Episode Transcript

Dr. Frank Keil Q&A [1:58]

Eric Olsen: On today's episode, Dr. Frank Keil, professor of cognitive science and director of the cognition and development lab at Yale University and author of the book Wonder: Childhood and the Lifelong Love of Science, which has received stellar reviews from both nature and American scientists joins the podcast to talk about the academic superpower that is wonder, why generalist polymaths make for better specialists than specialists, and how to reignite wonder in our children if they've lost it along the way.

Frank, what's so great about wonder?

Dr. Frank Keil: Well, it's interesting because the word wonder's been used in different ways and different times over history. So I have to describe my sense, and I propose that it's an early emerging passion for learning about the workings of the world. We are wired for wonder and we do this through spontaneous exploration and discovery of everything around us, and it's extremely early. We see it in infants, we see it in toddlers and it's very important to see it as far more than dumbstruck awe. There's an archaic version of wonder, which is, oh, far out. And you're sort of not really doing anything other than being kind of overwhelmed. I see it as much more enabling and active and exploratory. And what it does, is allows you to go beyond surface impressions of reality, mere glimpses, and look for deeper causal structure. Much of the book is about discovering mechanisms that explain the world around us, and children have this.

Why our students stop asking questions [3:37]

Eric Olsen: Yeah, let's talk about how they have it and then they lose it. So our kids pop out when they're really young, they start asking a whole lot of questions. To the parents of toddlers, that can be overwhelming at some point, but you suggest it's so good. And the question is, why do so many of them eventually stop asking?

Dr. Frank Keil: Yeah. I mean, first of all, just to amplify that, it accelerates. In the fourth year of life, children just totally go nuts with why and how questions. Not just fact questions, but why and how questions. They're the most interesting ones. And some kids can ask well over a hundred why questions a day. It can exacerbate parents, but usually not if they have time to listen. Sometimes they ask why questions just to anno, but most of the time they're sincere and real. So then the question is why is it when they hit elementary school, the acts of wondering and questioning can go down to one or two a day or none a day. The drop is catastrophic and been observed all over the world by study after study for many years. It hits all social classes. It's not something about low SES groups or anything else like that. It's everywhere. And it seems to happen for a variety of reasons, but they converge in a very negative way.

One is that most adults, including teachers, have a misguided view of what kids minds are like. They see them as bundles of cognitive deficits, limited, having highly concrete thought that can't think in abstract ways. And so from that, devolved to teaching them facts and little factoids rather than their deeper, kind of relational structures, but every bit of recent research shows the opposite. So they're very good early on at forming abstractions and looking for deeper structure. In fact, perhaps sometimes more than they should. And so that's one thing. Secondly, high stakes testing in the schools is a very mixed game. If you have a large class and you have to test the kids, a fact based assessments is the easiest way to go and easiest, least controversial things to check.

A lot of us have served on national science boards recommending what STEM curriculums should look like, and it's all about deep conceptual understanding, causal mechanisms, but we send those out to the states and they implement them in tests and they become back as facts, like how far is it to the sun and so on. And that's not just because they're lazy, it's because it's really hard to come up with good assessments that get deeper structure that you can scale up to a whole school district or a whole state. And if you don't know much as a teacher, it's easy to do that. If you have classroom management issues, like you have 25, 30 kids in a class, it's very easy to set yes, no questions as opposed to asking to explain something and then monitoring explanation and dealing with it. And finally, there's a whole bunch of these.

Another thing is we do this horrible thing where we reward kids for spontaneous pleasure of learning. And when we do that, we turn play into work. So when we tell kids we're going to give you a star or a reward for knowing this and that, they say, "Well, it must not be very much fun if they're rewarding me." And this happens all over. The good news is it doesn't have to happen that way. There's always a few kids who just preserve this wonder. Probably because of parents are luck or a great teacher. And I'm convinced it's absolutely in everybody's potential. I've never met a case where you couldn't get it inspired. Sometimes it takes a little work, but I think you can do it.

