EyeDetect for Theft Investigations: Employer’s Guide

What Is EyeDetect and How Does It Work?

EyeDetect is an ocular-motor deception test developed by Converus, Inc. and based on research conducted by scientists at the University of Utah [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. The technology was conceived in 2002 when Dr. John C. Kircher and Dr. Doug Hacker explored whether changes in cognitive load would affect the eyes in measurable ways [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. They formed the Converus Science Team in 2003, joined by cognitive scientists Anne Cook and Dan Woltz, with Dr. David C. Raskin joining in 2009 [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. The technology was commercially released in April 2014 [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position.

Unlike a traditional polygraph, which measures physiological responses like blood pressure, respiration, and galvanic skin response through sensors attached to the body, EyeDetect tracks involuntary changes in the eyes using an infrared camera built into a specialized computer station [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. The underlying science is rooted in cognitive load theory — when a person lies, their brain must work harder than when telling the truth [2]Verified 12th Peer-Reviewed Publication Further Validates Eye Behavior as an Accurate Deception Detection Method
Confirms 86-88% accuracy claim and details of published validation research involving the Converus Science Team. This increased cognitive demand produces measurable changes in pupil dilation, eye fixation patterns, reading speed, blink rate, and other oculomotor behaviors [2]Verified 12th Peer-Reviewed Publication Further Validates Eye Behavior as an Accurate Deception Detection Method
Confirms 86-88% accuracy claim and details of published validation research involving the Converus Science Team.

During an EyeDetect test, the subject sits in front of a computer screen and reads a series of true-or-false statements. The infrared eye-tracking camera takes approximately 60 measurements per second of each eye [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Proprietary algorithms analyze the data to produce a Converus Credibility Score ranging from 1 (most likely deceptive) to 99 (most likely credible) [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Understanding what to expect during an EyeDetect test can help both employers and employees feel more prepared for the process.

Two Types of EyeDetect Tests Relevant to Theft

Converus offers two primary test formats relevant to theft investigations:

Multi-issue Comparison Test (MCT) / Screening Test: This is a broader screening test typically used for pre-employment or periodic screening. The MCT has a reported mean accuracy of 88% based on initial lab studies conducted by the Converus Science Team [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Subjects answer true-or-false questions covering multiple topics, and the algorithm compares eye behavior patterns to determine credibility. MCT tests are useful when an employer wants to screen a group of employees who had access to stolen property.

Directed Lie Comparison (DLC) / Diagnostic Test: This is a more focused, specific-issue test designed for targeted investigations. According to field data gathered by Kircher and Raskin in December 2018, the DLC test has a mean accuracy of 90% [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. It zeros in on a particular incident, such as a specific theft event, and is the more appropriate choice when there is a defined theft incident with identified suspects.

How EyeDetect Differs from Polygraph in Practice

The practical differences between EyeDetect and polygraph are significant for workplace investigations. A traditional polygraph requires a professional examiner to conduct a pre-test interview, attach physiological sensors, ask questions in a controlled format, and then interpret the resulting charts. The APA Standards of Practice specify that examinations shall be scheduled for not less than 90 minutes [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit, and examiners shall not conduct more than five examinations per day [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit.

In contrast, an EyeDetect test is largely automated. After initial setup and instructions, the subject interacts with the computer, and the algorithm generates results without human interpretation [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Converus states that one test proctor using three stations can administer up to 30-40 tests daily [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. This is a critical distinction — polygraph instruments cannot independently detect deception; only trained examiners who study physiological responses can make credibility determinations [4]Verified Investigations: Polygraph Use
Confirms that polygraph instruments cannot independently detect deception; only trained examiners can make credibility determinations.

A polygraph examination includes the human element of the pre-test interview, during which an experienced examiner can observe behavioral cues, ask probing questions, and sometimes elicit admissions before the actual test even begins. EyeDetect, by design, removes this human element in favor of standardization and speed. For a detailed side-by-side analysis, see our comparison of EyeDetect vs. polygraph accuracy.

The Employee Polygraph Protection Act Overview

The Employee Polygraph Protection Act of 1988 (EPPA) is the cornerstone federal law governing the use of lie detection in the workplace [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. Signed by President Reagan on June 27, 1988, EPPA broadly prohibits most private employers from requiring, requesting, suggesting, or causing employees or job applicants to take a "lie detector test" [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. The law defines "lie detector" to include polygraphs, deceptographs, voice stress analyzers, psychological stress evaluators, and any similar device used to render a diagnostic opinion about the honesty or dishonesty of an individual [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements.

