It is one of the most searched questions about lie detection in the UK, and it is usually asked by two very different people: someone worried about being falsely accused, and someone hoping to hide something. The honest answer is that the countermeasures circulating online were written about the polygraph, most of them do not work well even there, and almost none of them apply to the eye-based testing we use.
This article explains what people actually try, what an examiner sees when they do, and why an invalid test is not the same thing as a pass.
Polygraph folklore is decades old and remarkably consistent. The usual advice is to control breathing during control questions, cause yourself pain to spike arousal, take a sedative, or occupy your mind with mental arithmetic so that lying feels no different from telling the truth.
All of it targets the same thing: the autonomic nervous system. A polygraph records breathing, pulse, blood pressure and skin conductance through sensors attached to the body, so anything that muddies those signals in theory muddies the chart.
EyeDetect® does not measure arousal. It records involuntary eye behaviour — pupil dilation, blink rate and timing, fixation, reading behaviour and response latency — while you answer true/false statements on a screen. Nothing is attached to you, so there is no sensor to fool by breathing differently or tensing a muscle.
The measurements also happen far faster than deliberate control allows. Pupil response to cognitive load is sampled many times per second and is not something people can consciously drive. Deciding to lie takes more mental effort than telling the truth, and that extra effort shows in the data whether or not you feel calm.
Deliberate non-cooperation is visible. Refusing to read the statements, answering at random, staring away from the screen or answering before the statement can possibly have been read all produce data that fails quality checks. The outcome is an inconclusive or invalid test, recorded as exactly that in the written report.
That matters, because an invalid test is not a pass. If you commissioned the test to get an answer, you will be told plainly that no reliable result was produced and why. In practice, an obviously uncooperative examinee tells the person who asked for the test almost as much as a failed one.
Far more people search this question because they are innocent and frightened than because they intend to cheat. Nerves are normal and expected. Being anxious, upset or angry about being accused does not fail you — the technology compares your responses across a structured question set rather than judging whether you seem nervous.
What genuinely affects data quality is different: severe sleep deprivation, intoxication, certain eye conditions and some medications. We run a short suitability check before every appointment so that nobody travels for a test that cannot produce a clean result.
EyeDetect® is 86-88% accurate on validated question sets. That is high, comparable to or better than a well-run polygraph, and it is not certainty. Any provider telling you a lie detector is 100% accurate is misrepresenting the science.
Our reports state the accuracy of the method alongside the result for each question, so you can weigh the outcome properly rather than treating it as a verdict.
People who plan to interfere with a test rarely think through what they are actually buying. There is no version of the report that says "passed". The two realistic outcomes of interference are an inconclusive result or an invalid one, and both are written down along with the behaviour that produced them: statements not read, answers given faster than reading is possible, the screen ignored, repeated movement.
In practice that reads far worse than a straightforward failed result. A failed result at least looks like someone who engaged with the process. An invalid one looks like someone who tried to break it, and whoever asked for the test — a partner, an employer, a solicitor — draws the obvious conclusion.
If you genuinely do not want to be tested, saying so is a cleaner option than sitting down and sabotaging it. Consent is required for every test we carry out and it can be withdrawn at any point, including part-way through the examination.
Most people who search this question are not planning anything. They are worried that being nervous, or being on medication, or having a bad memory of the event will count against them. It is a fair worry and the answer is reassuring: the test looks for the extra mental effort involved in deciding to lie, not for signs of stress.
Two things genuinely help. First, agree the wording of the questions in the pre-test discussion until each one is a plain yes-or-no fact you are certain about — vagueness is the enemy of a clean result, not nerves. Second, be honest about anything that might affect concentration, such as medication, tiredness, a migraine or a recent night shift. We would rather move an appointment than run a test in poor conditions.
No. Nervousness is normal and expected. The test compares responses across a structured set of statements rather than measuring how anxious you appear.
They do not produce a pass — they degrade data quality and usually produce an invalid test. Attending intoxicated means the appointment cannot go ahead.
No. An inconclusive or invalid test means no reliable result was produced, and the report says so explicitly.
They reduce physical arousal, which is what a polygraph records, so the folklore has some basis there. Eye-based testing does not use arousal at all — but heavy sedation reduces concentration, which degrades data quality and pushes the result towards invalid rather than towards a pass.
No method of deliberate control has been shown to produce a passing result on eye-based testing, because the measurements happen in fractions of a second and are not under conscious command. What trained interference does produce is data that fails quality checks.
Services · Pricing · How it works · Book a test · Contact