Play like Dunning and Kruger aren’t watching.Eric Weinstein
Success is best enjoyed when it over-delivers. You thought you were a decent guitar player. Until you get complimented for your skills by a master of the craft. Unfortunately, the same can’t be said about failure. Few things hurt more than realising how terrible we are after thinking we nailed a task. Imagine seeing yourself as a guitar god. Only to be put in place by a 10-year-old. The kid shows you what your six string can really do and lets you know: Your cringe-inducing experience has a name. It’s known as the Dunning-Kruger Effect.
Much has been written about it. For what it’s worth, here’s my take on the often misunderstood phenomenon. Just bear in mind, this is not a science paper. We’ll explore what the Dunning-Kruger Effect is, whether it’s real and how — if at all — we can avoid it. But we’ll also look at it from a pop culture and philosophical perspective and see how it ties in with other ideas about knowing and not knowing.
What Is the Dunning-Kruger Effect?
The Dunning-Kruger Effect is a cognitive bias we fall for when we overestimate our abilities in a given domain. This seems particularly true for people who are not very good at a task. Roughly speaking, the less skilful people are, the worse they are at recognising their lack of skill.
The term goes back to the 1999 paper Unskilled and Unaware of It by Justin Kruger and David Dunning. In a series of experiments, the psychologists tested participants on their ability to spot humour, reason logically and recognise correct grammar. Then they compared the results to people’s self-assessments of their skills. This is where they discovered the counterintuitive relationship between confidence and competence.
Appraising humour, for example, required a subtle understanding of other people’s tastes. Some participants, however, turned out to be rather unskilled at spotting a good joke (as determined by professional comedians). It was the same people who perceived themselves significantly better at knowing “what’s funny” in relation to their peers. The better a participant’s test score was, though, the more realistic their estimation turned out to be.
The researchers concluded that those who overestimated their own abilities suffered from “a dual burden: Not only do these people reach erroneous conclusions and make unfortunate choices, but their incompetence robs them of the metacognitive ability to realize it.” In other words, they lack the ability to think critically about the knowledge and skills they didn’t know existed.
Beyond Kruger and Dunning
Truth be told, there are doubts about whether the Dunning-Kruger Effect is real. While it certainly rings true, it has (kinda) failed to replicate in later studies. That is to say, the effect was there but not as significant as the original paper suggested. There’s a good chance it may just be a statistical fluke. Still, the Dunning-Kruger Effect has a strong foothold in psychology. Since its inception, it has been studied in a variety of other contexts.
In a previous newsletter, I wrote about the Bullshit Blindspot, a phenomenon from the scientific study of bullshit. According to Canadian researchers, it’s those who consider themselves experts in detecting bullshit who are most likely to fall for it. The psychologists likened the effect to the infamous findings of Dunning and Kruger. Then there was the question if the Dunning-Kruger Effect also applies to physical skills.
A study from 2021 found that a similar effect occurred when the task was physical rather than cognitive: “The worst performers,” (in a grip strength task) “were the most miscalibrated and significantly overestimated themselves.” Researchers also suggested there’s more than just a “metacognitive deficit” at play when it comes to assessing your own motor skills. The problem may be motivational.
What the Dunning-Kruger Effect is Not
Real or not, the effect’s popularity beyond the academic world is undeniable. Ironically enough, it’s often been misinterpreted and misused. So it’s important to note what the Dunning-Kruger Effect is not.
For one, the effect is not a measure of arrogance or how confident people are. The least competent people were found to be significantly less competent than they think they were. They were “unskilled and unaware of it”. Not unskilled and cocky about it. This is why referring to the Dunning-Kruger Effect tends to backfire when used as an insult for people we perceive as incompetent or arrogant.
By the same token, the Dunning-Kruger Effect is not synonymous with stupidity. It’s not an appropriate label for fellow humans we consider rather simple-minded because it’s not a condition some people are forever burdened with and others are free of. On the contrary. If real, the cognitive bias would be a flaw we’re all susceptible to. Particularly when we venture into the uncharted territory of knowledge.
Escaping the Dunning-Kruger Effect
The idea of a universal bias corresponds to the remedies mentioned in Dunning and Kruger’s original paper: “Paradoxically, improving the skills of participants, and thus increasing their metacognitive competence, helped them recognize the limitations of their abilities.” This solution is reminiscent of Aristotle’s observation that with increased knowledge we discover more of our ignorance. Who would have that learning enables critical thinking?
Lastly, let’s consider the other end of the spectrum, which is the underestimation of our skills. Because the Dunning-Kruger Effect can also be considered the opposite of impostor syndrome: No matter how much we’ve accomplished and how much recognition we got, we perceive ourselves as incompetent frauds who do not deserve the accolades. We may have escaped Dunning and Kruger, only to end up skilled and unaware of it.
With this dichotomy in mind, let’s take a closer look at the relationship between competence and confidence.
