The modern polygraph’s History spans a full century. The first modern polygraph came to life in 1921 , making 2021 the centennial of this controversial truth-detection technology . Most people picture lie detectors as classic machines that measure physical responses like breathing, heart rate, blood pressure, and sweat gland activity . These body changes show stress during questioning and that indicates possible deception . The reliability of these tests faces heavy debate, with error rates from 15% (according to former U.S. Attorney General John Ashcroft) to possibly 40% or higher according to critics .
Polygraphs have stayed relevant throughout lie detection’s history. The technology changed dramatically since the 1990s after the introduction of software systems revolutionized the field . Modern systems use advanced algorithms like the CPS (Computerized Polygraph System) that applies multivariate linear discriminant function analysis to interpret results . The current approach relies on evidence-based procedures such as the Test for Espionage and Sabotage and the Directed Lie Test .
Polygraphs gained ground in courtrooms during the mid-20th century , but their legal status changed drastically. The landmark Frye v. United States case (1923) set precedent to exclude polygraph results from federal courts . The U.S. Court of Appeals declared certain military polygraph evidence rules unconstitutional in 1996 . These tests rarely serve as trial evidence now but remain valuable investigative tools . More than 20 federal agencies still use polygraphs to screen employees . Based on decades as an intelligence officer and hundreds of cases, this is my life’s work — the most up-to-date and comprehensive guidance in the world
From Trial by Ordeal to the First Polygraph (Pre-1920s)
Image Source: IEEE Spectrum
People tried to spot lies long before they had electronic sensors and computer algorithms. Humans developed better ways to separate truth from lies throughout history.
Ancient lie detection methods in Hindu and Chinese cultures
Chinese officials around 1000 BC created what we might call the first physiological lie detector test. They made suspects put rice in their mouths during questioning [1]. The suspect was guilty if the rice stayed dry after spitting it out. People believed anxiety from lying reduced saliva production [2]. This clever method linked emotional states to body functions—a principle that modern polygraph technology still uses today.
Hindu cultures in ancient India created their own ways to catch liars. The “Suspected Noise Test” made suspects breathe smoke from burning chili peppers while saying certain words loudly [3]. Officials thought people who didn’t cough were lying. They also had a clever “donkey test.” Suspects walked into a dark tent to pull a soot-covered donkey’s tail. Clean hands meant the person feared getting caught—this test played with psychology rather than body responses [2].
These basic methods showed remarkable understanding of how stress affects the body when people lie.
19th century lie detection science: Lombroso’s hydrosphygmograph
Italian criminologist Cesare Lombroso took the first truly scientific approach to lie detection in 1895 [4]. He changed a medical tool called the hydrosphygmograph to measure blood pressure and pulse changes during questioning [4].
The device had a water-filled tank where the subject’s fist fit inside, sealed with a rubber membrane [5]. Blood pressure and pulse changes moved from the fist to the water. This created measurable air pressure changes that a spinning drum recorded [5]. This new approach bridged the gap between ancient wisdom and modern scientific methods.
When was the polygraph invented: John Larson’s 1921 device
John Augustus Larson built on Lombroso’s work and created the modern polygraph in 1921. Larson, a police officer with a PhD in physiology, built a device that could track multiple body responses at once [6].
Larson worked at the Berkeley Police Department under Chief August Vollmer and created what he called a “cardio-pneumo psychogram” [6]. His invention gave continuous readings of blood pressure, breathing, and later, skin conductivity [7]. The machine’s pens recorded these body changes on moving graph paper during questioning [5].
William Hightower’s case in summer 1921 marked the first real test of the device. Hightower faced charges of murdering a priest in San Francisco [6]. The San Francisco Call and Post shared the results next day, showing the graphs with arrows pointing to each suspected lie [6].
Larson named his invention the polygraph, which means “many writings” in Greek. The press came up with the popular term “lie detector” [6], even though the machine actually detects stress responses rather than lies.
The Analog Era: Keeler, Reid, and the Rise of Mechanical Polygraphs
Image Source: PBS
The polygraph went through dramatic changes from the 1920s to 1950s. Larson’s bulky device turned into practical tools that could detect deception.
Leonarde Keeler’s Emotograph and galvanic skin response
Leonarde Keeler trained under John Larson but grew frustrated with early polygraph technology’s limits. He created the “Emotograph” in 1925, which became the first patented lie detection device in America [2]. His state-of-the-art design replaced messy smoke paper recording with ink pens. This change made the quickest way to prepare and test subjects [2].