Eric Olsen: Yeah. Let's talk about the get it because it seems apparent that we... Parents, society, cultural, test, traditional education system, we have the ability to stifle one sense of wonder. Can we teach it? Can we help instill it in our students? Instill it back into our students?

Dr. Frank Keil: Yeah. I definitely think we can. I would not use the word teach. I would use more like the word model and it's a subtle difference, but teaching sort of implies they don't know how to wonder. We have to teach them of that skill. They know how to wonder. We're the ones that crush it. So I think the most important thing we have to do is let them know that this is a terrific thing to do and that we're partners in wonder. We're not the authorities that know everything. We're senior partners cause we know more, but there's lots that we don't know. And so I think one of the strategies we can use on kids is at dinner, come up with a question that you don't know the answer to. One of my favorites is, so how come humans need to eat fruits and veggies to get vitamin C, but dogs don't?

That's a really interesting question. It has all sorts of details, sort of consequences. Or why are electric cars so much quieter than combustion cars? This is another one. Why do leaves fall off trees so neatly in the fall and what controls their kind of precision? Any question you want, you can ask that can open up discovery, but you should tell the kid, I don't know all this. And that's what makes it really exciting. Ignorance is not a curse, it's a blessing. It means there's all sorts of neat stuff to discover, but requires a different kind of mindset. And we all can do it.

I teach a senior seminar at Yale. The seniors last semester at Yale, a lot of them become kind of jaundiced about science and they have to present each week. Something in the basic hard sciences, they know nothing about, explain it without any overheads or slides, just usually their words and show how they've come to understand it and convince the rest of the class of useful insight. And it is an eyeopening experience to them because they realized they didn't think they could do it. And it's amazing what they come up with and how we all learn together.

One other example. I think this is like getting better lenses on the world. When you learn more science, it's incredibly rewarding because you see things in technicolor before they were sort of blurry gray. I had cataracts a couple of years ago that I had removed in both eyes and I couldn't believe how the world popped. Well, that's what causal understanding does. You can see the underlying machinery and everything looks different. We've had a whole bunch of presentations last spring, on spring. The phenomena of spring. How early plants come up, how they go through changes, bird song. Gosh, how bird song emerges, and brain nuclei explode. It's a fascinating stretch. So when I listen to a bird sing, it's a totally different experience than before I had that kind of knowledge and it's richer. And that's what I try to convince people. And parents need to model that for kids and they should enjoy it genuinely. And the kids know how to do that and how to reawaken. That help answer it?

The academic research on wonder [9:48]

Eric Olsen: It did. And it just restated the gravity of the situation and how important it is for us to solve this. I know there is so much exciting research coming out of this space. What's some of the most exciting research that made you want to write this book, Wonder: Childhood and the Lifelong Love of Science?

Dr. Frank Keil: Well, there's a whole bunch. It's hard to say one particular study, but I'll mention a couple. There's a now classic study, done by Brandy Frazier, Henry Wellman, and Susan Gelman at Michigan where it's a really fun study. You show kids pictures that are anomalous, just something doesn't make sense like a turtle sitting in a bird's nest in a tree. And that prompts in the kid, a why or how question. Why is that turtle sitting up there? How'd that turtle get up there? Then they have different conditions and one condition they respond back, "I don't know," or, "It is." And that's a terrible response. Or they give an empty circular response. The kid says, "Why is that turtle in that nest?" And he says, "Yes, it is things that have shells and crawl who would like to sit in the structures in trees." Or they give a mechanistic response. "Well, there was a certain kind of bug and turtles can climb trees better than you think," or someone put.

And if you're a three year old and you hear the non-mechanistic response, you keep asking why. You're not satisfied with that. They know that they've been kind of cheated. And so they're sensitive to the value of mechanism. And there's been a lot of other studies that show those too. Younger kids are remarkably precocious in some of these respects. Susan Hespos and Lance Rips at Northwestern has shown that pre-verbal infants know something about fluid dynamics versus solid dynamics. If I show you a beaker full of fluid... Sorry, a beaker full of something and I tilt it and the stuff inside tilts... No, it doesn't tilt, stays level to the horizon. As I tilt it, they know it's a liquid. If it tilts with the beaker, they know it's a solid. And they immediately know then, you can push something through it versus you can't and they make all sorts of other inductions about it.

