The Dunning-Kruger Effect

Did you hear the one about the bank robber—get this: 5 feet and 6 inches, 270 pounds—who thought that if he rubbed lemon juice on his face and walked into a bank, no one would recognize him?  That he would be completely invisible, even to surveillance cameras?  When arrested shortly after the surveillance tapes were played on the local news, the robber was incredulous.  “But I wore the juice!”

I came across this story in Errol Morris’s series in the NY Times titled "The Anosognosic’s Dilemma: Something’s Wrong But You’ll Never Know What It Is."  While the entire series is fascinating, what hooked me was the story about the bank robber.  In fact, it’s also the story that hooked Professor David Dunning of Cornell.  Dunning’s interest went something like this:  The robber was clearly too stupid to be a good bank robber.  Was it possible that he was also too stupid to know he was too stupid to be a good bank robber?  Enlisting his graduate student Justin Kruger, Dunning went on to research the matter and conduct studies.  The result was the paper “Unskilled and Unaware of It: How Difficulties of Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments.”  His research also resulted in what is frequently described as the Dunning-Kruger Effect: the incompetence to recognize one’s own incompetence.

Reading this, I immediately started thinking about a particular relative I have, as well as one or two students I’ve encountered, and even a few friends.  (The Dunning-Kruger Effect also fits into a recent segment on Jon Stewart’s The Daily Show about obesity, which you can watch here (about 9 minutes in)).  But mostly, the Dunning-Kruger Effect got me thinking about criminals and my days as a federal prosecutor.

So many of the criminals I prosecuted were, let’s say, not very bright.  They were so not bright that catching and convicting them was relatively easy. (These were mostly the ones that resulted in guilty pleas.)  One thing I always found troubling as a prosecutor was that, in a way, we were fooling the public.  We let the public think that law enforcement and prosecutors were doing a brilliant job of keeping the streets safe, or maintaining the integrity of the financial markets, or ferreting out public corruption.  Sure, this was true some of the time. But a lot of the time, we were arresting low-hanging fruit, the bad guys who were so naïve they didn’t realize they were leaving a trail behind them that practically glowed in the dark.  In short, in a world where there are smart criminals and not-so-smart criminals, we were disproportionately getting the latter. Are we comfortable with that kind of justice?  I don’t know.

Another place where the Dunning-Kruger Effect gets played out is probably police interrogations, especially defendants who offer false-exculpatories.  There’s a reason why Miranda doesn’t seem to have made a real difference in suspects’ willingness to talk to the police.  So many of the criminals we arrest are too stupid to know that they’re too stupid to fool their interrogators.  So here’s my thought for today: It would be nice to think we’re successfully locking up the bad guys.  But if we were honest, I think we’d have to concede that we’re mostly locking up the not-so-bright bad buys.  Right?