The original statute set a maximum civil penalty of $10,000 per violation [6]Verified Department of Labor Federal Civil Penalties Inflation Adjustment Act Catch-Up Adjustments
Confirms EPPA penalty was adjusted from $10,000 to $19,787 per violation under inflation adjustment requirements. Under the Federal Civil Penalties Inflation Adjustment Act, this amount has been adjusted for inflation and was increased to $19,787 per violation as of 2016 [6]Verified Department of Labor Federal Civil Penalties Inflation Adjustment Act Catch-Up Adjustments
Confirms EPPA penalty was adjusted from $10,000 to $19,787 per violation under inflation adjustment requirements. Employees can also sue for lost wages, benefits, and attorney fees [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. Our comprehensive EPPA guide covers all the details employers need to know.

Importantly, EPPA regulations clarify that "polygraph tests, but no other types of lie detector tests, are permitted under limited circumstances subject to certain restrictions" [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. This distinction is crucial for understanding EyeDetect's legal position.

Does EPPA Apply to EyeDetect?

This is one of the most important legal questions for employers considering EyeDetect. Converus itself acknowledges on its FAQ page that "although the EPPA law specifically mentions polygraph, the law has been applied in practice more broadly to any lie detectors" [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Converus further states that "any lie detector is prohibited" for private-sector employment use, with exceptions for government employees and law enforcement [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position.

Converus developed EyeDetect+ specifically to address this legal gap. Released in 2020, EyeDetect+ adds physiological sensors (cardiovascular, respiration, and electrodermal activity) via a "Physio Tracker" device, allowing it to meet the federal government's functional definition of a polygraph [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Converus states that EyeDetect+ "should qualify to be used under the three exceptions to the Employee Polygraph Protection Act (EPPA) where polygraph — and no other lie detectors — may be used" [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position.

For the standard EyeDetect (without Physio Tracker), the EPPA applicability remains legally uncertain. As of 2026, no published federal court decision has specifically addressed whether the standard EyeDetect system is covered by EPPA. The most prudent approach is to treat EyeDetect as if it were covered by EPPA and follow all the same compliance requirements. Understanding employee rights relating to lie detector tests is critical for employers navigating this area.

State Laws Add Another Layer

Beyond EPPA, many states have their own laws restricting or prohibiting workplace lie detection. Some state statutes are even broader than EPPA and may explicitly cover technologies beyond traditional polygraph. Converus identified 17 states plus the District of Columbia where statutes defined "lie detector" in ways that the original EyeDetect did not meet — including Alabama, Arkansas, Delaware, Illinois, Kentucky, Maine, Michigan, Mississippi, Missouri, Nevada, North Dakota, Oklahoma, South Carolina, Tennessee, Vermont, Virginia, and West Virginia [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position.

States with particularly restrictive laws include Massachusetts, which broadly prohibits lie detection in employment, and California, which has strong privacy protections. Employers must check both federal EPPA requirements and their specific state's employment laws before using EyeDetect in any workplace context. Our state-specific guides like the Massachusetts polygraph training guide and Nevada NRS 613.440 employer guide provide details on individual state regulations.

What the EPPA Ongoing Investigation Exception Allows

The most relevant EPPA provision for theft investigations is the "ongoing investigation exception" found at 29 U.S.C. § 2006(d) [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. This exception allows certain private employers to request a polygraph test of an employee as part of an ongoing investigation involving economic loss or injury to the employer's business [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. The statutory text specifies that the test must be "administered in connection with an ongoing investigation involving economic loss or injury to the employer's business, such as theft, embezzlement, misappropriation, or an act of unlawful industrial espionage or sabotage" [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements.

To qualify for this exception, an employer must simultaneously meet all four requirements: (1) the test is connected to an ongoing investigation involving economic loss; (2) the employee had access to the property that is the subject of the investigation; (3) the employer has a reasonable suspicion that the employee was involved; and (4) the employer provides a written statement to the employee at least 48 hours before the test describing the specific incident, the economic loss, and the basis for suspicion [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements.

Learn more about whether employers can polygraph employees suspected of theft and the detailed compliance requirements.

Applying the Exception to EyeDetect

Here is where the legal complexity deepens for employers considering EyeDetect. The ongoing investigation exception, by its statutory text, applies specifically to "polygraph examinations" [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. If a court determines EyeDetect is not covered by EPPA at all, then no exception is needed. However, if a court determines EyeDetect is covered by EPPA as a "similar device," then the ongoing investigation exception might not apply, since it specifically references polygraph examinations.

