
In 1995, a man named McArthur Wheeler robbed two banks in Pittsburgh in broad daylight. He made no attempt to disguise himself. He walked in, handed over notes, and walked out, apparently convinced that no one would be able to identify him. When the police showed him the surveillance footage, he looked genuinely confused. “But I wore the juice,” he said.
Wheeler had rubbed lemon juice on his face before the robberies, under the belief that it would make him invisible to cameras. The logic: lemon juice is used as invisible ink. Therefore, a face covered in lemon juice should not show up on film. The logic was completely wrong. But Wheeler was not aware that it was wrong. He was, in fact, confident.
This story reached David Dunning, a psychologist at Cornell University, who found in it a question worth investigating. How was it that a man could be so wrong about something and have no idea that he was wrong? And was this unusual, or was it in fact the normal condition of the human brain?
In 1999, Dunning and his graduate student Justin Kruger published a study that tested the self-assessments of Cornell undergraduates across logical reasoning, grammar, and humour. Their finding was simple and uncomfortable: people who performed worst on the tests rated their own performance highest. Bottom-quartile students consistently believed they had scored above average. They were not lying. They genuinely believed it. And the reason they believed it, Dunning and Kruger argued, is that the same skills required to perform a task well are the skills required to recognise when you have performed it badly. If you lack the skills, you also lack the ability to notice that you lack them.
The finding won the Ig Nobel Prize in 2000, an award for research that first makes you laugh and then makes you think. It entered popular culture under the name the Dunning-Kruger effect, and it has been misunderstood ever since in a way that is itself a small illustration of the phenomenon it describes.
The popular version of the Dunning-Kruger effect is about other people. The incompetent who think they are competent. The amateur who argues with the expert. The person in the meeting who speaks most confidently about the thing they understand least. In this telling, the Dunning-Kruger effect is a category error that happens to someone else, and recognising it in them is a mark of your own clarity.
This is not what the research shows.
The research shows that the effect applies to everyone, including experts, including the people who study it, and including, explicitly, David Dunning himself. “We are all Dunning-Kruger at some point in our lives,” Dunning has said in interviews. The bias does not attach to a type of person. It attaches to domains of knowledge. An expert cardiologist may be spectacularly overconfident about their driving ability, their parenting, or their understanding of economics. Expertise in one area does not transfer to accurate self-assessment in another. The overconfidence follows the ignorance wherever the ignorance goes, and the ignorance follows all of us into every domain we know less well than we think.
The second thing the research shows is that high performers are not exempt. Dunning and Kruger found that top-quartile students consistently underestimated their performance relative to others. They scored genuinely well but assumed that if the task felt manageable to them, it probably felt manageable to most people. The curse of knowledge runs in both directions: experts underestimate how much they know relative to beginners, and beginners overestimate how much they know relative to experts. The calibration is off at both ends of the spectrum.
The third finding is the one in this article’s title, and it is the hardest one to accept. Learning about the Dunning-Kruger effect does not protect you from it. Studies that informed participants about cognitive biases before testing their self-assessments found no significant reduction in overconfidence. Knowing the name of the trap does not help you avoid stepping into it. The bias operates below the level at which conscious information can intervene. You can believe, with genuine sincerity, that you are now more calibrated than you were before reading this article. That belief may itself be the bias in action.
What actually helps, according to the research, is specific and targeted feedback from sources outside yourself. Not general awareness that you might be wrong, but concrete, detailed, external evidence of where your assessments diverged from reality. The person who reviews their predictions against outcomes. The student who goes back to the test and identifies exactly which questions they were confident about and got wrong. The professional who asks for direct criticism rather than general impressions. The intervention is not self-reflection. It is external calibration against real data.
David Dunning was once asked whether he himself is subject to the effect that bears his name. He said yes, immediately and without hesitation. “The first rule of the Dunning-Kruger club,” he has said, “is that you don’t know you’re a member.”
You are probably a member. So am I. The difference between people who handle this well and people who do not is not awareness. It is the habits they build around getting honest, specific feedback from the world, rather than relying on their own sense of how things went.
That is the only exit from the club that the research has found.
Now you know. The question is what you do with it.
Marco Bellinari
Want to understand yourself a little better? These books will help.
Sources
- Nautilus. Why We Think We Know More Than We Do. David Dunning interviewed on the effect, its misunderstandings, and self-application.
- BizNews. Too Confident to Know You Are Wrong: Unpacking the Dunning-Kruger Effect. Cornell 1999 study, Ig Nobel Prize, real-world implications.
- Fact Myth. The Dunning-Kruger Effect. McArthur Wheeler bank robbery origin story, spectrum from prosopagnosia to expertise.







