Katy Milkman: The famous English author Neil Gaiman gave a memorable and much-discussed commencement speech at the University of the Arts in Philadelphia in 2015. At the end of the speech, he encouraged students to "make interesting mistakes, make amazing mistakes, make glorious and fantastic mistakes." The idea being that the way we learn best is by failing. But what if that idea is wrong?
In this episode, you'll hear about a pattern in the way we process information that may seem a bit counterintuitive, one that contradicts a lot of what we've been taught about failure. Northwestern University Professor Lauren Eskreis-Winkler joins me to discuss what we miss when we fail. And you'll hear from an artist who has become an expert at learning from his mistakes.
I'm Dr. Katy Milkman, and this is Choiceology, an original podcast from Charles Schwab. It's a show about the psychology and economics behind our decisions. We bring you true stories involving life-changing moments, and then we explore how they relate to the latest research in behavioral science. We do it all to help you make better judgements and avoid costly mistakes.
Joseph Herscher: So I had this idea. What if I had a machine that the moment you finished your dinner, it served you dessert? And so the whole machine starts with me eating dinner. And I take a sip of my drink. And when I place my drink back down, I place it on a little platform, and that kick-starts this chain reaction.
Katy Milkman: This is Joseph.
Joseph Herscher: Hi. My name is Joseph Herscher, and I am a kinetic artist and YouTuber.
Katy Milkman: Joseph makes elaborate chain-reaction machines that use household items in overly complicated ways to perform simple everyday tasks. You may know them as Rube Goldberg machines. You might have seen them in cartoons or movies like Wallace and Gromit or Back to the Future. Or maybe you've seen them in museums or airports. Joseph has been building these contraptions since he was a kid.
Joseph Herscher: I made my first machine when I was six, and it was a machine to store my candy. And obviously that was very useful for a six-year-old. But I also noticed that it made my parents smile. And that kind of combination of solving problems in my life with entertaining people around me, I really loved it.
Katy Milkman: That early success spurred more inventions.
Joseph Herscher: I kept making machines my whole childhood. I made a machine once so when my mom came home late from work at night, she opened the front door around midnight, and there was a string that got pulled that pressed play on a cassette player. And there was a message like, "Welcome home, Mom. I love you."
Katy Milkman: This whimsical approach continues to inform Joseph's designs, including his dessert-serving machine.
Joseph Herscher: So I take a sip of my orange juice, and when I place that down, it triggers this mechanism that actually tips the glass of orange juice over. And the liquid pours all out into a big tray, and the glass tumbles across the table. And these two things happen at the same time. So you've got all the liquid draining in this tray, the glass rolling down the table, and then the glass does this flip, falls off the table, lands perfectly upright, just as the liquid starts pouring through a spout and into the same glass.
So then that triggers this block of butter to slide down a metal track. And there's a candle underneath the track that heats up the track, and that makes the butter glide, because it melts it just enough. And then that leads to a visual gag where you see this huge hammer about to hit a computer, but at the last minute it's caught by a chain and doesn't quite hit it. And that triggers a tiny mini hammer to fall down and press a key on the keyboard. And that starts a Skype call, which calls my phone. The phone starts vibrating. It slides down a ramp. It falls off the ramp. It's hanging by its cable. And then a baby emerges on the scene. And of course babies can't resist phones that are ringing because of the sound and the noise, and they just love them.
Katy Milkman: This was the plan. The baby would grab the phone, which was connected by a cable to a laptop. The cable would pull the laptop off the table, which would smash and trigger the next part of the machine. But when Joseph and his team went to film the sequence, the baby didn't quite cooperate.
Joseph Herscher: He wasn't behaving in the way I assumed he would. So he was very trepidatious around the phone. He didn't think he was allowed to touch it. Or if he did, he just did it very slowly and kind of awkwardly, not the pacing that I wanted. And then he'd finally would touch it, but he wouldn't yank it hard enough to pull the laptop off the table.
Katy Milkman: If the baby didn't pull on the phone, then the machine wouldn't work. Joseph would have to rethink his approach to the project, a project he had been working on for months.