This is precisely why Converus developed EyeDetect+ — by adding physiological measurement channels that meet the functional definition of polygraph, EyeDetect+ is designed to fall squarely within EPPA's exceptions [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Arizona became one of the first states to formally recognize EyeDetect+ as meeting polygraph requirements for law enforcement screening after the Arizona Peace Officer Standards and Training Board adopted the U.S. Department of Labor's official polygraph definition in 2023 [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position.

The safest approach for employers using standard EyeDetect is to follow all EPPA procedures: document the economic loss, establish individualized reasonable suspicion, provide the 48-hour written notice, obtain informed consent, and respect the employee's right to refuse without retaliation.

Critical Compliance Mistakes Employers Make

The most common compliance mistakes in theft investigations include testing employees without individualized reasonable suspicion, failing to provide the 48-hour written notice, using test results as the sole basis for termination, retaliating against employees who refuse testing, and testing employees who did not have access to the property in question.

EPPA specifically provides that the ongoing investigation exemption "shall not apply if an employee is discharged, disciplined, denied employment or promotion, or otherwise discriminated against in any manner on the basis of the analysis of a polygraph test chart or the refusal to take a polygraph test, without additional supporting evidence" [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. This means employers must always have corroborating evidence beyond the test results alone. Our HR step-by-step guide to investigating employee theft walks through the proper procedures in detail.

Speed, Cost, and Scalability

EyeDetect offers significant advantages in speed and throughput. Tests take 15 to 30 minutes with results available within 5 minutes [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. One proctor can manage up to three EyeDetect stations simultaneously, enabling approximately 30-40 tests per day [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Proctor training requires just 3 to 4 hours [7]Verified EyeDetect Training & Certification
Confirms proctor training takes 3-4 hours, administrator training 2-3 hours, and advanced test writer training 5-6 hours.

Traditional polygraph examinations require a minimum of 90 minutes according to APA Standards of Practice [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit, and a member examiner shall not conduct more than five examinations per day [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit. Polygraph examiners undergo approximately 10 weeks of training and ongoing evaluation [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position.

On cost, EyeDetect tests typically range from approximately $125 to $400 per test depending on location, provider, and test type [8]Verified The eyes expose our lies. Now AI is noticing.
Confirms EyeDetect machine cost of $4,000 plus $50-$150 per test scored, and that the federal government has paid $2,200 for some polygraph exams. One CNN report noted that an EyeDetect machine costs approximately $4,000, with per-test scoring fees of $50 to $150 [8]Verified The eyes expose our lies. Now AI is noticing.
Confirms EyeDetect machine cost of $4,000 plus $50-$150 per test scored, and that the federal government has paid $2,200 for some polygraph exams. Traditional polygraph examinations for theft-specific investigations generally range from $300 to $800, though government agencies have paid significantly more [8]Verified The eyes expose our lies. Now AI is noticing.
Confirms EyeDetect machine cost of $4,000 plus $50-$150 per test scored, and that the federal government has paid $2,200 for some polygraph exams. For detailed pricing, see our EyeDetect cost guide.

Accuracy Comparison

Converus reports EyeDetect accuracy of 86 to 88% for screening tests and up to 90% for diagnostic (single-issue) tests [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. However, independent scrutiny has raised important questions. A report by the European Polygraph Association noted that Converus derives its 86% accuracy rate from lab and field studies, but acknowledged that accuracy varied from study to study [9]Verified EyeDetect Marketer Converus Spends Hundreds of Thousands of Dollars Lobbying Federal Agencies
Confirms independent scrutiny of accuracy claims, unpublished Colombia field study, and criticism from academic experts regarding EyeDetect validation. One peer-reviewed study found that EyeDetect was 85% accurate in correctly classifying participants [10]Verified Generalizability of an Ocular-Motor Test for Deception to a Mexican Population
Confirms EyeDetect correctly classified approximately 85% of truthful and deceptive participants in peer-reviewed study. Experts outside Converus have expressed concern that the correlation between cognitive load and lying "doesn't always apply" [9]Verified EyeDetect Marketer Converus Spends Hundreds of Thousands of Dollars Lobbying Federal Agencies
Confirms independent scrutiny of accuracy claims, unpublished Colombia field study, and criticism from academic experts regarding EyeDetect validation.

