The Dunning-Kruger-Deboer Effect

Isn’t it strange how every theory about intelligence just becomes a stick to beat other (hypothetical) stupid people over the head with? I mean, I can’t think of a better way to make me feel better about myself or about the world in general.

The Dunning-Kruger effect is one of those things. Its popular formulation is something like “stupid people aren’t smart enough to know how stupid they are”, which is kind of a nice shorthand way to slag other people off: Just invoke the Dunning-Kruger effect!

But this kind of misunderstands what Dunning-Kruger is about. It’s not primarily about intelligence, but about skill. Sure, intelligence comes into the mix somewhere (genuinely stupid people will probably have a harder time learning skills and applying metacognitive filter to the skills they have) but that’s a related but different topic.

The point is that going from novice to amateur at anything is easy. This is because skills level isn’t linear to practice. If it takes, say, 10 hours of practice to graduate from novice to amateur, a linear relationship would mean 10 more hours of practice to go from amateur to expert, and then 10 hours more to from expert to master (and then grand wizard, etc, etc).

Of course we all know that’s not how skills work. If the relationship between practice and mastery were plotted it would probably be some kind of inverse exponential curve. It takes a lot more practice to get from level 3 to level 4 than it does to get from level 0 to level 1. (Makers of MMOs get this. In Eve Online for instance there’s a very tangible diminishing return in skill levels. Training from level 0 to level 1 is quick but training from level 4 to level 5 can take months, while providing the same percentile boost in attribute. This makes a certain amount of intuitive sense to people, as we’re used to real world skill working that way.)

So if you’re the guy on the guitar who just learned Wonderwall, it may see like guitar is actually pretty darn easy! Which leads you to overestimate your own skill level, underestimate the amount of skill you need, and (critically) underestimate the skill level of other more skilled people.

Anyways, that’s the Dunning-Kruger effect, a sort of cognitive bias that gets in the way of proper metacognition.

What I’m proposing is the Dunning-Kruger-Deboer effect, which is when your lack of understand of the Dunning-Kruger effect leads you to apply the Dunning-Kruger effect incorrectly. Which is to say it’s a metacognative bias that gets in the way of proper metametacognition. I’d wildly speculate that about 90% of mentions of the Dunning-Kruger effect online are actual examples of the Dunning-Kruger-Deboer effect in action.

The only thing left now is for someone named, say, Jonathan Smith to come along and coin the Dunning-Kruger-Deboer-Smith effect which is when the author of a blog post creating a new Dunning-Kruger variant hasn’t himself quite understood the Dunning-Kruger effect and is subject to a metametacognitive bias which gets in the way of proper metametametacognitive bias.

Or maybe the Dunning-Kruger-Deboer-XKCD effect where somewhere in a hovertext XKCD did it first.