Joseph Herscher: So usually I have an idea or a concept, and I start playing and exploring and just seeing which ideas will work and which won't work. And then there's a long build period of up to two, three months. This is just for a two-minute-long machine. And then leading up to the filming for about four days, I'll just test it over and over and over again, each step, and make sure that it works most of the time. But when you've got them all in one long chain reaction, and you're trying to film it in one continuous take, without any cuts, it's very, very hard. It doesn't work every time. Most of my machines work one out of a 100 times, to be honest.
Katy Milkman: One out of 100 times. Not great odds, especially when you're relying on a baby.
Joseph Herscher: So we film them over and over again for a whole day, with the camera person like meticulously following the action. If they miss the action on the one take where it works, we'll be very upset. So it's a very high-pressure, high-stakes environment.
So at that point I was worried that my idea to use a baby was not a good idea and that it wasn't going to be possible. And I was almost going to rewrite that part, take it out, change it, because it seemed like it was too hard or that wasn't going to work. The baby started to cry at one point. They're babies. He was confused, or I don't know, and that really stressed me. I felt bad and I wasn't sure. I was like, "This isn't going to work." So then I just took a breather, thought about it. His mother was like, "He'll get this eventually. I'm telling you." She had more faith than me.
Katy Milkman: Joseph and his team took some time to rethink the problem.
Joseph Herscher: It's important you don't get too attached to your original ideas. You need to be able to bend with the wind. If something doesn't work the way I think it's going to work, that's just life, reality. When you're dealing with a problem, give yourself some play time, like a kid would play, without restrictions, without constraints, just two hours or four hours to just brainstorm and explore and not worry about whether it's going to work or not.
Katy Milkman: This approach led the team to a new idea. They thought about a way to make the phone more enticing for the baby.
Joseph Herscher: Someone, can't remember who, had the idea to put bells on the laptop so when he pulled the phone, it would pull the cable, which would pull the laptop a little bit and make it shake and the bells would ring. And so he liked that. And so we started training him like that. And we just did it slowly by building up. We started with just the bells. Then we added the laptop, which made it a bit harder for him to pull it but at that point he'd already learnt the bells could be triggered. So he eventually got that to work. And then we'd removed the bells finally, before the final take.
Katy Milkman: They set up the camera, and they reset the machine. It was time to try it all again. The pressure was on.
Joseph Herscher: I'm very nervous and stressed and anxious. And it's sort of a sense of impending dread because I just know how hard it's going to be. You've got that one day to get it. You've hired a camera person. You've baked 10 cakes already, and you've put a lot of work into this. And then you just don't know if you're going to get it or not. And relentless failure of take after take after take, where every single time something doesn't work, and it's every time it's something different that doesn't work. It's really infuriating, and you just have to keep trying and trying and trying until you get it. It usually takes between 50 and 150 takes to get a perfect run-through in one take.
Katy Milkman: Joseph had invested months of work into this machine. Now comes the moment of truth.
Finally, it worked. The baby pulled the phone. The laptop fell.
Joseph Herscher: And then that triggers another ball to roll. And then this giant wagon wheel rolls over my head, and there's a little hole in the wagon wheel that fits perfectly around my head, so I'm unscathed. Then that triggers 20 pots to fall off the wall and smash. And finally, a chandelier swings over my head as I duck to wipe my face with a napkin, which triggers the slice of cake to get placed onto my plate. And then I eat the cake.
Katy Milkman: After dozens of failures, Joseph finally succeeded, and it paid off.
Joseph Herscher: When you get that one beautiful take when everything just works, it's really an amazing feeling. It's like winning the lottery. And that's what gets you hooked. That's why I keep doing it. Despite all the failures, it's worth it for that one success. It's definitely one of my most popular videos. At the time it got a lot of views, especially on Facebook. I think like hundreds of millions of views on Facebook. And on YouTube, it's had about 30 million.
Katy Milkman: What those millions of people didn't see is the many times that different parts of this machine failed. But once in a while, Joseph puts his skills to the ultimate test. He runs his machines in front of a live audience where he can't hide failure.
