A weird world, full of true stories · Every fact checkedSubscribe
WeirdBank

Stories · Psychology

Dunning-Kruger Effect: The Lemon Juice Robber Who Inspired It

The Dunning-Kruger effect is the idea that being bad at a skill can also make you bad at noticing it. It began with a 1995 bank robbery near Pittsburgh, and scientists still argue over how much of it is real.

What is the Dunning-Kruger effect?

The Dunning-Kruger effect is the idea that people who are bad at a skill are often also bad at noticing it. It is named after two psychologists, David Dunning and Justin Kruger. They described it in a 1999 scientific paper called "Unskilled and Unaware of It."

In the paper's words, people with little skill "suffer a dual burden." They make mistakes, and "their incompetence robs them of the ability to realize it."

The effect is about specific skills, not a type of person. The authors make "no claim" that someone weak in one area is weak in others. Dunning later wrote that "we are all poor performers at some things."

Who was the lemon juice robber?

On January 6, 1995, a Mellon Bank branch in Swissvale, a town near Pittsburgh, Pennsylvania, was robbed in broad daylight by two men, one of them with a handgun. Neither wore a mask.

The two men were McArthur Wheeler and his partner, Clifton Earl Johnson. Wheeler later said his partner had told him about a trick: rubbing lemon juice on your face would hide it from security cameras. Wheeler believed it.

Police said he even tested it. According to Sgt. Wally Long, the juice burned his face and eyes, so he had to squint. Wheeler then took a Polaroid photo of himself, and his face did not appear in it. "When the Polaroid didn't show him, he thought it worked," Long said. Detectives guessed the film was bad or the camera was pointed the wrong way.

How was the lemon juice robber caught?

The bank's cameras recorded the robbery clearly, but Wheeler stayed free for months. In April 1995, his picture from the security footage was shown on the 11 o'clock news in Pittsburgh. Less than an hour after the broadcast, he was arrested.

When detectives showed him the surveillance pictures, Wheeler was shocked. "But I wore the lemon juice. I wore the lemon juice," he told them, according to the Pittsburgh Post-Gazette.

Johnson testified against Wheeler and was sentenced to five years in federal prison. Wheeler was convicted in federal court of the Swissvale robbery.

What did the Dunning-Kruger study find?

David Dunning, a psychology professor at Cornell University, read about the case in the Pittsburgh Post-Gazette. With his graduate student, Justin Kruger, he set out to test whether being bad at something also makes you bad at noticing it.

They gave college students tests in logic, English grammar and humor. Then each student guessed how they had done compared with the others, as a percentile. A percentile is a ranking out of 100: being at the 62nd percentile means you did better than 62 percent of the group.

On the logic test, the students in the bottom quarter scored around the 12th percentile. They guessed their score was at about the 62nd percentile, and rated their logic ability at about the 68th.

The best scorers did the opposite and guessed a little too low.

One detail often gets lost: the low scorers did not think they were the best. They placed themselves a little above average, still below where the top scorers placed themselves.

Is the famous Dunning-Kruger graph real?

Online, the Dunning-Kruger effect is often shown as a mountain-shaped curve, with beginners' confidence sitting at the top of the peak. That curve is not in the 1999 paper.

The paper's charts are much simpler. Each one has four points, one for each quarter of the students, sorted by their test score. The charts show percentile rankings, not confidence. Each comes from one test taken at one time, so none follows people as they learn.

"For years, I've been teaching a fake graph," wrote Dennis G. Jerz of Seton Hill University in 2020.

Is the Dunning-Kruger effect real?

This is where scientists disagree. Several research teams argue that much of the classic pattern is a statistical artefact: a pattern that comes from how the numbers are handled, not from how people think.

Two things can create it. Most people rate themselves a little above average, which is called the better-than-average effect. And our guesses about our own scores are only loosely tied to our real scores, so when people are sorted by real score, the lowest group tends to have guessed too high and the highest group too low. This pull of extreme results back toward the middle is called regression to the mean.

Researchers have tested this with random numbers. When made-up data with no Dunning-Kruger effect built in are split into quarters, the famous pattern still appears. In 2022, a team including McIntosh and Della Sala concluded that the classic pattern is "a statistical regression artefact that tells us nothing much about metacognition" (metacognition means thinking about your own thinking).

Dunning disagrees. In a 2022 reply, he pointed to later studies that he says avoid the problem, and wrote: "the effect is still real; the quarrel is merely over what produces it." Even Gignac and Zajenkowski, who called the effect "(mostly) a statistical artefact" in 2020, allow that the effect "may be to some degree plausible for some skills," but say it may be "much smaller than reported previously."

We think the honest takeaway sits in the middle. People do often misjudge their own skill. But the least skilled are not the most confident, and part of the original pattern may come from the math. The useful lesson still holds: when you are new at something and feel sure, ask someone with experience to check your work. Feeling confident is not the same as being right.

Questions people ask

What is the Dunning-Kruger effect in simple terms?

It is the idea that people who are bad at a skill often cannot tell they are bad at it, so they rate themselves higher than they should. David Dunning and Justin Kruger described it in 1999.

Who was the lemon juice robber?

McArthur Wheeler, a man who robbed a bank near Pittsburgh in January 1995 without a mask. He believed lemon juice on his face would hide him from the cameras. His story opens the 1999 Dunning-Kruger paper.

Has the Dunning-Kruger effect been debunked?

Not fully. Some researchers say much of the classic pattern is a statistical artefact, and the popular mountain-shaped graph is not in the original study. Dunning says the effect is real and the debate is about what causes it.

