Why Intelligent People Need More Confidence (And Stupid People Need Less)

The Dunning-Kruger effect states that overestimation of one’s abilities is a part of why incompetent people perform so poorly. The original study by Dunning and Kruger showed that, when measuring the ability to perform “tests of humor, grammar, and logic,” those in the 12th percentile estimated themselves to be in the 62nd. When the skills of these participants were improved, the participants had a more realistic evaluation of their abilities.

The problem is that those in the top quartile actually underestimated their abilities. We can see in this graph, which charts the results from the humor test, that our improvement in a skill increases at a slower rate than does our belief in how competent we are at that skill. There is a point in which, as we become better at a skill, we become more competent than we believe we are.

What this means, if we are to make a general claim from this specific piece of data, is that everyone considers him or herself to be in the 3rd quartile. In social psychology this phenomenon is called illusory superiority (also called the Lake Wobegone effect), and it has profound implications both as to why bad ideas propagate and why outstanding ideas have a hard time becoming universally accepted.

This is not solely a problem of stupidity overvaluing itself: all of us have a too narrow perception of competency. Again, speaking generally, we see the distance between competency and incompetency to be a 1/4th smaller than it actually is, since most of us imagine that there exists this fictional “bottom 60 percent” in which we are all superior to and a fictional “top 20 percent” which is superior to all of us.

Of course, this is not the only instance when our natural intuition with numbers is incorrect. The birthday paradox and the Monty Hall problem are two famous examples which exploit our false intuitions with probability, and our notion of the size of large numbers (a trillion is a million millions) is often incorrect.

One Million (1,000,000) in U.S. $100 bills

One Trillion (1,000,000,000,000) in U.S. $100 bills  (Man depicted on bottom left corner.)

A trillion is a much larger number than we imagine it to be. We know that anyone who wishes to have an accurate understanding of mathematics must remind him or herself of these false intuitions and look at the numbers objectively, but it is less obvious that anyone who wishes to have an accurate understanding of their cognitive abilities must remind him or herself of the Dunning-Kuger effect, especially if his or her evaluation happens to land them neatly in the 3rd quartile with everyone else.

If we cannot see past this cognitive bias, we do more than unfairly credit the untalented. We run the risk of slowing down progress, which relies upon our ability to cull the outstanding from the excellent, the excellent from the good, and the good from the average and lousy. (How can we do that if everything is just above average?) Not only that, but if you are highly skilled, you are likely to have inadequate confidence in your abilities, which will stifle your much needed ability to contribute, not to mention give you a headache.