We did a study a few years ago, that still blows my mind, showing that 11 month olds know that only agents with intentions and goals can create order out of disorder. So if I show a bunch of objects all randomly scattered, a screen comes up and either a little agent comes behind it or a random bowling ball. They're totally shocked that goes from disorder to order, but they're not surprised if it goes from order to disorder. Either the bowling ball or other. So they're very sensitive to... And now you'd state that and it sounds impossibly complicated, but they know. It's almost like knowing what things can do to local reverse entropy. And we have no idea how they do it, but it's very clear. It's been widely replicated.

So these kinds of amazing talents are what we seize upon. And as they get older, they get good at what we call the division of cognitive labor. You heard about Adam Smith and the division of labor. Well, this is a division of cognitive labor. They learn who knows what around them and how to data mine. And a lot of it again, knows what to prefer. So we've shown that they know that people who know mechanism are your most important people to consult, not people who know silly facts and they understand that mechanism acknowledge has a special power. It generalizes more, it allows more inductions and so on. That gives you some sense of the studies.

Eric Olsen: It's beautiful and exciting. For the 11 month olds, was the goal to try to notice their wonder, their shock? Were we looking at the expression of the face?

Dr. Frank Keil: Yes, yes we did. We looked at what they call looking time. So they just looked a lot longer. Actually, when you measure facial expressions... I don't do much infant research because it's kind of frustrating. They don't go whoa, double takes. That's what you want them to do. But instead, they just kind look a little bit longer and we have these coders show it, but they definitely show a strongly significant difference. But yeah, in essence, they're surprised, but they don't show it the same.

Eric Olsen: It's fascinating. So fully convinced, we need to collectively figure out a way to not only not stifle in future generations, but reignite this sense of wonder that is just inherent and natural in kids. Is there any downside to wonder? Are our wanderers sometimes too inclined to believe in the fantastical over the Occam's razor reasoning of a scenario?

Dr. Frank Keil: I don't think so. I think this gets into different interpretations of wonder might be. Historically, wonders often had a bad name because it's seen as challenging authority. It's seen as... I mean, Adam and Eve were wondering about knowledge and look what happened to them.

Saint Augustine has a terrible thing about wonders being an evil kind of... It's the ultimate hubris. It's pride because how dare you question the received authority of God or whatever authority you think there is. So I think there's some versions of wonder where it seems like you're questioning dogma and that's terrible, but that's not the way I see it. I see it as you should never be discouraged from asking why and how. There's no way that should be a bad thing. And as you embrace that, I don't think it leads to problems. Occam's razor and logic is certainly a part of science, but science is far more than math and logic. It's also discovering the geography of causal terrain.

If you'd look at scientists in the lab and many people who have done this, Kevin Dunbar is an authority on this system and many studies. You'll see that they spend a lot of time saying, "What is out there? What's the regularities out there? How do we map these?" And I think it's a lot like explorers discovering new terrain. We do think quasi logically in the lab. But if you go to lab meetings, even the best things, they're not just writing equations and doing a deductive calculus, they're trying to brainstorm about possibilities. They're wondering. They're saying, "Could it be this mechanism?" Look at the current debates about the immune system or about cosmology. I have a friend who's a very eminent cosmologist and he says, "I'm just blown away by how much more vast the universe is than I thought it was a year ago based on the new web results. And so now I'm changing everything I think. How could this work? How can the dark energy..." And so on. So he's not running his equations. He's very gifted mathematically, but that's not where he starts.

