American Scientist

With various streams of voluntary and involuntary information sources, it can be challenging to determine if our thoughts are truly our own. Tools like generative AI are raising deep concern among educators, as these tools are taking the "hard" part of critical thinking away from students.  In this episode of Wired for This, University of Virginia professor of psychology Dan Willingham shares decades of his knowledge and research on learning, memory, and his most recent work in applying cognitive psychology to education from kindergarten through undergraduate levels.

TRANSCRIPT

[Music: Wandering by Nat Keefe FADE IN]

[Celia]

Welcome to Wired for This—a deep dive into how we think, believe, change, and connect.
In this limited series, we explore the psychology of human behavior and neuroscience—what drives us forward, what holds us back, and how we navigate a world bursting with noise, contradiction, and complexity.
Today, we’ll hear from Dr. Dan Willingham, a professor of psychology at the University of Virginia. He’s taught there since 1992, and until about 2000, studied the neural basis of learning and memory. But today, all of his research concerns the application of cognitive psychology to kindergarten through undergraduate (K-16) education. He’s the author of several books, including the best-selling Why Don't Students Like School?, and most recently, Outsmart Your Brain. His writing on education has appeared in twenty-three languages. In 2017, he was appointed by President Barack Obama to serve as a Member of the National Board for Education Sciences.
I learned so much from our conversation, and I can’t wait to share it with you. From American Scientist, I’m Celia Ford, and you’re listening to Wired for This.

[Celia]

Before we get any deeper, can you break down what self-direction or independence means across different stages of education? I’m sure that learning how to be an independent critical thinker looks different if you’re in preschool versus middle school versus high school.

[Daniel Willingham]

It does, but there is a common thread. I think there are two main things that self-direction might mean. One is that you, the learner, select the goal of what it is that you’re wanting to learn about, or what problem it is that you seek to work on.
The second thing that might mean is that, given you have a particular goal, a particular problem you want to solve or a thing you want to learn, you figure out on your own how to go about learning it.
I’ll note right at the outset—both of those are pretty rare in most schools. Typically it is not the students who are selecting the goals of what it is they want to learn. Once the goal is set, the teacher tells them. "This is a standard problem in mathematics. Here’s the standard algorithm to solve it." Or, "Today we’re going to learn a five-paragraph essay. Here’s the format for writing a five-paragraph essay."
Now, that sounds pretty bad. It sounds like a straightforward criticism. Isn’t it self-evident that students should be selecting goals? And there’s something to that.
But I want to point out—when you don’t know very much, and you try and do either of those things—select the goal of what you’re going to learn or try to figure out how to solve the problem—you’re usually not very good at it. You spend a lot of time spinning your wheels.
When you think about being a teacher—you’re a third-grade teacher. You have all these things that you want to introduce students to, books you want to read, and so on. The idea that you’re going to set a goal for them, perhaps, and then not give them a lot of guidance about how to solve it—what do you think is going to happen?
The likely thing is, if you’re really just giving them no guidance at all, they’re very quickly going to get discouraged, and not a whole lot is going to happen. The alternative is that you want to coax them along and give a little bit of guidance, a little push, a little bit of information at just the right time.
That’s difficult to do when you’re tutoring one student. If you have 25 students who come to you with different background knowledge, knowing different things, different aptitudes, and comfort with numbers or whatever it is, then it becomes clear that you have 25 students progressing at 25 different paces. It’s enormously challenging. This is why it’s the case that most of the time, students are given a lot of direction. There’s not a lot of self-direction in classrooms.

[Celia]

So, what does it take for students to transition from being dependent on their teacher’s guidance, to having more intellectual independence?

[Dan]

I think a lot of it is stuff that is teachable, and it’s not very skill-like. It’s more—you learn a bag of tricks. When I’m confronted with a problem and I have no idea how to address it, here are some things I can try. Those are, again, I think highly teachable, and they’re not things students get a ton of practice in exercising what they’ve learned when it comes to that.
When we think about being self-directed, that almost sounds like a personality characteristic. And I think that’s part of it. Part of it is confidence. There is that feeling of, this is something I’ve done before, and I’ve shown myself that I can be resourceful. I can get it done and figure out what to do. But I think a lot of it is learning concrete strategies that tell you how to cope with these situations.
If I’m being asked to set my own goal—I haven’t been set a problem, but instead someone says, "I want you to think of something that would be interesting and important to learn about the US Civil War," where do I start with that? How can I, again, look up information that would tell me what’s known about this, what’s unknown, what would be a manageable problem?
All of these are things I think you can learn, but it's a little bit counterintuitive when we think about being self-directed.