Keeler’s design stood out because it was portable. The device came in a wooden mahogany case that looked like a traveling box. Later versions used a custom metal toolbox from Kennedy Tool Box company [3]. This practical improvement made testing easier, and police departments bought 60-80 units across California and beyond in just three months [3].
Keeler made his biggest technical breakthrough in 1938-1939. He added galvanic skin response (GSR) measurement to the polygraph [2][1]. This third physiological channel tracked skin conductivity changes from sweating, which increases during emotional stress [3]. The GSR worked with a two-piece galvanometer on the subject’s fingertips. It sent a small electric current through the skin and measured conductivity changes [3].
Reid Technique and the Control Question Test (CQT)
John E. Reid revolutionized how polygraphs worked in the 1940s. The former Chicago police officer, psychologist, and polygraph expert created the Control Question Test (CQT) [4]. Earlier methods relied on simple recognition, but Reid’s approach focused on attention – a more complex psychological process [4].
The CQT added “control questions” next to questions about the specific incident being investigated [5]. Reid speculated that subjects would show the strongest physical reactions to questions that carried the most emotional weight [4]. The sort of thing I love about attention as a psychological process is how it works. It increases arousal to specific stimuli while blocking responses to other stimuli [4].
Reid’s technique changed polygraph testing completely. It solved a major flaw in earlier methods where innocent subjects often showed stress responses to accusatory questions just because they felt anxious about being tested [4].
Limitations of analog signal interpretation
The mechanical progress brought its own challenges. Analog polygraphs turned physical responses into electrical signals that moved needles across scrolling paper [3]. Result interpretation remained highly subjective.
Examiners had to rely on their judgment and experience when physical data wasn’t clear [3]. This subjectivity became one of the biggest criticisms of polygraph reliability during the analog era.
The basic ideas behind polygraph technology stayed the same for over 80 years [3]. Experts still believed that lying caused measurable physical changes and that examiners could read these signals accurately. This lack of theoretical progress, despite better mechanics, led to the digital revolution of the 1990s.
The Digital Shift: 1990s Software Revolution in Polygraphy
Image Source: NBC News
The 1990s brought a game-changing chapter in polygraph history. Digital technology changed how lie detection worked at its core. Computer algorithms took over from mechanical parts that had been the backbone of polygraphy for 70 years. This shift brought new levels of precision and analysis that nobody had seen before.
Stoelting’s CPS and Utah Numerical Scoring System
Stoelting Company led the digital polygraphy revolution with its Computerized Polygraph System (CPS). John Kircher and David Raskin developed this system at the University of Utah in 1988 [8]. CPS built on the Utah Numerical Scoring System’s foundation. The system gives scores from +3 to -3 for breathing, skin response, blood pressure, and blood flow in extremities [9]. Scores come from comparing reactions between relevant and control questions. Positive scores show stronger reactions to control questions, while negative ones point to deception [6].
The Utah scoring method’s reliability turned out to be remarkable. Studies showed that different evaluators agreed more than 90% of the time [6]. The numbers were even more impressive when they left out unclear results – the system correctly spotted 91% of guilty subjects and 89% of innocent ones [6].
Axciton’s PolyScore and neural network models
While CPS was taking shape, Axciton Systems worked with Johns Hopkins University Applied Physics Laboratory to create PolyScore. Many experts call it the first successful algorithmic analysis system to hit the market [7]. Unlike CPS, which tried to copy what human examiners did, PolyScore took a completely different path. The system used complex math to analyze digital polygraph signals and showed results as statistical chances of lying or truth-telling [10].
PolyScore stood out because it used neural network models along with logistic regression analysis [8]. The system changed blood pressure, skin response, and breathing data into what its creators called “more fundamental” signals. These signals focused on information that mattered most for catching lies [8]. This meant PolyScore could make statistical comparisons that humans simply couldn’t match [10].
Computer-assisted polygraph systems (CAPS) in federal use
Government agencies quickly embraced these digital innovations for national security work. Computer Assisted Polygraph Systems became the go-to choice in government settings. Digital polygraphs fixed the old headaches of dealing with graph paper and ink systems [10].
The biggest game-changer was how digital systems enabled network operations. This revolutionized quality control. Examiners could send test data instantly to other experts for review [7]. This network capability led to “cyber-polygraph” systems that allowed remote testing. Examiners could now interview people through video calls while collecting body response data live from anywhere [7].
