Definition
Have you ever walked a local hiking trail a few times and suddenly felt sure that climbing Everest couldn't be that hard? The Dunning-Kruger effect is a cognitive bias (a twist in how we think) in which beginners in a field rate their own ability much higher than it really is, while true experts tend to rate themselves lower. To notice your own gaps, you need the skill to judge that field properly. Beginners can't even tell what they don't know, and that is where this illusion comes from.
Why do beginners feel so sure of themselves?
Picture a new driver who just got a license and has been on the road for a few days. Once keeping to the lane and using the gas and brake feel natural, it's easy to think, 'Driving is way easier than I expected!'
But that driver hasn't yet faced skidding on ice, a heavy rainstorm at night, or a sudden surprise at a busy intersection. With no experience, there is no yardstick in their mind for what to watch out for or how risky things can get.
That's because the skill of grading your own performance comes from the same knowledge as the skill itself. When you lack the eyes to look back at yourself critically, you end up, ironically, fearless and filled with confidence that has no basis.
Why does deeper learning make people humble?
Once you keep studying seriously or start getting real-world experience, everything changes. You finally see how many people are far better than you, and you realize the sea of knowledge you still need to learn is as huge as a mountain.
At this stage, the vague confidence you started with melts away like snow. You realize that what you knew was only the tip of the iceberg, and a deep sense of 'I still have a long way to go' sets in.
This feeling of falling short doesn't come from losing skill. It's actually a great sign that your metacognition (the ability to check your own thinking) has grown well enough to see your limits objectively. Admitting that you don't know what you don't know is where real learning begins.
To be more precise: the internet meme vs. the original study
You've probably seen a roller-coaster-shaped graph online with labels like 'Peak of Ignorance' and 'Valley of Despair.' That curve is a famous internet meme (a popular metaphor) made to help people understand the idea. It is not the actual graph from the original 1999 paper by Professors Dunning and Kruger.
In the real study, people took tests of logic and grammar and then guessed their own scores. The bottom 25%, who actually landed in the teens in percentile rank, guessed they scored around the 60th percentile, above average. That was the biggest overestimate of all. The less skilled people were, the larger their error in judging themselves.
The true top performers, the top 25%, rated themselves in the 70s percentile, a bit lower than the truth. They figured, 'If it was this easy for me, it must have been easy for everyone.' However, some critics argue that much of this result can also be explained by statistical illusions (regression to the mean, and the tendency of everyone to see themselves as better than average), so the size of the effect is still debated today.
Common misconceptions
The Dunning-Kruger effect only happens to people with low intelligence.
It's a common psychological phenomenon that anyone can run into when first meeting a new field or unfamiliar knowledge, no matter how intelligent they are.
Where you meet it
A person makes a profit one week after starting to invest in stocks and believes they're an 'investing master.'
You learn a few basic phrases in a foreign language and decide, 'Learning a language is no big deal.'
In one sentence
Beginners feel confident because they lack the eyes to see their own gaps, and as they learn, they realize how much they don't know and become humble.