Joseph Herscher: I did a live machine in Berlin for an event, and I had spent three weeks building this machine on site. And I went all the way around this huge shopping mall actually. And it went up three flights and then across through the air. These bicycle wheels rolled down this track, and then there was a zip line, and it was very big. And we'd spent three weeks on it, like 12 hours a day, testing it and testing it. On the day, there was a big audience, and it was extremely nerve wracking. And then we ran it for the audience. And one of the steps didn't work, in the middle of the machine. Something got stuck. This T-shirt kind of got stuck on the string that it was meant to glide past. I was sort of devastated because I'd spent three weeks on this, and it had all come down to this one moment where this T-shirt had got stuck on a string, and everyone's watching and people are like, "Oh, oh, it didn't work. It's not working." And people don't understand that these things don't work every time. So, they'll view it as a total failure.
Katy Milkman: But all was not lost.
Joseph Herscher: My friend was in the audience standing right by the T-shirt, as the T-shirt sort of glides over their heads. And I just was like, "Pull the string, pull the string." And she quickly pulled it, flicked it. And then the machine kept going. And it was quite funny. I don't know how many people noticed that she did that, and in the end the rest of it worked, and it was OK. But like that was a little bit disappointing, that kind of failure, because it's a public failure and it's at the end of a project. You don't have the chance to solve that problem. That's it. You're not going to run it again.
Katy Milkman: So it wasn't a catastrophic failure, but it still stung.
Joseph Herscher: When things don't work the way that I thought they would work, that can be a little bit demoralizing. But over the years I have a really good strategy that I've developed to deal with that, which is that I remind myself that sometimes things are not going to work, and that's OK.
Katy Milkman: Joseph has become an unusual type of expert.
Joseph Herscher: You could say I'm an expert at failing. You get more information; it's data, right? So when I'm trying to make something work, and things are not working, every single one of those failures is data for me, like how not to do it, basically. And it teaches me. And that knowledge is going to help me make a mental map of the world.
Katy Milkman: That's a really important point. Joseph's failures contain critical information that help him learn what it takes to build truly incredible machines.
Joseph Herscher: The alternative is that I live in a universe where I only do things that are really safe and that work all the time, and they're much less ambitious and less exciting. If you're going to do things that push the envelope and that are new, the only way to do that is to take risks, and not everything's going to work. Those failures are part of the process. Those failures go with the territory.
Katy Milkman: Joseph Herscher is a kinetic artist based in London and an expert at learning from failure. You can find a link to the video of his cake-serving machine and his YouTube channel, Joseph's Machines, in the show notes and at schwab.com/podcast.
Errors are a constant in Joseph's line of work. He's been building complex Rube Goldberg machines for years, and he's come to expect that certain elements won't work right away or that they'll fail at inopportune moments. He's had enough success, in the form of millions of online views of his creations, to work through the disappointment he feels when something doesn't turn out exactly the way he'd like. And as he mentioned, failures are chances to gather useful data.
Unfortunately, whether we realize it or not, most of us are not terribly adept at taking advantage of those learning opportunities. That's because failure can be uncomfortable. As a result, research shows we tend to ignore the often-valuable information that failure contains. This idea that we learn a lot by failing has been around for a long time. In 1897, an American theologian named Tryon Edwards wrote that "Some of the best lessons we ever learn, we learn from our mistakes and failures. The error of the past is the wisdom and success of the future."
This notion has been repeated by many well-known figures, from Albert Einstein to Neil Gaiman to Winston Churchill, but is it true? Society often celebrates failure as a teachable moment, but in several studies, Lauren Eskreis-Winkler, along with her collaborator Ayelet Fishbach, found that failure did the opposite. It undermined learning. Lauren joins me to talk about why failure often teaches us less than it could and certainly less than many famous figures have suggested.
Lauren Eskreis-Winkler is an assistant professor of management and organizations at the Kellogg School of Management at Northwestern University.
Hi, Lauren, thank you so much for joining me today. I really appreciate it.
Lauren Eskreis-Winkler: Thanks for having me, Katy.
Katy Milkman: So I'm excited to talk about your research showing that people learn less from their errors than from their successes. And I'm wondering if you could describe how you show this pattern in your work.