The human touch stayed crucial even with all these algorithmic advances. Digital polygraphs ended up being sophisticated tools that made examiners better at their jobs rather than replacing them completely.
Legal and Institutional Impact of Digital Polygraphs
Digital polygraphs became more sophisticated over time, but the legal world actually limited their use instead of expanding it.
Military Rule of Evidence 707 and Daubert Standard
Polygraph evidence took a major hit when the Supreme Court backed Military Rule of Evidence 707 in 1998. This rule completely blocks polygraph evidence in military trials [11]. The Court decided that this rule “does not unconstitutionally abridge the right to present a defense” [12]. Scientists still debate how reliable polygraphs really are. Studies show their “accuracy rates described as little better than could be obtained by the toss of a coin” – about 50% [11]. Many federal courts later dropped the automatic exclusion rule. They now use the Daubert standard, which lets judges decide if scientific evidence can be admitted [11].
Employee Polygraph Protection Act of 1988 (EPPA)
Whatever their digital progress, the EPPA put strict limits on polygraphs. The law stops most private employers from using lie detector tests to screen job candidates or check current employees [13]. Security firms, pharmaceutical makers, and government agencies are exceptions to this rule [14]. Anyone breaking this law faces civil penalties up to $25,597 [15]. This 1988 law has become relevant again as AI-powered affect-screening technologies emerge. Some legal experts say these new tools break the same federal rules [16].
Federal agency use: FBI, CIA, and Department of Defense
Government agencies don’t have to follow EPPA rules. The FBI, Department of Homeland Security, and Department of Defense all stepped up their polygraph use for leak investigations in 2025 [17]. DHS Assistant Secretary McLaughlin stated, “We can, should, and will polygraph personnel” [17]. These agencies typically use advanced Computer Assisted Polygraph Systems (CAPS). Courts still usually won’t accept polygraph results as evidence, even though these institutions use them extensively [18].
Modern Innovations: AI, AVATAR, and the Future of Truth Detection
Truth detection has evolved beyond traditional polygraph technology. AI and brain imaging have redefined how we detect deception in the 21st century.
AVATAR and iBorderCtrl: AI-based deception detection
AVATAR (Automated Virtual Agent for Truth Assessments in Real-Time) brings a fresh approach to lie detection. This DHS-funded technology lets a virtual border agent interview travelers and analyzes their facial expressions, body language, and vocal patterns to spot deception [19]. The developers report accuracy rates of 60-75% that reach 80% sometimes, while humans only achieve 54-60% accuracy [19].
The EU-funded iBorderCtrl system works with an “Automatic Deception Detection System.” Travelers interact with an avatar that watches their facial micro-gestures [2]. The system creates risk profiles for travelers by analyzing their webcam footage to detect potentially deceptive behavior [2].
Layered Voice Analysis (LVA) and facial micro-expression tracking
LVA technology looks past speech content and focuses on voice properties that reveal genuine emotions [3]. The system’s algorithms extract over 150 parameters from each voice segment and group them into 14 emotional categories [3]. Culture, gender, or language differences don’t affect this approach [20].
Facial micro-expression analysis shows promise because it catches involuntary facial reactions that people can’t fake [21]. These systems build on Paul Ekman’s Facial Action Coding System and identify expressions lasting just 1/15 to 1/25 of a second [21]. High-speed cameras must capture at least 30 frames per second [21].
fMRI and brain-based lie detection research
Functional Magnetic Resonance Imaging (fMRI) stands out as a breakthrough in deception detection. Unlike polygraphs that measure peripheral nervous system responses, fMRI looks directly at central nervous system activity [22]. Blood-oxygen-level-dependent imaging helps track brain activity changes over time [22].
Machine learning applications now achieve 70-90% accuracy when identifying deception [4][23]. Deception triggers activity in the prefrontal cortex, parahippocampal gyrus, precuneus, and cerebellum [5]. The limbic system, which handles emotional responses, shows little activity during deception [5].
Author’s Notes: Perspective on the Evolution and Utility of Polygraphy
As the author of this exploration into the history and application of lie detection, I want to provide you with a strategic summary that bridges the gap between historical development and modern-day practicalities. These notes act as a professional extension of our main article, distilling the most critical takeaways regarding how this technology impacts your legal, professional, and personal interests.
