Dunning-Kruger Effect

What the Dunning-Kruger effect means

The Dunning-Kruger effect is the cognitive bias where people with limited knowledge or skill in a domain significantly overestimate their own competence. At the same time, genuine experts tend to underestimate theirs. The gap between what people think they know and what they actually know is widest when they know the least.

The effect was formally described in 1999 by psychologists David Dunning and Justin Kruger at Cornell University. Their original study asked participants to rate their own ability in logic, grammar, and humour - and found that those who scored lowest were the most likely to dramatically overrate their performance. Those who scored highest tended to slightly underrate theirs.

This happens because competence and the ability to assess competence require the same underlying knowledge. If you don’t know much about epidemiology, you also don’t know enough to recognise what you’re missing. The gaps in your knowledge are invisible to you. Meanwhile, an expert is acutely aware of everything they still don’t know, which makes them more hesitant. Real expertise can also intensify complexity bias - the more you know, the more nuances you can add, and the harder simple answers start to feel - which is one reason why experts and novices can talk past each other so completely.

The effect doesn’t mean uninformed people are stupid. It means they’re operating with an incomplete map and don’t yet realise how much territory the map is missing.

How the Dunning-Kruger effect works

The Dunning-Kruger effect is often visualised as a curve that maps confidence against actual knowledge, and the shape of that curve is revealing.

The peak of “Mount Stupid”

When someone first encounters a new subject - reads an article, watches a documentary, sits through a single lecture - they experience a rapid spike in confidence. They’ve gone from knowing nothing to knowing something, and that something feels like a lot. This is sometimes called the peak of “Mount Stupid”: maximum confidence built on minimum understanding. The mirror image, once expertise sets in, is the curse of knowledge - the expert can no longer imagine what it was like not to see what they now see, which makes it harder to teach, explain, or even talk to people earlier on the curve.

At this point, the subject looks simpler than it is. The nuances, exceptions, contradictions, and unsolved problems haven’t come into view yet. The person has learned enough to form an opinion, but not enough to know how much their opinion is missing.

The valley of despair

As knowledge deepens, confidence collapses. This is the painful stage where you start to see the real complexity of the subject - the things you didn’t know you didn’t know. A medical student in their third year knows vastly more than someone who read an article online, but feels vastly less confident, because they can now see how much there is still to learn.

This valley is where many people give up. The discomfort of realising how little you know - what psychologists call “conscious incompetence” - creates cognitive dissonance that’s hard to sit with. It’s tempting to retreat to the comfortable confidence of the peak, or to avoid the subject entirely.

The slow climb of genuine expertise

Beyond the valley, confidence gradually returns - but it’s a different kind of confidence. It’s calibrated, hedged, and aware of its own limits. An expert might say “based on the current evidence, it seems likely that…” where a novice would say “it’s obvious that…” The expert isn’t less sure because they know less. They’re less sure because they know more.

This is closely related to circle of competence - the idea that genuine expertise means knowing the boundaries of what you know, not just the contents. The Dunning-Kruger effect is what happens when someone can’t see those boundaries at all.

The Dunning-Kruger effect in the real world

The Dunning-Kruger effect on social media

Social media has amplified the Dunning-Kruger effect dramatically. Someone who watches a few videos about a complex topic can feel sufficiently informed to argue with specialists. During the pandemic, this played out at enormous scale - people with no medical training felt confident enough to dismiss consensus medical advice, sometimes with fatal consequences.

The structure of social media rewards confidence and punishes nuance. A bold, simple claim gets shared. A carefully qualified one doesn’t. This means the most visible voices on any given topic are often the ones experiencing the peak of Mount Stupid, while the people who know the most are being cautious in ways that algorithms don’t amplify.

Confirmation bias makes this worse. Once someone has formed a confident opinion on limited knowledge, they’ll selectively seek out information that supports it and dismiss information that challenges it. The Dunning-Kruger effect gets them to the peak. Confirmation bias keeps them there.

The Dunning-Kruger effect in the workplace

In professional settings, the Dunning-Kruger effect shows up in meetings where the person with the strongest opinion is often the one who has engaged least deeply with the problem. The people closest to the work are more likely to hedge, qualify, and express uncertainty - which paradoxically makes them less persuasive.

This creates a dangerous dynamic. The confident generalist sounds like they know what they’re doing. The cautious specialist sounds like they don’t. Organisations that reward confidence over calibration end up making decisions based on the worst-informed opinions in the room, because those opinions sound the most decisive.

It’s compounded by authority bias - our tendency to defer to people who project confidence and status. A senior person who speaks with certainty about a domain they don’t deeply understand can override the quieter expertise of someone more junior who knows the territory far better.

The Dunning-Kruger effect in politics and public debate

In political discourse, the Dunning-Kruger effect helps explain why complex policy issues get reduced to simple slogans. Trade policy, immigration, healthcare systems, climate science - these are areas where genuine expertise takes years to develop, but where everyone feels qualified to have a strong opinion after reading a headline.

The effect also explains a particular kind of frustration: experts who can’t understand why the public doesn’t listen to them. But the public isn’t being wilfully ignorant. They genuinely don’t know what they don’t know. The complexity that’s obvious to the expert is invisible to the novice. This isn’t a moral failing - it’s a structural feature of how knowledge works. It’s closely related to naive realism - our deep-seated assumption that we see the world as it is, and that anyone who sees it differently must be uninformed or biased.

What the Dunning-Kruger effect doesn’t mean

It’s worth being clear about what this effect isn’t, because it’s frequently misused.

The Dunning-Kruger effect is not a way to dismiss anyone who disagrees with you. Saying “that’s just Dunning-Kruger” to someone who holds a different view is itself a form of motivated reasoning - using a psychological concept as a weapon rather than a lens. It applies to everyone, including (especially) the person confidently deploying it.

It also doesn’t mean that experts are always right and non-experts are always wrong. Experts can be captured by groupthink, blinded by paradigms, or simply mistaken. The Dunning-Kruger effect is about the relationship between knowledge and confidence calibration - not about who gets to have an opinion.

The ancient insight here is much older than the 1999 paper. Socrates built his entire philosophical method around the idea that wisdom begins with recognising how little you know. The Dunning-Kruger effect is the scientific confirmation of something the wisest people have always understood: the more you learn, the more you realise you have left to learn. And the moment you feel certain you’ve mastered something, that certainty itself is worth questioning.