See previous 74 interviews, including former secretary of state Condoleezza Rice, Senate Education Committee Chairman Lamar Alexander, former education secretary Arne Duncan, and researcher Kirabo Jackson. Full archive here.
To Ulrich Boser, there’s a huge topic missing in the prevailing debate about education policy: the science of teaching and learning. Boser — a senior fellow at the Center for American Progress, a progressive think tank — is the author of a new book, Learn Better, which walks through the research on how people can improve their knowledge and skills. Boser believes that both the education system at large and individuals trying to master new topics could benefit from these findings. Among other myths, Boser challenges the idea of learning styles — that some people are visual learners while others are auditory.
In a recent interview with The 74, Boser discusses the importance of direct instruction (e.g., lectures) for beginners, what he sees as the problems in teacher preparation programs, the role of knowledge in critical thinking, and how people are overconfident in their own abilities.
The interview has been lightly edited for clarity and length.
The 74: For how long did you work on the book?
Ulrich Boser: I forget when I really started working on it exactly. I remember being at Education Week becoming fascinated with some of these ideas about learning, and being like, oh, maybe there’s a book here.
At a high level I’ve become disappointed in the education debate because I feel like it’s very focused on vouchers vs. not vouchers, school boards vs. not school boards, charters vs. not charters. Many of us, I think, just ignore what happens in classrooms even though you can see these potentially really big gains for students there.
So that was sort of what your recent ThinkProgress piece argued. I gave that a lot of thought personally as a journalist. One thought I had was, “Well, the science of learning is not a policy issue, and, as an education reporter, I mostly focus on policy and politics.” Do you see it as a policy issue or do you think the point is that education reporters should expand beyond just the policy realm?
On one side, I do find it remarkable, as I pointed out in that piece, that there are education reporters who I’ve reached out to and they’re like, “Ahh, that’s actually something I don’t cover,” and that seems odd to me. It’s sort of like, “I cover health but actually what doctors do? I don't really care.” So I find that strange.
Isn’t the analogy like covering the practice of how doctors do heart surgery? Most health reporters don’t cover that.
No doubt there are health care reporters for, say, The Wall Street Journal and what they do is the ins and outs of health care policy. I think that is great. I think when you look at a lot of health reporters, actually what they do is some health care stuff, and then they do the 10 ways to lose weight. So there’s a lot more interest in how some of these studies get applied to your daily life.
I just feel like in education there isn’t enough of that. For instance, learning styles don’t exist — and yet the idea of it appears everywhere! It’s so crazy. So to my mind I think there are these hands-on things that reporters could do to treat parents as people who are invested in their kids’ education that’d be really simple.
Can you talk about myths about learning — and you already mentioned one, learning styles — that you discovered in your book? First, maybe elaborate on learning styles and why it’s a myth?
Yeah, I think the learning styles one is a favorite of people these days, and you see it often in education media, popular conversation. I got something from my own kid’s teacher about what learning style my kid was; it was a survey that came home and I had to do everything that I could not to be the jerk parent who sends research documents to their kid’s teacher.
You could have sent your book!
Yeah, I could have sent my book, I know. So these myths gain traction because there’s something intuitively to like about them. On one side, it is great to learn basketball by playing basketball; it’s great to learn art by doing art. If you want to learn analysis of literature, it’s great to read and write about books, and so there’s a sense that some things are best engaged in that way.
Then there’s also evidence around dual coding — learning things visually and auditorily — so we do have these two types of channels. But the fundamental learning styles idea — that you are a kinesthetic learner, so everything you learn, whether it’s math, or science, or basketball — everything should come at you kinesthetically is basically the heart of the learning styles idea, and it’s kind of wacky, for one. You shouldn’t learn statistics by running down the road. And yet there’s something appealing about it.
In terms of other myths, there are things like, right-brained learners are more creative. We at the Center for American Progress published something recently on this. I think there’s others that are more subtle. Highlighting text and re-reading is another one where there’s this big analysis by John Dunlosky that sort of shows really actually very little evidence for those practices helping you learn. I often work at a law school, and it’s incredible how pervasive highlighters are and yet there’s actually no good research evidence behind those.
One thing that seems to work to help people learn, you say, is testing. Some people would think testing — how can that be a form of learning? Can you explain that?
