Lying and remembering differently are not the same
In a classic 1974 experiment, people watched footage of an accident and were asked about speed. Those whose question said the cars had "smashed into" each other thought they had been going faster than those whose cars had "collided." A week later, the first group more often "remembered" broken glass that was not in the footage. No one was lying: the question added to the memory, and memory presented it as its own.
By the time of the oath, the story has already been told several times: to a friend, a patrol officer, an investigator, in a written statement, to a lawyer, to yourself before sleep. Each retelling is not replaying a recording but rebuilding it, and it picks up details from other people's questions, from the news, from the version that sounded more coherent. The story gets polished: hesitations disappear, details appear, and the voice grows firmer.
A diagram, not measurements: confidence grows with every retelling. Memory does not keep up - repetition strengthens not accuracy but the feeling that you remember well.
Why an oath holds less power than it seems
A deliberate lie is arithmetic. The truth costs a lot right now: a job, a relationship, a brother who will be sent to prison, admitting that you already lied before. A lie will cost a lot later, and only if you are caught. And people are rarely caught: you need to prove not a memory error but intent - that the person knew the truth and said something else. From the outside, there is almost no visible difference between "lied" and "mixed it up."
The second mechanism: the point of no return is passed long before the courtroom. A version is chosen in the first conversation with police, frightened and rushed. After that, every retelling nails you to it more firmly - backing out now means admitting you were untruthful every previous time.
And third: lying in court often works. The courtroom reads behavior - whether someone makes eye contact, fidgets, or answers smoothly. In a large research review, people distinguished lies from truth correctly in about 54 cases out of 100, compared with 50 for a coin toss, and professionals - police officers and judges - did no better than anyone else.
This many lies and truths are identified correctly; a coin toss gets 50
If you yourself are a witness, what works is not willpower but the form of your answer: keep separate what you saw from what you understood or filled in later.
- Say "I don't remember" when you do not remember: it is a normal answer, not a failure
- Separate what you saw from your conclusion: "he was shouting" is a fact, "he was angry" is a conclusion
- Ask for the question to be repeated if it already contains the answer
- Fill out the picture into a complete whole: gaps in memory are normal
- Fit today's story to yesterday's just to make it smoother
Finish reading in the Miqo app
1 more minute, then a quiz at the end. Find it under “Crime → Perjury”
Point your phone camera here to open the App Store or Google Play.
Sources
- Loftus & Palmer, "Reconstruction of Automobile Destruction," 1974
- Bond & DePaulo, meta-analysis of lie detection accuracy, 2006