This is the question innocent people ask, and it deserves a straight answer: yes, a truthful person can be wrongly flagged. Credibility testing is 86-88% accurate, which means errors exist in both directions.
What matters is understanding what actually causes those errors, because most of them are avoidable and almost none of them are about nerves.
An 86-88% figure describes performance across validated question sets — the proportion of correct outcomes, not a confidence score for your particular answer. Errors split into false positives, where a truthful person is flagged, and false negatives, where a deceptive person is not.
That is why a result should be read as strong evidence rather than a verdict, and why we print the accuracy of the method alongside the outcome in every report.
In practice the biggest cause is a badly written question. If a statement is ambiguous, covers two facts at once, or uses a word the examinee understands differently from the person who asked for the test, the data reflects that confusion rather than deception.
Physical factors matter too: severe sleep deprivation, intoxication, some eye conditions and certain medications degrade data quality. These are the reasons we run a suitability check before every appointment and will postpone rather than test someone unfit.
Being frightened of the test does not fail you. EyeDetect® measures involuntary eye behaviour against a structured, repeated question set, not how anxious you look or sound, and there is no examiner in the room judging your demeanour.
Feeling angry or upset about being accused is normal and does not change the outcome either.
Read the report. It lists each question as it was actually asked, and quite often a disputed outcome traces back to wording that did not mean what the examinee thought it meant.
A retest with corrected questions, on a properly rested examinee, is the appropriate remedy. Where the stakes are high — a job, a legal matter, a marriage — a second test on refined wording is a reasonable step, not an admission of anything.
When a truthful person gets an unexpected outcome, the question is usually at fault. Three patterns cause most of it. Questions containing a word the two people define differently — "cheating", "flirting", "stealing". Questions with no time boundary, so the person is silently reaching back across years of memory. And compound questions that bundle two facts together, where one is true and one is not.
This is exactly why the pre-test discussion exists and why it is not a formality. If you read a question and think "well, it depends what you mean", say so at that moment. Once you are comfortable that every question is a plain fact you know the answer to, the main source of error has been removed.
Tell us. The first step is reviewing the report with you question by question, because a surprising number of disputes turn out to be about what a question meant rather than about the data. If the wording was ambiguous, that is a defensible reason to re-test on a rewritten question set.
Retesting is a normal part of the process, not an admission of failure. Where the original conditions were poor — illness, no sleep, an interrupted room — a second examination in better conditions is usually the right answer. We will always tell you honestly if we think a retest is unlikely to change anything.
Testing is 86-88% accurate overall, so errors occur in a minority of cases in both directions. Good question design and a proper suitability check are what keep that minority small.
Anxiety itself does not fail you. Some medications and severe sleep disruption can affect data quality, which is why they are covered in the suitability check.
Yes. Where a result is disputed, a retest with reviewed question wording is the right approach.
Tell the examiner what you take and when. Most routine prescriptions are not a problem, but anything strongly sedating reduces concentration, and poor concentration shows up as weak data rather than as deception.
Yes, though a repeat on identical wording adds less than a rewritten set. If the concern is that a question was ambiguous, rewriting it is the more useful course.
Services · Pricing · How it works · Book a test · Contact