Should we be trying to raise specialists or polymaths? [15:41]

Eric Olsen: It's funny when you talked about that and how wonder leads you to so much more than math and science. It is funny how in my mind I go to, oh, so wonder creates generalists out of specialists. And I'm thinking of the pros and cons here. So let's say many of our parents listening have a specific career goal in mind for their kids. Maybe it's their own kids' goal or one they're projecting. Let's say engineering for the sake of argument. Talk about how we might want to have a different potential goal for parents to get excited about and to point for their students to become polymaths.

Dr. Frank Keil: This is complicated. First of all, it's very dangerous to try to dictate your goals to your children. I mean, I have great sons who've all done wonderful things, but none of them did what I think I would've hoped they do at first. Now, I'm much glad that they do what they did. But having said that, I think you do have a burden as a parent to expose the children to a wide variety of opportunities to make sure they don't miss something. Suppose you think your kid doesn't appreciate engineering and engineering's a marvelous place to start. You're not sure you want them to be an engineer, but you want them had thought about it and appreciate what it is. What would you do?

I would wonder with them about daily things. I'd gesture to bridges. I'd look at a bridge on a small town that crosses a brook and how it's simply at one beam. Then, look at the bridge that crosses the highway into New Haven and notice how it's a suspension structure. And they'd say, "Why don't they just scale up the beam?" Well, it has to do with the way volume is a cube and mass overwhelms the support structure. You'd have to change the whole architecture. So statics, the structures, and support, and volumetric versus kind of area calculations, are a fascinating on train to engineering. I know that detracts for you what I'm trying to talk about, but you can just get that to kids very, very quickly.

Another engineering case would be... I wrote some of these down because I was trying to think about... Well, you just can come up with examples that are remarkable in the world. How did they solve this problem and how did it work? So I think that's the way I would get it. I would stick to something you generally don't know all about. Even something simple, how refrigerators work? It gets tricky. I mean, you can get into Carnot engines and all sorts of complicated kind of subtleties, but you don't have to. Now, comment about math. I'm not anti math at all. I think math is incredibly wonderful as a way of posing more questions and mysteries. It doesn't give the answers. I mean, when I first encountered special relativity in terms of algebra and saw how it fell out so beautifully and simply. I was blown away, but that doesn't mean you can't teach science and STEM subjects initially with a small amount of math and then show them how math makes you articulate and pose things in a much more deeper way. I hope that helps.

Eric Olsen: Yeah. That is one of the questions that I was thinking about and you kind of teased it already. This concept that I would think that wonder is easier in the arts, that quantitative subjects are just hard science, just the facts, no room for mystery. You spoke about how that came to life for you. And so you do think that wonder can be correlative in STEM subjects.

Dr. Frank Keil: Yeah, I think actually there's no evidence that we wonder more in the arts. I mean, arts are often factoid things. You learn different styles of music theory or art history or something like that. I don't know any major scientist who isn't filled with wonder. Feynman, one of the most quantitative of all physicists was if you read his quotes I described in the book, it's all about wonder. Some people talk about the end of science and I guess they're humanists. They say, "Oh, we've done all the big discoveries and now we just have a few little..." That's just not true. The mysteries are getting more and more complex and more and more interesting. That may happen someday, but I don't see it as even close. And like I say, I think you want to get kids into math and introduce them to math because it sheds enormous light.

Sadly, I don't think math has taught that well in most cases. Most kids can do much more than they think. I like to teach kids math is a system of structures rather than procedures. It was such an eye opener to me to realize that differential equations could be thought as families with different kind of structures, than as simply procedures you crank through, when I was in college. And so I think how you approach math and teach it can make a big difference and insinuate what it can do. It's not giving you an answer, it's giving you a tool.

Eric Olsen: Yeah. And maybe I'm leading the witness at this point, but I feel like you've hit at this at a bunch of different angles today. But just to reinstate it, even if I have a specific career goal for my kid or my student does themselves and they want to be a computer engineer. And so in their head, they want to get good at Python or different computer languages. When I see them getting curious about all these other things, it might appear to be distractions from their goal. Remind us why that's not the case.