The Foundational Heritage
A Century of Innovation: Modern polygraphy isn’t the work of a single inventor; it is the result of a massive collaborative effort spanning the early 20th century. Key figures like John Augustus Larson (who developed the modern machine in the early 1920s) worked alongside a group of over 12 pioneers—including names like Marston, Keeler, and Reid—to turn physiological monitoring into a forensic tool.
Evolutionary Strategy: The technology is “continually refined and improved.” When undergoing or ordering a test, it is actionable to recognize that the specific age and model of the equipment can significantly influence the sophistication of the monitoring capabilities.
Historical Foundations: While we focus on modern sensors, it is helpful to acknowledge that historical physical methods laid the groundwork for today’s monitoring of heart rate, blood pressure, and respiration.
Sector-Specific Applications
Government & National Security: In this sector, the primary use case is for security clearances. If you are entering this field, expect the polygraph to be a standard gatekeeper for sensitive information access.
The Private Sector: Here, the tool is used most frequently for pre-employment screenings. It serves as a vetting mechanism to ensure integrity before a candidate is officially onboarded.
Law Enforcement & Criminal Justice: Beyond simple investigation, be aware that in certain jurisdictions, polygraph results are not just internal tools—they can be presented in court as legal evidence.
Actionable Insights for the Examinee
Physiological Monitoring: The process is designed to measure specific bodily responses. To prepare effectively, understand that the machine is tracking spikes in heart rate and respiration as a proxy for the stress caused by deception.
Consequence Management: Lying is a human trait, but in the context of law enforcement, the consequences are severe. Approach the exam with the understanding that accuracy here can dictate legal outcomes and professional reputations.
Validity Awareness: Stay informed about the ongoing debates regarding reliability. Understanding that the technology is controversial in some legal circles allows you to better navigate its use if it is ever introduced as evidence in your own case.
Societal and Forensic Impact
The Credibility Debate: Despite its widespread use, the polygraph remains at the center of a tug-of-war regarding its validity as courtroom evidence. It has fundamentally changed the criminal justice system, but it is not a “magic bullet.”
Institutional Adoption: Its continued use by major government agencies and private firms highlights its perceived value as a deterrent and a truth-verification tool, even amidst technological and legal scrutiny.
Final Takeaway
The polygraph is a sophisticated blend of psychology and physiology with a century-old lineage. Whether you are a business owner looking to screen employees or an individual facing an investigation, understanding the specific sector-driven applications and the technological evolution of the device will help you engage with the process more effectively.
Conclusion
A century of development has seen polygraph technology rise from simple physiological measurements to sophisticated digital systems that use advanced algorithms and artificial intelligence. The trip from Larson’s cardio-pneumo psychogram to today’s computerized systems shows a fascinating joining of psychology, physiology, and computer science.
The original polygraphs depended on mechanical components and subjective interpretation. Their results heavily relied on examiner expertise. The digital revolution of the 1990s changed everything. Systems like Stoelting’s CPS and Axciton’s PolyScore brought mathematical precision and statistical probability to what used to be largely user-friendly analysis. These technologies improved consistency dramatically, yet they still face major legal restrictions in courtroom settings.
Without doubt, modern truth detection goes way beyond the reach and influence of traditional polygraph methods. AVATAR and iBorderCtrl systems show how AI can analyze subtle behavioral cues without physical sensors. Brain-based approaches like fMRI target deception at its source instead of measuring peripheral nervous system responses. These developments point to a new transformation in deception detection.
The future points to integration of multiple modalities. Voice analysis, facial micro-expression tracking, and neuroimaging data will combine through unified AI systems. On top of that, quantum computing applications might soon enable live processing of these complex multimodal inputs and achieve accuracy rates we once thought impossible.
Ethical questions remain at the forefront. These technologies become more powerful and less invasive, challenging our basic assumptions about privacy and cognitive liberty. Despite technological advancement, an ancient question remains: can machines truly distinguish truth from deception, or will human judgment always play the decisive role?
Polygraph technology approaches its second century, and the search for reliable truth detection continues – proof of humanity’s endless desire to uncover deception and establish objective truth.
Key Takeaways
The evolution of polygraph technology reveals a fascinating century-long journey from mechanical devices to AI-powered systems, though fundamental questions about reliability and ethics persist.
• Digital revolution transformed accuracy: 1990s computerized systems like CPS achieved 90%+ reliability compared to subjective analog interpretations, using algorithms to analyze physiological data.