Lauren Eskreis-Winkler: So there's a few different research paradigms that we've used. One of the ones that I think is most compelling as a researcher is a simple multiple-choice test. So one sample we studied was a group of telemarketers, and we gave them a simple multiple-choice question. It said like, "How many Americans in the United States use this type of customer service?" And it was a simple two-answer choice. And people are guessing. They don't actually know what the right answer is, but they guess, and then either they receive success feedback that they got it correct or failure feedback that they got it incorrect.
And we've used this paradigm not just with telemarketers, but with online participants, with lots of different samples. And the key is that people are guessing an answer to a question, and afterwards they get feedback on whether they got it right or wrong. Now because they didn't know the answer to begin with, the feedback, that's the lesson, that's the information. And the reason I love this paradigm is because so often in the world, there really is different information in failure versus in success. So imagine the hair stylist who gets a compliment versus a criticism. You can imagine there actually are differences in the thoughtfulness and the intelligence of the customers who provide you with a criticism versus a compliment. So this paradigm is really useful because it 100% equates the information in success and failure. And what we do after giving feedback to telemarketers or people online or children or whoever it is in our studies is we ask if they learned from the feedback, and what we find consistently is that people learn much more from success than they do from failure.
Katy Milkman: So let me just make sure I sort of got this right in my head. You asked me a question. It's a question I couldn't possibly know the answer to, like how many people have ever had a dream about going to the moon or something. And the answers are 90% or 10% or something like that. And I guess. I get it wrong, and you tell me that was wrong. It was the other answer. Or I guess, and I get it right, and you say that was right. It was that answer. Either way, I now know the correct answer, but I only learn the correct answer, and I'm able to sort of regurgitate it at a higher rate when you're giving me that positive feedback? Is that right?
Lauren Eskreis-Winkler: Exactly. Yeah. We actually do find across studies that there is some learning from failure. So sometimes people are literally at chance level. It's as though they receive no information at all. But often they do learn from failure. It's just that they don't learn as much as they learn from success.
Katy Milkman: Got it. That's really helpful. It's interesting that sometimes they learn nothing, but good to know that we learn a little from failure. It's just that we're learning more from success.
Lauren Eskreis-Winkler: Yeah.
Katy Milkman: So why is that? Why is it that we generally seem to learn more when we get positive feedback? "Hey, you got that question right." And then it's ingrained in your head, and you can answer it when it comes up again. But when you got it wrong, and you're corrected, you don't learn as much.
Lauren Eskreis-Winkler: So there could be multiple reasons. One that we've explored extensively, and this is all research with Ayelet Fishbach, is that failure is really ego threatening. It makes people feel bad about themselves. And so you can imagine in any situation in which you fail, there really are two motives. One is to learn, to gain something from the experience. And the other is not to feel bad about yourself.
And so what we find often is that when there's these two competing motives, you might think when you're outside the situation that what's really going to trump the other is the desire to learn. But what we find over and over again is that's not how human beings are created. Even in relatively inconsequential failures, things that you think would not even be all that ego threatening, people choose to tune out and not pay attention and not look. And a prerequisite to learning is paying attention. And so just by nature of tuning out and not paying attention people basically make it impossible for them to learn.
Katy Milkman: That's really interesting. It makes me think of Carol Dweck's work on growth mindset, which of course is all about reframing failures as opportunities to grow as opposed to demonstrations of your fixed capacity. And, of course, when you have a growth mindset, lots of things turn out to be better. I'm wondering if that feels relevant to you, or if there's anything besides trying to reframe failures as growth opportunities that you think can reduce this bias.
Lauren Eskreis-Winkler: Yes. So I think that's totally relevant and would be incredibly useful. So if indeed the mechanism here, if the psychological process is that people are feeling very threatened by failure and not paying attention, then I think the key question is how do you get people to stop feeling threatened? And so one way is to change people's underlying belief system. And Carol Dweck has really been the pioneer in this sense in figuring out how to create lessons and how to get people to think differently about failure so it's not as threatening. A much simpler way to do this is something we did in our studies, which is, hey, if you don't want people to feel ego threatened, just remove the ego, which is to say that it's very hard for people to learn from their own failures because their own egos are involved, but we found that people are more than able to learn from others' failures.