The American Polygraph Association's meta-analytic survey found that validated polygraph techniques intended for event-specific (single-issue) diagnostic testing produced an aggregated decision accuracy of 89%, with a confidence interval of 83% to 95% [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms APA meta-analysis found 89% aggregated decision accuracy for event-specific diagnostic testing with confidence interval of 83%-95%. Multi-issue screening techniques showed a mean accuracy of 85% [11]Verified Meta-Analytic Survey of Criterion Accuracy of Validated Polygraph Techniques
Confirms APA meta-analysis found 89% aggregated decision accuracy for event-specific diagnostic testing with confidence interval of 83%-95%. The APA requires that evidentiary polygraph techniques demonstrate an unweighted average accuracy rate of 90% or greater [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit.

Research on hybrid approaches is promising. A 2021 study on hybrid polygraph and ocular-motor deception tests found that the hybrid DLC protocol achieved 88% accuracy for specific-incident testing, while the hybrid MCT protocol achieved 91% accuracy for multi-issue screening [12]Verified Hybrid Polygraph and Ocular-Motor Deception Tests for Screening and Specific-Incident Investigations
Confirms hybrid DLC protocol achieved 88% accuracy for specific-incident testing and hybrid MCT protocol achieved 91% accuracy for multi-issue screening. This suggests combining eye-tracking with traditional polygraph sensors can enhance overall accuracy. Our EyeDetect accuracy analysis provides a deeper dive.

Legal Standing and Court Admissibility

EyeDetect has extremely limited court admissibility. The first time EyeDetect results were admitted as evidence was in 2018, when a New Mexico judicial district court judge granted a Daubert motion in a criminal case [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. According to court filings, as of late 2024, a jury has only considered EyeDetect results once in the U.S. — in that same New Mexico case [13]Verified Can a test detect those lyin' eyes? A Utah company says yes. Judges aren't so sure
Confirms that as of late 2024, a jury has only considered EyeDetect results once in the U.S., and prosecutors consistently object to its admission. Several Utah defense attorneys have attempted to submit EyeDetect results in criminal cases, with prosecutors consistently objecting [13]Verified Can a test detect those lyin' eyes? A Utah company says yes. Judges aren't so sure
Confirms that as of late 2024, a jury has only considered EyeDetect results once in the U.S., and prosecutors consistently object to its admission.

Traditional polygraph has decades more case law. Polygraph evidence has been allowed in over 20 states and in 9 of the 12 federal circuits under the Daubert standard [14]Verified The Polygraph and Forensic Psychiatry
Confirms polygraph evidence has been allowed in over 20 states and 9 of 12 federal circuits, and discusses the Daubert standard's impact on admissibility. While admissibility varies significantly by jurisdiction — some states ban it completely, others allow results by stipulation — polygraph has established legal pathways that EyeDetect has not yet developed.

The Human Element: Why It Matters for Theft Cases

One of the most significant practical differences is the pre-test and post-test interview process in polygraph examinations. When subjects fail their polygraph tests, examiners begin the post-interview stage and attempt to elicit admissions or confessions [15]Verified Polygraph Testing (EBSCO Research Starters)
Confirms post-test interview procedures where examiners attempt to elicit admissions or confessions from subjects judged to have failed. This human element has long been recognized as one of the polygraph's most valuable features in criminal and workplace investigations [4]Verified Investigations: Polygraph Use
Confirms that polygraph instruments cannot independently detect deception; only trained examiners can make credibility determinations.

Early research in the field demonstrated the value of physiological approaches in theft detection. A 1936 study comparing cardio-pneumo-psychograph methods with association methods found that the physiological instrument showed higher diagnostic accuracy for detecting deception in theft cases [16]Verified A comparison of the cardio-pneumo-psychograph and association methods in the detection of lying in cases of theft among college students
Confirms physiological instruments showed higher diagnostic accuracy than word association testing for detecting deception in theft cases. The role of skilled examiners in interpreting physiological data remains essential — polygraph instruments cannot independently detect deception without trained human interpretation [4]Verified Investigations: Polygraph Use
Confirms that polygraph instruments cannot independently detect deception; only trained examiners can make credibility determinations.

EyeDetect's fully automated nature means it cannot probe deeper with follow-up questions, observe behavioral cues, or create the psychological environment that sometimes leads to voluntary disclosures during the examination process.

Screening Large Groups of Employees

EyeDetect's strongest use case in theft investigations is screening larger groups of employees who had access to the area where theft occurred. Consider a retail store where cash has been disappearing from registers operated by 15 different employees. Testing all 15 with a traditional polygraph would take days and cost thousands of dollars. EyeDetect can screen the entire group much more efficiently, identifying individuals whose responses warrant further investigation.

