standardized option available. Whether for personal matters, legal proceedings, or organizational screening, working with a trained, professional polygraph examiner provides the most comprehensive and reliable assessment. Review our dos and don'ts of taking a lie detector test before your examination.

Can EyeDetect Supplement Polygraph?

While Converus markets EyeDetect as both an alternative and a complement to polygraph, the most defensible position is that it may serve as a preliminary screening tool in certain high-volume contexts where full polygraph examinations are not feasible. When used in combination with polygraph, Converus claims the joint probability of a false negative (a deceptive person passing both tests) drops to approximately 3.4%, yielding 96.6% accuracy.

However, this mathematical calculation assumes that the two tests are truly independent measures — a claim that requires further validation given that both systems rely in part on cognitive load as an effector [21]Verified Policing Project Five-Minute Biometrics Primers: Lie Detection II — 21st Century Tests
Confirms only in-house team conducted peer-reviewed studies; no independent academic studies found; notes algorithmic transparency concerns. Additionally, the practical value of any screening tool is only as good as the quality assurance and professional oversight surrounding its use. EyeDetect's minimal training requirements and automated scoring remove the professional safeguards that make polygraph a reliable forensic tool.

For organizations considering credibility assessment options, the comprehensive polygraph examination — administered by an examiner trained to APA standards following established standards and best practices — remains the clear choice for high-stakes applications. Explore our options for polygraph testing in Connecticut and other locations across the United States, or learn about polygraph training schools in Africa for international readers.