Dr. Frank Keil: No, I mean, it's complicated. Sometimes you do have to focus and knuckle down and learn the code. But it's more often than not, you can get enormous insights from analogies. You can get insights into chemical structures from analogies to other systems, the solar system and the atomic theory. I actually write in the book about other cases, which we should do more of. I'm fascinated by camouflage and deception in the immune system and how you could use these wonderful examples of camouflage and deception and literature as kind of different role models for it. I do discuss somewhat about the red queen hypothesis and how it was derived from Lewis Carroll. So there is for cross frivolousation and I'm all for learning about different disciplines. I also think, especially in the hard sciences, the physical sciences, the more you're really deep dive into one or two sciences, the more the rest starts to come easier and easier.

I had the great fortune of attending MIT as an undergraduate where the first two years was a vast, common shared curriculum. And we just all spoke a common language after two years. Having all had physics in biology and advanced math and stuff that made us... We were a little weird sometimes. We talked about people as random or the entropy in this room is... We had weird conversations on occasion, but we also saw things more quickly sometimes in those. And it's not hard. We need more of a larger common ground. Not because we want to be all technocrats, but because it allows us to talk about more things.

Eric Olsen: It's really fascinating, Frank. Finally, leave us with some next steps advice for parents and educators looking to reignite, looking to encourage wonder in our students.

‍Dr. Frank Keil: I'd like to say it's trivial. It isn't because we have to remind ourselves of what to do occasionally. One thing is to when you talk with your children, don't ask closed end questions. Ask open ended ones. So if your child went to the zoo, don't say, "Were there zebras there?" Don't say, "Did you have a good time? What did you eat?" Ask them... You might start with a fact question, like where were the hippos? They were all in this water. Why are they in water? And immediately convert it to a why, how question. And then the kid will start talking more. In a conversation, you don't want to be the dominant one talking, you want to be more of a listener and expander. Draw them out. So simple focusing on what you're asking and you might even keep a diary or put your iPhone down and record your conversation and say, "Did I just ask a whole bunch of yes, no questions or did I really probe?"

I have several grandchildren and I've noticed... Well, now it's three. But when I talked to the oldest one who's now three and a half, just the other day, she did an adventure. And I realized I'm not working hard enough to get to the why and how questions. I'm asking simply. If you kind of have a lot going on, you'd just say, "Was it fun?" "Yes." Okay, end of conversation. That's the wrong question to ask. So that's one thing, again and again, I say I should model wonder. Show how fun it is to yourself. I, as a matter of practice, try to learn something new that shakes me up at least once a month. If I haven't, I'm not doing my work. And boy, I have been shaking myself up like crazy. I mean, so many things I've learned about cosmology, about biology, about the immune system, about electromagnetic theory, whatever it is. And it's all accessible to us.

The internet is an amazing resource. And the more you start to understand things, the more you can tell fake news from real news because the more you start to see these common kind of ways of structuring good science. So I think we need to walk the walk. We can't watch garbage on TV and expect our kids to be sophisticated wonders. Now, most people listen to this podcast already know that, but they should monitor what they say to their kids. They should teach science through history. Historical narratives are remarkably compelling for kids. I had the best science teacher in my life who until the eighth grade, probably taught me more science than they ever knew any other way. By talking about the evolution of the steam engine, starting back with Chinese firecrackers and storing of chemical energy and then going through all.

And he did that for every part of science. And you might think that would take forever, but it didn't. It was actually what James Conant suggested for Harvard undergraduates, teach science through history because it makes it relevant, it makes it appealing. These were people who didn't know hardly anything and it took off. And so I think learning a little about the history of science can be very evocative in young children because it humanizes it in a way that's interesting. And you have to simplify sometimes because it goes in so many tangents, but even failures are interesting. I mean, the Phlogiston theory of fire. Is fire a stuff or an event? A lot of kids think of it as an object, when it's really just an event. And that's a wonderful insight to see them have, as well as many adults. Like rainbows, rainbows are not objects. They're events, but who thinks of them that way?