Katy Milkman: So watching other people's experience teaches us more than failing does. How does it compare to succeeding ourselves?
Lauren Eskreis-Winkler: It's basically all the same. So learning from others' failures and others' success is equivalent to each other, as well as learning from one's own success, and learning from one's own failures is lower.
Katy Milkman: That's really fascinating. And disheartening because there's so much to learn from failure. I'm wondering if you've thought about particular consequential settings where you think it's important to be aware of this issue. The most obvious is of course in education, but I'm curious about others beyond that as well.
Lauren Eskreis-Winkler: Yes. In education, in medical settings. I mean failure unfortunately is everywhere. It's in relationships. To what degree are you learning from these experiences or not? And I think probably where it's most difficult and most problematic to not learn from failure is where it truly has a lot of information. And where the information in failure may sometimes even be different than the information in success. And yet this whole sort of body of useful tactics and strategies is potentially being overlooked.
Katy Milkman: And I really appreciate your point there about the different information we often get in the real world when we succeed and get a pat on the back. Or fail, and the experience offers enormous insights into what can go wrong and how to avoid similar pitfalls in the future. So there's this asymmetry where failure can be a better learning experience. So we're really missing out if we bury our heads and ignore opportunities to learn when we fail.
So, Lauren, if you're a manager, and you're in a situation where you have to deliver negative feedback, you have to let someone know that they've made a mistake, is there some particularly good way that you can do that? Is it best to sort of start with a positive, and then transition to the negative, and then end up with a positive, or what are the best practices there?
Lauren Eskreis-Winkler: That's a great question about practically the exact ways in which to deliver feedback in order to have people actually respond to failure in ways that are most productive. So, yes, I think to the extent that people really are tuning in and paying attention to success and tuning out from failure, starting with something that leads them to tune in and leads them to engage will probably lessen the degree to which they're disengaging. I think there's millions of different factors in the environment that can help people to engage in failure. And the underlying current here is that the degree to which you can get people to think less about themselves to make the feedback less personal, so it feels less threatening. That's going to facilitate engagement and having people learn.
Katy Milkman: Really interesting. OK. What do you do differently in your life now that you've learned all of this from your research? Is there anything you do differently?
Lauren Eskreis-Winkler: So I do think I've just realized how much more responsive people are to success and praising success. It doesn't arise everyone's defenses. It makes them very receptive to the information and able to learn from it. And so I think to the same extent, coaching kids, or honestly adults, and pointing out what people are doing right, I think they have a much easier time taking in that information than when they're told what they're doing wrong. And I think this is really Ayelet Fishbach's work on experts and showing that novices in any domain, whether it's kids or adults, no matter who, they have a really hard time learning from failure, and it takes true expertise to be able to take in that information and look at it and really learn from it.
Katy Milkman: I really love that advice and I think it also relates to work that's been done by Nick Epley at U Chicago showing that we don't give people enough compliments, and how good it makes us feel and how good it makes them feel. And this is like an amazing bonus benefit of the compliments we give. It's actually also a better way to teach people how to learn and grow. So maybe the key takeaway should just be like give compliments constantly, like 10 times as much as you think is appropriate.
Lauren Eskreis-Winkler: Totally.
Katy Milkman: What would you advise listeners to do differently now that they know about your work? So we've covered that they should compliment people more. Is there anything else that you think a person who's thinking about their investment portfolios, thinking about managing other folks at work, or thinking about managing a family, someone who's looking at those everyday decisions, what would their key takeaway be besides give more compliments?
Lauren Eskreis-Winkler: Yeah. So I think like sometimes just failure happens, and you can't just focus on a success. You really do need to address the failure. And so I guess in situations like that when you're trying to look at your own failures or look at other people's failures, again, what you want to do is try and minimize the ego threat. And so it doesn't feel as threatening. And I think there's different ways to do that. One is simply time. All of our studies, we're looking at the immediate effects of failure and the fact that people really don't learn from the failure experience itself. Luckily, most failures in life are not like that. It's like the rejection letter. It's still there. You can look at it a week later. You can look at it a month later. And so trying to look at failure when you are in a mindset to take it in, I think that's useful. So trying to change your situation in order to feel better and feel like you really can address the information. I think that can help.