This screening approach is particularly valuable for warehouse and distribution center theft investigations where dozens of employees may have had access to missing goods. The Association of Certified Fraud Examiners estimates that organizations lose 5% of revenue to fraud each year [17]Verified Occupational Fraud 2024: A Report to the Nations
Confirms CFE estimate that organizations lose 5% of revenue to fraud each year, based on 1,921 cases across 138 countries, and asset misappropriation — which includes employee theft — accounts for 89% of fraud cases studied in the 2024 Report to the Nations [17]Verified Occupational Fraud 2024: A Report to the Nations
Confirms CFE estimate that organizations lose 5% of revenue to fraud each year, based on 1,921 cases across 138 countries.

Budget-Constrained Small Businesses

Small business owners often face the most devastating impact from employee theft but have the fewest resources to investigate it. The ACFE's 2024 data shows that median fraud losses for organizations with fewer than 100 employees were $141,000 [17]Verified Occupational Fraud 2024: A Report to the Nations
Confirms CFE estimate that organizations lose 5% of revenue to fraud each year, based on 1,921 cases across 138 countries — a potentially existential loss for a small operation.

EyeDetect's lower per-test cost makes credibility assessment accessible to businesses that otherwise could not afford any form of lie detection. A small restaurant experiencing ongoing cash shortages, a local shop dealing with inventory losses, or a small office suspecting embezzlement can use EyeDetect as a first step. Learn more about the scale of the problem in our employee theft statistics guide.

Preliminary Investigation Phase and Tiered Approaches

Even in larger organizations with ample budgets, EyeDetect can serve as a preliminary investigation tool. By screening a broad pool of employees first, investigators can narrow the field before investing in more expensive and time-intensive polygraph examinations. Converus positions this as a "successive hurdles" approach [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position, which aligns with APA Standards of Practice recommending that screening results be used in such a manner with additional testing of flagged individuals [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit.

When EyeDetect is used in conjunction with polygraph, and both tests agree on the same result, practitioners report combined accuracy approaching 97% [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. This tiered approach optimizes investigative resources while maintaining thoroughness.

High-Value Theft with Potential Litigation

When significant money is at stake and the investigation may lead to criminal prosecution, civil litigation, or contested termination proceedings, traditional polygraph remains the stronger choice. Polygraph has decades of legal precedent, clearer EPPA compliance pathways, and greater admissibility in legal proceedings.

Field studies of polygraph testing in actual criminal investigations have provided real-world validity data. Research by Elaad (1990) examining the Concealed Information Test in real-life criminal investigations demonstrated the practical applicability of physiological deception detection in actual field conditions [18]Verified Detection of Guilty Knowledge in Real-Life Criminal Investigations
Foundational field study providing real-world validity data for physiological deception detection in actual criminal investigations. A comparison study found that logistic regression classification accuracy did not significantly differ between laboratory and field settings, supporting the generalizability of controlled research to real-world applications [19]Verified A comparison of polygraph data obtained from individuals involved in mock crimes and actual criminal investigations
Confirms logistic regression classification accuracy did not significantly differ between laboratory and field settings, supporting generalizability of research.

When Admissions and Confessions Are the Goal

If an employer's primary goal is to recover stolen property or obtain a confession, the traditional polygraph examination with its structured interview process provides significantly more opportunity for this outcome. The polygraph process creates psychological conditions in which subjects who are deceptive may choose to make admissions during pre-test or post-test interviews [15]Verified Polygraph Testing (EBSCO Research Starters)
Confirms post-test interview procedures where examiners attempt to elicit admissions or confessions from subjects judged to have failed.

The polygraph's role in law enforcement internal affairs investigations has been well established. Judicial precedent permits agencies to require employees to submit to polygraph examinations during bona fide internal investigations, provided strict procedural safeguards are observed [20]Verified Use of Polygraph in Law Enforcement Internal Affairs Investigations
Confirms judicial precedent permits law enforcement agencies to require polygraph during internal investigations with strict procedural safeguards. This body of legal precedent provides a framework that employers can follow with confidence.

Learn more about why polygraph testing matters for protecting your business and how the examination process creates opportunities for resolution.

Cases Requiring Maximum Legal Defensibility

When the employer expects the investigation results to be challenged — whether through wrongful termination claims, union grievances, or civil litigation — polygraph's established legal framework provides greater defensibility. Courts have decades of experience evaluating polygraph evidence, examiner qualifications, and examination procedures.