Four Stages of Not the Dunning-Kruger Effect
I’m sure you noticed. We’re veering off the path of psychological statistics and are traversing into philosophical territory. We do so by plotting the journey of someone falling prey to their own ignorance and eventually recovering from it.
The following visualisation is based on an unofficial illustration that was featured on the Dunning-Kruger Effect’s Wikipedia page for a while. Let’s see how it corresponds to other theories of knowing and not knowing. Imagine we’re asked to assess our chess skills.
Our journey starts rather upbeat. With a steep yet swift climb to Mount Ignorance. We played chess as a kid. And we were pretty good at it. But it’s one of those games that can be learned relatively quickly but takes a lifetime to master. When it comes to our chess skills, our confidence level is high. Unfortunately, that can’t be said for our level of skill. We rarely manage to roll a six on the dice.
The idea of Mount Ignorance is not a new concept. Back in the 1960s, Martin M. Broadwell came up with four levels of teaching also dubbed the four stages of competence. Unconscious incompetence was the first psychological stage he described. At this level, we’re steeped in unknown unknowns, something Dunning called meta-ignorance in a 2011 follow-up paper. Our knowledge and skills are superficial and only seem sufficient.
That’s the insidious thing about Mount Ignorance. What’s more, if we start to get comfortable up there, we can end up worshipping a Cargo Cult. Trapped forever in a futile effort of mimicking the behaviours of the successful. Hoping that the results will one day manifest themselves. Oblivious to the myriad of false beliefs and lack of background knowledge that renders our efforts meaningless.
As we progress on the x-axis towards a more enlightened self, we soon reach a tipping point. We suddenly realise: We’re not as smart as we thought when it comes to playing chess. Our confidence level is in freefall. It drops as steeply as it has gone up. The fact that we’re now free of our ignorance and smarter than before is poor consolation. Eventually, we reach Cringe Valley. It’s the place where we reflect on our past uneducated selves. With a sense of cringe and a feeling of embarrassment.
In terms of the Four Stages of Competence, we’ve now reached the stage of conscious incompetence. We’re becoming aware of how much we don’t know about chess. Who is Gary Kasparov and since when can a pawn move like that? We’re ready to explore the known unknowns, slowly expanding our knowledge horizon. We’re still devastated, though. Our confidence won’t recover until we’re attempting the climb of humility.
The Climb of Humility
The Climb of Humility to regain our confidence is gradual. Slowly but surely, our confidence level increases again. Alongside our level of skill. Only this time, both have a strong footing in reality. This distinguishes it from our initial climb to Mount Ignorance. Our knowledge and metacognition have evolved enough to be acutely aware of our lack of skill. Perhaps we’re reading up on the history of chess. We study Gary Kasparov and Deep Blue. We learn about strategy and ELO levels. We discover castling tactics and en passant.
As we go up the slope of humility, we gain conscious competence. We now know how good we are at chess. Most likely because we applied what we learned and it worked. Perhaps we’re using a learning method such as the Feynman Technique. It prompts us to teach our new skills to others, which in turn exposes any knowledge gaps we might still have. We’re still far away from our initial confidence level. But we don’t mind.
As we progress further, our confidence level reaches a plateau. Our level of skill continues to grow, albeit more and more slowly. We now have a very detailed and differentiated idea of playing chess. We not only know how to win a game. We’ve also acquired negative knowledge about how not to lose. What’s more, we have a good sense of what else there is to know in order to progress even further.
This final stage corresponds to the idea of unconscious competence. We no longer need to make deliberate decisions, we’re instinctively good at a task. The only downside is: At the level of mastery, we’re so good at what we do, we may even be cursed with knowledge. That is we don’t even remember what it was like not to know how to play chess well and find it hard to relate to such people. But hey, look at how far we’ve come.
A Different Relationship to Failure
When people overestimate their expertise and underestimate the chances of them being wrong, the results are often cringe-worthy. I dare say this is true at every level of skill; from beginner to expert. Even if what you got wrong is merely an aspect of a field you thought you mastered a long time ago. In any case, it’s a feeling we’d rather avoid. I for one wouldn’t be keen on finding out that my understanding of the Dunning-Kruger Effect has been as off as our guitar player’s self-assessment from the beginning.
But the thing is, it beats staying ignorant. As philosopher Peter Boghossian put it: “Self-confidence will always be illusory if the activity does not correspond to reality. And the further from reality the activity, the more one must buttress oneself from challenges.” In other words, to meaningfully progress at anything, we need to take feedback from reality. And that includes having your confidence shattered on occasion.
As far as I can tell, to “play like Dunning and Kruger aren’t watching” means to be willing to take risks and fail. To override self-consciousness and the fear of being judged by others because our expectations didn’t match reality. It means to be willing to cringe yourself to success if you will.
What’s needed seems to be humble confidence in a learning process fuelled by failure. So we better get comfortable with the idea of looking like a fool. It’s a sign that we’re onto something. The cringe may feel like the end. But it’s only the beginning.