Line of Inquiry: Ayelet Fishbach on Learning from Failure

Ayelet Fishbach: We are taught to learn from failure, to celebrate failure, to fail forward. Graduation speeches often talk about how much you should dare to fail and learn from your failures, and managers talk about the lessons that they personally had from failures. If you just listen to public speaking, you would think that we are pretty tuned in to failures. However, this is not the case. When we experience failures in our everyday life, the common response is to ignore the failure, and we basically don’t learn anything if we ignore the experience that we just had and the lesson that we just got.

To the extent that failures are being ignored, to the extent that we actually tune out other than tune in, then there is no learning whatsoever from failures. And when there is no learning from failures, that’s quite in contrast with the general impression that failures were teachable moments in our life. Most of the time when we failed, we just didn’t pay attention.

My collaborator, Lauren Eskreis-Winkler, and I, were developing a paradigm where people are asked a question and there are two possible answers. And the first time they see these questions, they are basically guessing, and they have equal chance of getting it right or wrong. We asked telemarketers, for example, questions about consumer behavior; and there are two possible answers, and so they are kind of guessing, and they either got it right or wrong.

On the next phase, we test them on what they have learned; and now we present the same question, or a variation of the same question, with the same two answers, and we see whether they got it right. Now, on this phase, on the test phase, they are supposed to know either way, OK. If you got it right, then you know the correct answer. If you got it wrong, then you also know the correct answer; it’s the other answer. What we found is that while people mostly remember the correct answer, they don’t seem to remember what was the wrong answer, which means that they cannot tell us what is the right answer on the test phase.

We find that with questions that we presented to telemarketers about their work, basically. We found it when we asked people about a language that they didn’t know: so we present a symbol in some ancient language and ask them whether this is, let’s say, a bird or a piece of furniture. They are guessing. They make their guess; they are either correct or incorrect, and later on we test them about their knowledge. “So, remind me, what is this symbol again?” Those that guessed correctly can repeat the correct answer. Whereas, those that guessed incorrectly, even though they had full information at the point where we corrected them, don’t seem to know what is the correct answer.

We thought maybe it was harder to learn from failures, maybe just doing the switch—“This is not the right answer, therefore the right answer has to be a different one”—is harder for people. It turned out not to be the case. It turned out that when people give the wrong answer, they don’t even remember the wrong answer that they’ve given, so they really tune out. They don’t pay attention. It’s the response that, “This is not for me; I’m not good at it.”

With more experiments, what we were able to see is that it’s really a matter of self-esteem. It just doesn’t feel good to fail, and so people tune out. How do we know it’s about self-esteem? Well, Lauren and I had another variation of our task, where people are learning from somebody else’s failure. What we found is that they can; and they’re doing it pretty well.

If you are watching another person making the incorrect guess on our task, you will know the correct answer. You will know that if this symbol is not a bird, this person was mistaken to say it’s a bird. It means that it’s a piece of furniture, OK. People do this inference when they observe someone else failing; they don’t do it for their own failures, which is a pretty strong clue that this is about self-esteem.

Should we stop talking about failure as a wonderful thing? Well, failure was never a wonderful thing, whether big or small. No one likes to fail. We should acknowledge that it’s hard to learn from failure. We should understand that by exposing ourselves to failures, well, we might not be giving ourselves the best chance to learn.

However, what we found in another line of research is that when you encourage people to learn from failure and, in particular, what we did is encouraging people to come up with advice that they would give others based on personal failures, it’s not intuitive; it’s not easy. It’s not where our mind goes. Often when we fail, our mind goes elsewhere. Basically, “Let’s not do that; let’s engage in something else.” But we can learn from failure if we are encouraged to do so. What we also find is that we can learn from others’ failures, which means that the failures should be shared, that by talking about our failures, by telling others, they might be able to extract lessons that are often harder for us to extract because our ego is involved.