Sources

Every fact in this story is backed by at least two of these. How we check facts.

  1. 'Lemon juice didn't work; robber jailed', Centre Daily Times, Sunday 7 January 1996, p. 7A
  2. Michael A. Fuoco, 'Trial & error', Pittsburgh Post-Gazette, 21 March 1996
  3. Kruger and Dunning, 'Unskilled and Unaware of It', Journal of Personality and Social Psychology, 1999
  4. Errol Morris, 'The Anosognosic's Dilemma: Something's Wrong but You'll Never Know What It Is', The New York Times, 20 June 2010
  5. 'Swissvale bank robbed', Pittsburgh Post-Gazette, Saturday 7 January 1995, p. B-4
  6. 'Robber is sentenced', Pittsburgh Post-Gazette, Saturday 28 October 1995, p. C-2
  7. Brendan Rodela, 'How to Catch A Criminal: Lemon Juice Can't Keep Crook Concealed', Officer.com, 11 March 2025
  8. McIntosh et al., 'Skill and self-knowledge', Royal Society Open Science, 2022
  9. Steven Novella, 'Misunderstanding Dunning-Kruger', NeuroLogica Blog, 8 January 2019
  10. Dennis G. Jerz, 'Fake Graph: The Actual "Dunning-Kruger Effect" Is NOTHING Like I Thought It Was', 7 April 2020
  11. ScienceAlert staff, 'What Is The Dunning-Kruger Effect? You Might Be Using It Wrong', 26 April 2022
  12. Gignac and Zajenkowski, 'The Dunning-Kruger effect is (mostly) a statistical artefact', Intelligence, 2020
  13. Jonathan Jarry, 'The Dunning-Kruger Effect Is Probably Not Real', McGill University Office for Science and Society, 17 December 2020
  14. Nuhfer et al., 'Random Number Simulations Reveal How Random Noise Affects...', Numeracy, 2016
  15. Nuhfer et al., 'How Random Noise and a Graphical Convention Subverted...', Numeracy, 2017
  16. David Dunning, 'The Dunning-Kruger effect and its discontents', The Psychologist, 7 March 2022
  17. David Dunning, 'We Are All Confident Idiots', Pacific Standard, 27 October 2014
  18. Improbable Research, Ig Nobel Prize winners list, 2000
Video transcript

A man robbed a bank with lemon juice on his face.

The robber was sure the juice made him invisible to cameras, so where did that idea come from?

And how did his arrest end up starting a famous psychology study?

It was nineteen ninety five, near Pittsburgh. The man was McArthur Wheeler. He walked into a bank in broad daylight with no mask at all. He later said his partner told him that lemon juice would hide his face from security cameras.

Wheeler had even tested it. The juice stung his eyes, so he squinted and took a Polaroid photo of himself. His face was not in the picture, so he believed it worked. Police guessed the film was bad, or the camera was pointed the wrong way.

The bank's cameras caught everything. Months later, his picture ran on the eleven o'clock news. Less than an hour after that, police found him. When they showed him the photos, he could not believe it. He said, "But I wore the lemon juice." And that strange arrest gave a professor an idea.

A psychology professor at Cornell University, David Dunning, read about the case. He wondered if being bad at something also makes you bad at noticing it. So with his student, Justin Kruger, he tested college students on logic, grammar and even humor. Then each student guessed how they did compared with everyone else.

On the logic test, the bottom quarter beat only about twelve percent of the others. They guessed they had beaten about sixty two percent. The best scorers did the opposite and guessed too low. It became known as the Dunning Kruger effect.

But the version you have probably seen online is not what the study found. That famous chart, with confidence climbing a mountain, is not in the study. Some scientists also say part of the effect is a math illusion. Dunning still says the effect is real, and the argument is about why it happens.

So how do you avoid becoming the lemon juice robber? We think it takes one habit. When you are new at something and feel sure, ask an expert to check. Feeling confident is not the same as being right.

The robber tested the lemon juice with a camera, but the one thing he could not check was himself.

More weird true stories

Psychology

Learned Helplessness Experiment: Why 6 Dogs Stopped Trying

The learned helplessness experiment is famous for dogs that lay down instead of escaping a shock they could easily get away from. It began at the University of Pennsylvania in the 1960s, where two graduate students, Martin Seligman and Steven Maier, saw it show up by accident in someone else's experiment.

November 16, 2026 · ▶ 1:58

Psychology

Kitty Genovese: The 38 Witnesses Story That Was Mostly Wrong

The murder of Kitty Genovese in 1964 made the bystander effect famous, after a newspaper said 38 neighbors watched and did nothing. Here is what really happened that night, what the experiments it inspired actually found, and what newer research says about whether people help.

November 14, 2026 · 5 min read · ▶ 1:45

Psychology

Universe 25 Experiment: Why a Mouse Paradise Died Out

The Universe 25 experiment gave mice endless food, water and shelter, yet the colony stopped having babies and died out. Here is what scientist John Calhoun saw from 1968 to 1973, and what later research says it does and does not mean for people.

November 13, 2026 · 5 min read · ▶ 1:42

Psychology

The Batman Effect: Why Kids Dressed as Heroes Kept Working Longer

The Batman effect is the name of a real psychology study in which young children who pretended to be a hardworking character, like Batman, stuck with a boring job for longer. Here is how the experiment worked, what it found, and what it cannot tell us.

November 12, 2026 · 5 min read · ▶ 1:43