Psychology
Why We Are Bad at Spotting Lies, and How to Write a Lie Nobody Catches
Most of us are sure we can spot a liar: they look away, start sweating, stumble over their words. Research tells a very different story. When scientists measure how well people tell truth from lies, they find we do only slightly better than flipping a coin. Here is why, and what it means when you play a bluffing game like Bullshit.
54 percent: the number that surprises everyone
In 2006, psychologists Charles Bond and Bella DePaulo published a large review of lie-detection research. They pooled hundreds of experiments in which more than twenty thousand judges tried to decide whether someone was telling the truth. The average result was about 54 percent correct. Flipping a coin would give you 50.
The data also revealed an interesting bias: people are better at recognising truths than lies. When someone told the truth, judges usually got it right. When someone lied, judges caught the lie less than half the time. In other words, we tend to believe.
The truth-default
Researcher Tim Levine offered an explanation known as truth-default theory. According to the theory, most people automatically assume that what they are told is true unless there is a clear reason to suspect otherwise. That is not naivety: in everyday life the people around us really do tell the truth most of the time, so this default saves a lot of time and effort. The cost is that we struggle to catch the rare cases when someone does lie.
The myth of the averted gaze
Ask people how to spot a liar and many will say "they won't look you in the eye". A large international study from 2006 asked people in more than fifty countries about the signs of lying, and this was the most common belief almost everywhere. The problem is that research does not find it to be a reliable sign. Liars do not avert their gaze more than truth-tellers, and some liars make a point of holding eye contact precisely because they know it is expected.
The same goes for sweating, fidgeting or stammering. These are signs of stress, not of lying. An honest person who fears being disbelieved can look just as nervous as a liar.
So what does work?
Research suggests that what is said matters more than body language. True stories tend to contain concrete, checkable details that are hard to invent. Open, unexpected questions are harder on liars than on truth-tellers, because the liar has to make something up on the spot and stay consistent. Comparing a claim with facts you can verify helps far more than trying to "read" a face.
What this means for Bullshit
In Bullshit, our bluffing game, every round shows a true fact with a missing word. Each player writes a fake answer, and then everyone tries to pick the truth from all the answers. Because the lies are written, there is no body language, no eye contact and no sweating. What remains is exactly what the research says matters: content. Here are some tips that follow directly from it:
- Write a lie that sounds like a fact. The real answers in the game are usually surprising. A lie that is too sensible stands out because it is dull.
- Match the style. If the true answer is a number, write a number. If it is a place, write a place. An answer in a different style from all the others gives itself away.
- Be careful with jokes. A funny answer is fun, but most players will rule it out at once.
- When you guess, use knowledge, not gut feeling. What do you know about the topic? Which answer fits? That works better than guessing who looks like they wrote it.
- Remember the truth-default. Other players tend to believe. A good lie does not need to be brilliant; it just must not give anyone a reason to doubt it.
The bottom line
We are not built to detect lies, and that is fine: most of the time our tendency to believe helps us live and cooperate with others. But in a game, knowing this is an advantage. The other players will believe your lie more easily than you think.