The 1983 OTA report established an early framework for evaluating polygraph science that continues to inform legal proceedings [21]Verified The 1983 OTA Report: How Congress Evaluated Polygraph Science
Foundational congressional evaluation of polygraph science that informed EPPA legislation. While that report identified limitations, the subsequent development of validated techniques and standardized procedures has strengthened polygraph's legal position considerably. Our guide to the 1983 OTA report provides historical context that remains relevant today.

Preparation and Legal Review

Before administering any credibility assessment, consult with an employment law attorney familiar with both federal EPPA requirements and your state's specific regulations. Document the specific economic loss — date discovered, nature of the loss, estimated value, and evidence supporting the loss. Identify all employees who had access to the property or area where the theft occurred.

Develop individualized reasonable suspicion documentation for each employee you intend to test. General suspicion directed at all employees with access is insufficient for testing any one individual. Prepare the required 48-hour written notice for each employee, describing the specific incident, the economic loss, the basis for suspicion, and the employee's rights.

Selecting and Administering the Test

Choose the appropriate EyeDetect test format based on your investigation needs. For screening multiple employees with access to stolen property, the Multi-issue Comparison Test (MCT) is typically appropriate. For targeted investigation of a specific suspect regarding a defined theft incident, the Directed Lie Comparison (DLC) diagnostic test is the better choice [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position.

Converus requires operators to complete specific training before administering tests. Test proctor training takes 3 to 4 hours and includes practical exercises [7]Verified EyeDetect Training & Certification
Confirms proctor training takes 3-4 hours, administrator training 2-3 hours, and advanced test writer training 5-6 hours. Test writers who create customized tests require additional training of 5 to 6 hours [7]Verified EyeDetect Training & Certification
Confirms proctor training takes 3-4 hours, administrator training 2-3 hours, and advanced test writer training 5-6 hours. All tests must be administered in a distraction-free environment, and the subject must be positioned properly with eyes calibrated to the infrared camera [7]Verified EyeDetect Training & Certification
Confirms proctor training takes 3-4 hours, administrator training 2-3 hours, and advanced test writer training 5-6 hours.

Ensure informed consent from each employee. Explain the process, remind them of their right to refuse, and document everything. The workplace polygraph policy guide provides helpful frameworks that apply equally to EyeDetect.

Interpreting Results and Next Steps

EyeDetect produces a Converus Credibility Score from 1 to 99. Scores of 50-99 indicate credibility, while scores of 1-49 indicate deception [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. Scores closer to the extremes represent higher confidence levels.

Never use EyeDetect results as the sole basis for any employment action. EPPA requires additional supporting evidence beyond test results [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. If EyeDetect results flag an employee as deceptive, consider following up with a traditional polygraph examination conducted by a professional examiner. This combined approach provides the strongest investigative foundation.

Document all results, maintain strict confidentiality, and follow your organization's established investigation procedures. Results should only be disclosed to authorized persons as required by law.

EyeDetect Cost Breakdown

EyeDetect testing costs vary based on provider, location, and test type. Based on publicly available provider pricing, individual EyeDetect tests range from approximately $60 to $400 [8]Verified The eyes expose our lies. Now AI is noticing.
Confirms EyeDetect machine cost of $4,000 plus $50-$150 per test scored, and that the federal government has paid $2,200 for some polygraph exams. The underlying licensing costs from Converus involve purchasing an EyeDetect Station (approximately $4,000) plus per-test scoring licenses [8]Verified The eyes expose our lies. Now AI is noticing.
Confirms EyeDetect machine cost of $4,000 plus $50-$150 per test scored, and that the federal government has paid $2,200 for some polygraph exams. CNN reported per-test scoring fees of $50 to $150 from Converus directly [8]Verified The eyes expose our lies. Now AI is noticing.
Confirms EyeDetect machine cost of $4,000 plus $50-$150 per test scored, and that the federal government has paid $2,200 for some polygraph exams, while end-user provider pricing includes the provider's markup and operational costs.

For employers conducting their own in-house testing, the economics become more favorable at scale. With one proctor managing three stations and conducting 30+ tests per day [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position, the effective per-test cost can be significantly lower than outsourcing. Our EyeDetect cost guide provides current pricing details.