Recently, I was preparing for a speech and I kept going over my own notes and that’s basically like re-reading. There’s something comforting about it. But you’re much better off, and this bridges to the testing idea — you’re much better off putting away your notes and actively retrieving that information from your memory.
That’s really the same way that sort of low-stakes self-testing works. It’s not a yes-or-no question; it’s sort of self-explaining in a way and sort of goes to how long-term memory works. The more that we retrieve things from our memory, the more we’re able to remember them. That’s one.
Then two, when you do this kind of low-stakes self-quizzing, you’re also making connections. So if I ask, “What's the importance of 1776?” you start thinking of George Washington. And so there’s a lot of evidence behind self-testing as a way to improve memory.
Then there’s some research by Jeffrey Karpicke that shows that it’s better than some other learning tools that we associate with higher-order thinking. He’s done some studies comparing self-testing with concept maps [a graphic organization tool meant to help students make connections between ideas]. Concept maps generally are seen as a tool to promote that type of higher-order thinking, and it turns out that even on these types of measures, when you’re measuring sort of richer levels of thought and learning, that self-testing, this type of retrieval of facts, can work really well for that too.
So, like, high-stakes state standardized tests might not be very useful because you don’t get any feedback until after the school year’s over, if ever?
Yeah, so the type of testing here is very different than that type of testing.
Do the stakes matter per se? Is it worse if it’s higher stakes for the purposes of learning?
I think the purpose of learning is to have it low-stakes. You’re really just doing this to make sure, one, that you know what you know, and two, the more that you pull things out of your memory, that actually solidifies that learning itself.
So when we talk about tests, we talk about those state exams or LSATs or these other high-stakes exams that aren’t learning experiences, though potentially they could be. Really what we see is beneficial for learning are these different types of tests, and I think self-quizzing might be a better term to use.
A classroom exam, which I would say in most cases would be a moderate-stakes test if the teacher just gives you an exam. Then if they give you feedback on that, that actually seems like one of the most ideal things.
So we see that a lot and the evidence behind it is pretty good, though not fantastic. Like clickers in classrooms [where teachers ask questions of the class and everyone responds in real time using an electronic clicker] — it’s a way to activate prior knowledge, it’s a way for teachers to get a sense of who’s actually learning, and then it’s also a way for kids themselves to figure out if they know the material.
Let me ask you about knowledge. Some people make the argument that knowledge is somewhat obsolete now — you don’t really need to memorize facts because you can just Google it. Do you think knowledge is less important now?
So my usual response to this is, if I say something to you in German, you can Google those German words as much as you want, but unless you have a knowledge of those words, you’re still really not going to know what I’m talking about. So in my mind it’s just a great example of knowledge being a really key grounding and so even though you can get the German translation website up and running and get each of those words, unless you have some sense of what they mean, you’re not going to be able to get to this higher level of understanding. So “you can Google it” is not really true. Knowledge serves as a basis to these higher forms of learning.
And presumably if you lack knowledge when you Google something, you’ll have a hard time, one, sifting through the results, and two, determining what’s a reliable source or not.
Yeah, I mean, I think that the reliable-source thing has become a bigger issue with the fake-news debate, but I think it’s very easy to underestimate the power of knowledge. There’s not only reading about it, but you can read much faster if you have some knowledge over something. Agricultural Times is very hard for me to read because I simply don’t know enough. I’m not sure that publication exists, but I don’t know enough about it. But at The 74 or Education Week, where I’m pretty well grounded, I can engage pretty quickly. In a lot of these education debates we’ve set it up as this either/or. I think something in the middle is really where we should be.
Either/or between what two extremes?
There’s, like, knowledge vs. skills, and it’s a reasonable debate. Should we just have kids learning all the facts or should we be learning for these sorts of skills, and I think the answer is somewhere in the middle and probably will depend on the field as well.
Arguably they’re complementary. It’s not like there actually is necessarily a tension, but they work in tandem.
I would agree. And there is some interesting work being done on that. One study found that in classes where you’re teaching for this higher level of understanding, they found better recall of straight facts.
Staying on the knowledge point, it did strike me in your book that there was sometimes tension between that and some of the other things you were saying, like giving students more choices, like High Tech High School, experiential, hands-on learning. Do you see that as a tension?