There's also other research in psychology on psychological distancing, which is like, I talked about how to actually remove yourself from failure, which is to look at other people's failure. But there's also kind of these mental tricks you can do to think about your own experiences in the third person. And that just gives you a little bit of distance. It can be as simple as speaking to yourself in the third person. So I would talk to myself and say, "Lauren, what did you do or what happened there?" And instead of saying, "What did I do there? Or what did I cause to happen there?" And even those tricks can sort of distance you from yourself in this psychological way. And so I think any of these strategies can be really effective. Anything you can do to make that failure less personal and more just about information.
The last strategy which I alluded to, it's probably the hardest one, is like become an expert, become really good at something. And then when failures come up, first of all, you have a higher ratio of successes to failures. And also, I think that can just help focus you on domains in which you're probably more likely to learn from failure to think about what you already are expert in. People have strengths and weaknesses, and in areas of strength or areas in which you're more educated or knowledgeable, the research suggests those are areas in which you'd be most likely to tune in, pay attention, and actually be able to learn from failure.
Katy Milkman: That is so interesting. And I love that as a takeaway because hopefully all of our listeners are developing expertise in decision-making by tuning into this show. So as they become experts, they will learn more from their failures in this domain, I hope. Lauren, thank you so much for taking the time to chat with me today. It was wonderful to have you back on the show.
Lauren Eskreis-Winkler: Oh my gosh. Thank you so much, Katy. I always have time to talk to you.
Katy Milkman: Lauren Eskreis-Winkler is an assistant professor of management and organizations at the Kellogg School of Management at Northwestern University. You can find a link to her paper with Ayelet Fishbach called "Not Learning From Failure—the Greatest Failure of All" in the show notes and at schwab.com/podcast.
It's all too common to attribute good returns on an investment to skill but bad returns to market forces outside of our control. For tips on learning from failure when it comes to your portfolio, check out the recent Financial Decoder episode titled "How Can You Learn From Your Trading Mistakes?" You can find it at schwab.com/financialdecoder or wherever you get your podcasts.
When you make a decision, whether or not it turns out well, there's a lot of value in reflecting on what you learn, so you can do better in the future. The fact that we remember lessons accumulated from successes more readily than those drawn from failures is a clear bias. It prevents us from learning as much as possible from our experiences, and it relates to loss aversion, one of the most famous biases we've covered in this show. In general, losses loom larger than gains, so a failure stings your ego more than an equivalent success buoys it.
As Lauren explained, to learn from failure, you need to find ways to make yourself feel less threatened by it. A counterproductive way to do this might be to blame someone else for your failure or deny it happened in the first place. But research by psychologist Carol Dweck of Stanford University suggests one way to productively learn from failure may be by adopting a growth mindset. Teaching yourself to welcome failures, as opportunities to learn and grow, rather than treating them as diagnostic of you and your potential. Protecting other people's egos when you share feedback about failures is warranted as well. To help your kids, colleagues, and friends learn from their mistakes rather than choosing to ignore the lessons associated with bad outcomes, a bit of sugarcoating may go a long way.
You've been listening to Choiceology, an original podcast from Charles Schwab. If you've enjoyed the show, we'd be really grateful if you'd leave us a review on Apple Podcasts. You can also follow us for free in your favorite podcasting app. And if you want more of the kinds of insights we bring you on Choiceology about how to improve your decisions, you can order my book, How to Change, or sign up for my monthly newsletter, Milkman Delivers, at katymilkman.com/newsletter.
In two weeks, I’ll have the story of a tennis phenom who made huge sacrifices in her drive to be the next Venus Williams. And I'll speak with UCLA psychology professor Craig Fox about a quirk in the way we tend to estimate the odds of rare events.
I'm Dr. Katy Milkman. Talk to you soon.
Speaker 4: For important disclosures, see the show notes or visit schwab.com/podcast.