Polygraph Cost Comparison

Traditional polygraph examinations for theft-specific investigations generally range from $300 to $800, depending on location, examiner experience, and complexity. Government agencies have paid up to $2,200 for some examinations [8]Verified The eyes expose our lies. Now AI is noticing.
Confirms EyeDetect machine cost of $4,000 plus $50-$150 per test scored, and that the federal government has paid $2,200 for some polygraph exams. The APA standard limiting examiners to five tests per day [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit combined with the minimum 90-minute session length [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit means examiner availability often drives pricing.

For a scenario involving 15 employees, the cost comparison is stark: EyeDetect screening might cost $1,875 to $6,000 total, while polygraph testing for all 15 could run $4,500 to $12,000. The combined approach — screening all 15 with EyeDetect and then polygraphing the 3-4 flagged individuals — balances cost with thoroughness.

What the Research Actually Shows

Converus reports that EyeDetect achieves 86 to 88% accuracy for screening tests, based on published research primarily conducted by the Converus Science Team [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. The diagnostic DLC test is reported at 90% accuracy based on field data [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. These studies have been published in peer-reviewed venues — as of 2024, the Converus Science Team had published 19 articles or reports about the technology [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position.

However, important caveats apply. The European Polygraph Association reported that the accuracy varied from study to study [9]Verified EyeDetect Marketer Converus Spends Hundreds of Thousands of Dollars Lobbying Federal Agencies
Confirms independent scrutiny of accuracy claims, unpublished Colombia field study, and criticism from academic experts regarding EyeDetect validation. A WIRED investigation noted that the company decided not to publish results of its first field experiment in Colombia, which appeared to show EyeDetect working erratically — performing well with educated applicants but ineffectively with less-educated participants [9]Verified EyeDetect Marketer Converus Spends Hundreds of Thousands of Dollars Lobbying Federal Agencies
Confirms independent scrutiny of accuracy claims, unpublished Colombia field study, and criticism from academic experts regarding EyeDetect validation. Independent experts like Professor Ewout Meijer of Maastricht University have acknowledged that while the cognitive load approach has "some promise," there are "boundary issues" where the correlation between cognitive load and lying doesn't always apply [9]Verified EyeDetect Marketer Converus Spends Hundreds of Thousands of Dollars Lobbying Federal Agencies
Confirms independent scrutiny of accuracy claims, unpublished Colombia field study, and criticism from academic experts regarding EyeDetect validation.

Factors That May Affect EyeDetect Results

Several factors can potentially affect EyeDetect results in workplace theft investigations:

Reading ability and education level may influence performance on the text-based test format. Converus has acknowledged that early field testing showed the system was less effective with participants who had reading difficulties [9]Verified EyeDetect Marketer Converus Spends Hundreds of Thousands of Dollars Lobbying Federal Agencies
Confirms independent scrutiny of accuracy claims, unpublished Colombia field study, and criticism from academic experts regarding EyeDetect validation. Eye conditions, medications, and fatigue can potentially affect pupillometry measurements. The test requires adequate vision and the ability to focus on a computer screen for 15-30 minutes.

Base rate effects are significant — when screening large groups for a rare behavior (e.g., one thief among 50 employees), even a highly accurate test will generate proportionally more false positives among innocent employees. Converus's own chief scientist Dr. Kircher has advised against relying solely on EyeDetect or any single screening technology, noting that "even if a test is 90 percent accurate, about 10 percent of the tested population would fail it, and the vast majority of those individuals who fail the test would be innocent" [9]Verified EyeDetect Marketer Converus Spends Hundreds of Thousands of Dollars Lobbying Federal Agencies
Confirms independent scrutiny of accuracy claims, unpublished Colombia field study, and criticism from academic experts regarding EyeDetect validation.

A 2024 study examining polygraph reliability in criminal investigations found that polygraphs yield unreliable conclusions in approximately 50 to 65 percent of cases involving truthful innocent individuals [22]Verified How Reliable are Polygraph Examinations in Criminal Investigations? An Empirical Assessment
Provides empirical data on polygraph reliability in criminal contexts, finding unreliable conclusions in 50-65% of cases involving truthful innocent individuals. While this study focused on traditional polygraph, it underscores the importance of never relying on any single credibility assessment tool.