I think my argument, to a degree, the bigger argument of the book, is that learning is this process. So things that might be very good at the beginning of the process aren’t necessarily great at the end of the process. And so knowledge is very important when you’re getting a grounding in an area, and I think the same thing then is true around student autonomy. I think the work of John Sweller shows that direct instruction is really important at the beginning of learning a new skill, but they themselves have found an “expertise-reversal effect.” So you want less direct instruction the more skill that you get in an area. I think you’re right, that had crossed my mind too, and I think that specific tension, if you think of learning as this process, allows us to answer that in a way that’s more meaningful.
It seems like there’s a phenomenon of the rich getting richer in terms of knowledge. And in some ways it’s sort of depressing if you want to be a beginner to a topic, if you’re like, I know nothing about it, and now this book is telling me, well, the way to learn is to already have a lot of background knowledge. If you don’t have someone who can give you that direct instruction, that guided instruction, if you’re a beginner at something, how do you start off?
That’s a good question. One thing that we’ve seen, this is work from Lindsey Richland at the University of Chicago, is that activating prior knowledge through pre-testing is a good way to leverage what little you might have.
I think it would depend on what you’re starting to learn. But just to make sure you’re fluent in those basics. The direct instruction is sort of the parallel to that. You don’t just want the knowledge, you also want the insider tips, is another way of thinking about it, to help you gain a skill, and that’s kind of what direct instruction is in a way. It’s not just that they’re throwing a volume of information at you; it’s also helping you to engage that information in a meaningful way.
It does seem like for a lot of people, intuitively, the idea of direct instruction, or lectures, as a beginner — it doesn’t make sense to them because we’ve all been in lectures that are deathly boring on topics where we’re beginners and we’re like, this doesn’t work for me, and that’s why we need experiential learning, to be more engaged.
I think the other thing is sort of, just think of some certain skill you want to gain, we want to do ourselves. So I have little kids and I have these little scooters, and I can never figure out how to break them down, so what I do is I go to YouTube and I find someone who’s like, “You have to touch this little button and push it up.” That’s direct instruction, right? They’re giving me a very specific bit, and it’s not actually just the knowledge, they’re actually walking me through it.
So the problems of lectures is actually less about direct instruction in my mind; it’s often just a knowledge issue. Information is being thrown out there, and this is more of an issue in higher ed today, where information is just being thrown out there by professors, you get it or not — eh, who cares? So making that information applicable to what you know is going to be really important and making it engaging so it’s actually relevant to you in some way.
So maybe there’s a tension between direct instruction and making it engaging. And maybe direct instruction works differently depending on the size of the class. Like, tutoring is direct instruction, it’s just a very small class size, whereas college lectures, which we all fell asleep in, is direct instruction with an extremely large class size.
Yeah, the other thing where I find direct instruction sort of frustrating is — so, I ride a motorcycle, and I remember when I first started learning, I wanted to get on the motorcycle, that’s why I was taking the class, and a lot of it was like, “OK, here are some very clear basics. Here’s where the brake is,” and I was like, “No, just let me on the thing.”
So I think that would be the tension that I would highlight a little bit more. Getting these kind of frameworks in place, they will help you, I think, further down the line, kind of prevents you from that fun of getting your hands dirty.
How do you think teacher preparation programs have done in terms of imparting the science of learning to teachers?
I don’t think they’ve done a really good job. I think at a very high level education does a great job of siloing. So we see this in schools, where the finance department tends not to speak to the academic departments, and I think we see this a lot in education schools. Some of the work that I highlight in the book is really sort of memory research. So, spacing effects [spacing out learning over time], testing effects, interleaving [learning several related skills one after the other] — a lot of that was done in the memory field, so it hasn’t really shifted over into the education field, and I think a lot of scholars in education schools see themselves more as sociologists and so aren’t excited to really dig into this area. There’s some research to support this from the National Council on Teacher Quality.
Can you explain what metacognition is and why it’s important?
Metacognition is thinking about thinking. So it’s important because at its core we’re aiming to teach kids understanding, and one way to understand is to reflect on your own understanding. We often overestimate our own abilities. We saw this, ironically, in our own survey of the public. People believe that they’re really great at identifying great teaching, but then you give them the quiz — so it’s basically a metacognitive problem.
One issue is simply that people are overconfident. Then, deeper than that, these questions about, do I understand why I’m understanding, how do I know what I know? What we’ve seen in a lot of work is these questions promote richer understanding.