The Fallibilty of Holmes

20/01/2013 15:45 GMT | Updated 20/03/2013 09:12 GMT

Here's what I love about Sherlock Holmes: he makes mistakes. And then - and this is the kicker -he admits to them. Not only to himself, but to his faithful sidekick, Dr. Watson. As he remarks to the doctor when the latter wonders why they hadn't gotten on a case (that of Silver Blaze, to be precise) sooner, the answer is simple, really: "Because I made a blunder, my dear Watson - which is, I am afraid, a more common occurrence than anyone would think who only knew me through your memoirs."

It seems odd, to focus on someone's fallibility as an asset. Indeed, Holmes's particular faults leave quite a body count in their wake - Mr. Hilton Cubit, in the Adventure of the Dancing Men, Mr. John Openshaw in The Five Orange Pips, to name but a few. But in his willingness to admit his errors when he does make them lies perhaps the greatest secret to Holmes's continued success.

Without mistakes, we can't learn. In something known as the reward prediction error, or RPE, our brain compares what actually happens in our environment to our expectations of what would happen. If there is a mismatch, bingo: dopamine is released into the brain and we learn. The error can be a positive one - the so-called reward exceeds our expectations - or a negative one - it falls below our expectations. In either case, there is a discrepancy between prediction and experience, and learning takes place. No RPE, no dopamine, no learning.

But that's our brain. We, on the other hand, don't like to admit it when we're wrong. Others, sure. They make mistakes all the time. But me? Falter? Absolutely not. That reluctance to accept error is the basis of one of the most stable, classic findings in psychology: cognitive dissonance, our ability to hold seemingly conflicting beliefs at the exact same time. When there is a mismatch between expectation and reality, we adjust either the reality or our beliefs to reduce the resulting conflict in our minds. And usually, the belief is the easier thing to change. We can't well adjust an event that has already happened, but we can certainly adjust our view of it. I thought the world would end and it didn't? I wasn't wrong. The evidence was just misleading. I knew what would happen all along, now that I think about it. It's self-justification gone wild.

When we self-justify, we don't learn. We've, in a sense, erased the possibility of the RPE from the get-go. How can there be an inconsistency between outcome and expectation if we've just adjusted the expectation to match the observed outcome?

But Holmes is different. Holmes makes mistakes, just like everyone else. In that, he is far from being some superhuman paragon of perfection. What is superhuman about him - or at least, uncommon - is his willingness to admit to those mistakes. Take, for example, his behavior in The Yellow Face, one of those rare instances where his initial theory of the case ends up being totally off base. Instead of shrugging it off - I knew what was going on all along - he bows his head and acknowledges his error to Watson. "Watson", he tells him, "if it should ever strike you that I am getting a little overconfident in my powers, kindly whisper 'Norbury' in my ear, and I shall be infinitely obliged to you." How easy it would have been to sweep the mistake under the carpet. But then, Holmes wouldn't be Holmes.

Being wrong isn't what keeps us from learning. It's admitting to being that's the culprit. So perhaps, the next time we want to say I was right all along, we'd do better to pause for a moment and whisper 'Norbury.' We may find ourselves surprised at the result.

Maria Konnikova is the author of Mastermind: How to Think Like Sherlock Holmes, out now published by Canongate.