Why Combining Technologies Provides the Best Results

The most effective approach for many theft investigations combines EyeDetect's efficiency with polygraph's depth. A 2021 study published in Psychology Applications & Developments VII examined hybrid polygraph and ocular-motor deception tests and found that the hybrid DLC protocol achieved 88% accuracy for specific-incident testing, while the hybrid MCT protocol achieved 91% accuracy for multi-issue screening [12]Verified Hybrid Polygraph and Ocular-Motor Deception Tests for Screening and Specific-Incident Investigations
Confirms hybrid DLC protocol achieved 88% accuracy for specific-incident testing and hybrid MCT protocol achieved 91% accuracy for multi-issue screening. This research demonstrates that combining eye-tracking technology with traditional polygraph sensors enhances accuracy over either method alone.

The practical workflow typically involves: (1) Using EyeDetect to screen the broader group of employees with access to the theft area; (2) Identifying individuals whose EyeDetect results warrant further investigation; (3) Conducting traditional polygraph examinations on flagged individuals with a professional examiner; (4) Using the polygraph pre-test and post-test interviews to explore issues further and create opportunities for admissions.

Practitioners report that when both EyeDetect and polygraph indicate deception on the same issue, the combined outcome confidence approaches 97-99% [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. This "successive hurdles" approach is consistent with APA guidance on using screening results to inform follow-up testing [3]Verified APA Standards of Practice
Confirms APA requirements for 90% minimum accuracy for evidentiary techniques, 90-minute minimum examination scheduling, and 5-exam daily limit.

For employers considering the combined approach, our guide on cultivating trust in the workplace with lie detector tests explains how to implement credibility assessment programs that strengthen rather than undermine workplace culture.

Retail and Point-of-Sale Theft

Retail environments are among the most common settings for employee theft investigations. The ACFE's 2024 Report to the Nations found that asset misappropriation — the category encompassing employee theft — accounts for 89% of all fraud cases, though it produces the lowest median loss at $120,000 per case [17]Verified Occupational Fraud 2024: A Report to the Nations
Confirms CFE estimate that organizations lose 5% of revenue to fraud each year, based on 1,921 cases across 138 countries. For retailers with multiple register operators, EyeDetect's screening capability allows efficient testing of all employees with cash-handling access.

The deterrent effect of credibility assessment programs should not be underestimated. When employees know that their organization uses lie detection technology as part of its loss prevention strategy, they may be less likely to engage in theft. EyeDetect's non-contact format can be incorporated into routine screening without the psychological weight of being connected to a polygraph machine.

Warehouse and Distribution

Warehouse and distribution center theft presents unique challenges due to the large number of employees with access to inventory. Our warehouse theft polygraph guide covers the specific considerations for these environments. EyeDetect's scalability makes it particularly well-suited for initial screening in facilities where 50 or more employees may have had access to missing goods.

The ACFE data shows that wholesale trade has one of the highest median losses per fraud incident at $361,000 [17]Verified Occupational Fraud 2024: A Report to the Nations
Confirms CFE estimate that organizations lose 5% of revenue to fraud each year, based on 1,921 cases across 138 countries, while manufacturing sees median losses of $267,000 [17]Verified Occupational Fraud 2024: A Report to the Nations
Confirms CFE estimate that organizations lose 5% of revenue to fraud each year, based on 1,921 cases across 138 countries. These figures underscore the financial justification for investing in credibility assessment technologies.

Financial Services and Embezzlement

For financial services companies dealing with embezzlement or funds misappropriation, the stakes are often higher and the investigations more complex. The ACFE found that banking and financial services had 305 fraud cases in their 2024 study [17]Verified Occupational Fraud 2024: A Report to the Nations
Confirms CFE estimate that organizations lose 5% of revenue to fraud each year, based on 1,921 cases across 138 countries. EyeDetect can serve as a preliminary screening tool, but these cases frequently require the deeper investigative capability of traditional polygraph.

The EPPA provides specific exemptions for firms providing security services, including armored car, security alarm, and security guard companies [5]Verified Employee Polygraph Protection Act of 1988 (Full Text)
Confirms the ongoing investigation exception at 29 U.S.C. § 2006(d), penalty provisions, employer prohibitions, and all four exception requirements. EyeDetect+ meets the federal definition of polygraph that enables use under these EPPA exemptions [1]Verified EyeDetect Product Information and FAQs
Confirms EyeDetect accuracy claims of 86-88%, test duration of 15-30 minutes, development history at University of Utah, and EPPA position. For fraud-related investigations beyond theft, see our guide on polygraph testing for workers' compensation fraud.

Educational Institutions

Schools, colleges, and universities face their own theft and fraud challenges, from equipment theft to financial misconduct. Our guide on polygraph testing for educational institutions discusses how credibility assessment can be implemented in academic settings while respecting the unique dynamics of educational environments.