Eric Olsen: Frank, your own self wonder is palpable and it's so troubling that you've had so little success as an academic and a professional because of it.

Dr. Frank Keil: I mean, I have to say I don't think I did as much. I only regret as I'm old as I am, I wish I'd done more of this when I was earlier. I was too worried about getting those papers out. I should have wondered more because it's so rewarding. You talk to these people, these great polymaths of history, when you're old and you're near the end of your life, it's the thing that gets you the most joy, not how many cars you have or how big your house is. It's learning more. See the world clearly more. You see these people in their death bed still asking questions. Well, what about this? I think that's what I want to be in my death bed.

Enroll in AoPS Academy’s Math and Language Arts Courses This Academic Year [26:07]

Eric Olsen: So how do we help our kids to wonder? Well, we might be able to help. Our academic year advanced math and language arts classes at AoPS Academy are the perfect way to make sure your student is maxing out their wonder potential. Whether you're near our 12 physical learning centers across the country, or want the flexibility of learning online from our AoPS Academy virtual campus, our after school and weekend courses are a flexible way to make sure your students are getting the deep problem solving skill set they deserve.

Visit AoPS Academy Math & Language Arts Courses today to learn more and find a time that works for your family.

Dr. Frank Keil Rapid Fire [27:01]

Eric Olsen: It's now time for our rapid fire segment called Problem Solved where we ask the guest to solve incredibly complex and difficult education issues in single soundbites. Frank, what's one thing about K-12 education you wish you could snap your fingers and problem solved, it's fixed?

Dr. Frank Keil: Give teachers a chance to talk to kids and listen to them back in conversations. If I could expand that a tiny bit, have teachers talk about their children after classes as kind of like, well, how can we make this kid be smarter or work better or know more? This is almost like doctors adjusting. Every kid can be improved and in some school systems they do this. They meet as teams at the end of the day, discuss every child. What they've done today or how they've met targets or not met them. That would be great.

Eric Olsen: If you could go back and give your kid-self advice on their educational journey, what would it be?

‍Dr. Frank Keil: When you get a great teacher, suck them dry. You use them every way you can because one great teacher can carry you through a lot of bad ones. Not get discouraged. And I don't mean bad. I don't want to blame teachers. One of my sons taught in Teacher for America for a couple years in one of the toughest school districts in the world. And he said it was the hardest thing has ever done. He has a PhD and MD and he said it was nothing compared to that. So I think hats off to teachers. They're amazingly dedicated and the burnout problem is severe. But if you do get someone who's gifted and amazing and isn't that way, really I should have tended them more. I think I did it quite a bit, but I could have done it more.

Eric Olsen: What part of education do you think or hope looks the most different 10 years from now?

Dr. Frank Keil: I hope we ennoble the art of teaching more. I hope we appreciate how hard it is to do well. And we get people who are super talented and love science rather than afraid of science, even at the elementary school levels teaching, because we need them and they should do it. It's a form of national service, which would be extraordinarily value to our country.

Eric Olsen: And what’s your best advice for parents looking to raise future problem solvers? 

‍Dr. Frank Keil: Don't do one thing and have them start to do another. If you want them to be brilliant, creative, wondering souls, be one yourself and share it with them. It's a shared project. It's not something where you're imparting information to them. You're both doing it together and appreciate what they can bring to the table.

Eric Olsen: Such great stuff, Frank. Thanks so much for joining us today.

Dr. Frank Keil: Thank you, too.

Episode Summary & Conclusion [29:20]

‍Eric Olsen: Is there a better person to write a book on wonder? It was so fun to watch Dr. Keil's brain ping pong and connect to new ideas in this conversation. His curiosity about everything made him such a powerful player in the field he chose to specialize in. It was a great reminder that world class people are often great at a lot of things. A great reminder to help steer our kids toward, not away from interests. And may you continue your journey alongside us raising the great problem solvers of the next generation.