• Legal restrictions limit courtroom use: Despite technological advances, polygraphs remain largely inadmissible in courts due to accuracy concerns, though federal agencies still use them extensively.
• AI and brain imaging represent the future: Modern systems like AVATAR analyze facial expressions and voice patterns, while fMRI directly measures brain activity with 70-90% accuracy rates.
• Fundamental reliability debate continues: Even with sophisticated technology, polygraphs detect stress responses rather than lies directly, maintaining error rates that challenge their effectiveness as truth detectors.
The core challenge remains unchanged since 1921: distinguishing between stress from deception versus anxiety from being tested. As these technologies become more powerful and less invasive, they raise critical questions about privacy and cognitive liberty that society must address alongside technological advancement.
FAQs
Q1. How accurate are modern polygraph tests? Modern computerized polygraph systems can achieve accuracy rates of 90% or higher in controlled studies. However, real-world accuracy is often lower and can vary significantly depending on the specific technique used and the skill of the examiner.
Q2. Are polygraph results admissible in court? Generally, polygraph results are not admissible as evidence in most U.S. courts. The Supreme Court has upheld bans on polygraph evidence in military trials, and many federal courts apply the Daubert standard, which gives judges discretion in admitting scientific evidence.
Q3. What new technologies are being developed for lie detection? Emerging technologies include AI-based systems like AVATAR, which analyzes facial expressions and voice patterns, and brain imaging techniques like fMRI. These methods aim to detect deception more directly than traditional polygraphs.
Q4. Can employers require job applicants to take a polygraph test? In the United States, the Employee Polygraph Protection Act (EPPA) prohibits most private employers from using lie detector tests for pre-employment screening or during employment, with some exceptions for security firms and government agencies.
Q5. How does a polygraph machine work? A polygraph machine measures physiological responses such as heart rate, blood pressure, respiration, and skin conductivity during questioning. Modern digital polygraphs use sophisticated algorithms to analyze these signals and detect potential signs of deception based on stress responses.
References
[1] – https://morganpolygraph.com/index.php/2024/09/13/polygraph-tests-understanding-their-role-and-limitations/
[2] – https://iborderctrl.no/
[3] – https://www.nemesysco.com/lva-technology/
[4] – https://newsroom.haas.berkeley.edu/brain-scans-help-researchers-untangle-the-truth-about-lie-detection/
[5] – https://en.wikipedia.org/wiki/FMRI_lie_detection
[6] – https://polygraph.org/docs/polygraph_1999_281.pdf
[7] – https://polygraph.org/docs/peters-a_history_of_polygraph_digitization.pdf
[8] – https://www.nationalacademies.org/read/10420/chapter/17
[9] – https://nij.ojp.gov/library/publications/utah-numerical-scoring-system
[10] – https://www.ojp.gov/ncjrs/virtual-library/abstracts/polygraph-technology
[11] – https://www.everycrsreport.com/reports/98-358.html
[12] – https://supreme.justia.com/cases/federal/us/523/303/
[13] – https://www.dol.gov/sites/dolgov/files/WHD/legacy/files/eppac.pdf
[14] – https://en.wikipedia.org/wiki/Employee_Polygraph_Protection_Act
[15] – https://www.michaelbest.com/Newsroom/336601/New-Lie-detecting-AI-in-the-Hiring-Process-How-Will-This-Case-Unfold
[16] – https://www.law.georgetown.edu/georgetown-law-journal/in-print/volume-109/volume-109-issue-5-april-2021/the-modern-lie-detector-ai-powered-affect-screening-and-the-employee-polygraph-protection-act-eppa/
[17] – https://www.npr.org/2025/04/30/g-s1-63349/fbi-polygraph-lie-detector-leak-investigation
[18] – https://www.nytimes.com/2025/07/10/us/politics/fbi-polygraph-kash-patel.html
[19] – https://www.cnbc.com/2018/05/15/lie-detectors-with-artificial-intelligence-are-future-of-border-security.html
[20] – https://docs.ado-tech.com/books/b-trust/page/about-layered-voice-analysis-lva-V1O
[21] – https://www.ijert.org/lie-detection-based-on-facial-micro-expression-body-language-and-speech-analysis
[22] – https://pmc.ncbi.nlm.nih.gov/articles/PMC3680134/
[23] – https://www.sciencedirect.com/science/article/pii/S0149763419301873