[Celia]

I want to dive into some of those concrete strategies. You’ve written about how students tend to gravitate toward study practices that make them feel like they’re mastering the material, but might not actually lead to deep understanding. I know that when I was in college cramming for stuff, I was absolutely doing this: highlighting printouts of slides, hoping that would somehow magically do something. It frequently was not that helpful. How can educators help students overcome this tendency? What should they be doing instead?

[Dan]

It’s really hard. The analogy I used in one of my books is that it’s like doing pushups on your knees. If someone said to you, we’re going to have a pushup contest, so you need to train and be able to do a lot of pushups, and you come back in a month and you see the person and they say, "Things are going great, let me show you!" Then they start doing pushups on their knees. You’d say, what are you doing? "Well, I tried doing pushups the way you said I should do them, but those are very hard. I could barely do any of those. And look, I can do lots of these!"
And so that’s essentially what students do when they’re trying to learn and trying to study. They do things where they see, "Oh, look, my performance is really good. I’m doing lots of pushups really fast." And it feels good. This is not hard. But it’s the same thing. In order to learn effectively, you pretty much need to do things that are difficult, and feel difficult. The fact that it feels hard doesn’t mean that you’re learning, but if you’re learning it’s probably going to feel hard.
But there is this unfortunate characteristic, that frequently your brain does mislead you. For example, when you’re trying to commit things to memory, the environment can support your performance, and you can think that everything you’re doing is based on memory. But actually, some of it is coming from the environment.
The most obvious instance of this—the most common strategy that students use to try and learn is to read things over. They read over their notes. If they have the slides they look at the slides. What this does is it makes the content feel very familiar.
It’s similar to—think about your listeners now. If they re-listen to this podcast, how would you react? Your reaction would not be, "Oh, this is fantastic, this is an opportunity for me to hear all this content again and to really get into it." Instead, you think, "No, I heard this before." And so the fact that you’re recognizing it makes you feel like you know it. When you read over the content, it’s very familiar. You say to yourself, "Yeah, of course, this makes sense, I got this."
But the truth is, if you couldn’t see your notes, you probably wouldn’t be able to produce the content. Recognizing is not the same as being able to produce. You can mistake what’s actually happening in your mind, how much you get out of something, and unsurprisingly, we usually make mistakes in directions that favor the way we want things to go. You usually overestimate how much you’re learning.

[Celia]

Totally. I remember seeing this when I was in grad school. I was a TA for some really bright students taking advanced undergrad classes at UC Berkeley, which is a pretty intense school. A lot of my students, despite being very smart, would express a lot of anxiety when they were given open-ended assignments. They wanted to know what the right and the wrong answer was. Things where they could rely more on that recognition and familiarity seemed more comfortable for them.
I’m curious about how teachers can help students feel more comfortable with doing the harder thing, gaining that deeper understanding, approaching assignments with a greater sense of independence—maybe even in the face of ambiguity.

[Dan]

I see two parts to the question that you just asked.
Students feel uncomfortable because they’re very used to excelling. They want to be sure that they’re going to excel on this assignment. Open-ended assignments have characteristics where I as the instructor can’t tell you exactly what it’s going to look like. When I ask my students–I just finished this semester. It’s a senior seminar on high-level cognition. The final paper, and short papers too, you have to have an interesting idea. I’m grading you mostly based on whether your idea is any good.
Understandably, my students, some of them, get kind of irate. They’ll say, look, just tell me what to do and I’ll do it. How high do you want me to jump? But that’s part of—when you get out of here, not everyone is going to tell you what to do. It’s going to be up to you to come up with an interesting idea. It’s not like baking a pie. I can’t tell you exactly what the recipe is.
Getting back to your question, how can we make students feel more comfortable with that? Well, for starters, we can give them more assignments like that. Especially initially, we can make them lower stakes, so that people are not—when they’re feeling a bit at sea with this kind of thing, because they’ve not done it before, they feel like, well, it’s not contributing that much to my grade. It doesn’t matter that much.
The other part of this is, with open-ended assignments and students feeling more comfortable, we can give them tools to help them know how to do it. If you like this idea, if you want to give more assignments where students are encouraged to be independent and so on, well, if they have very little experience doing that, then you need to teach them what makes, in my case, a question interesting.
I just finished saying, I’m going to grade you based on how interesting your question is. Well, I have some criteria for that. The least I can do is tell my students, here’s what the rubric is. Here’s how I judge whether a question is interesting. A lot of times faculty may not be that comfortable doing that. It’s a moment of self-realization for them. If you’re asking your students to do something and they don’t know how to do it, it’s on you to teach them.

[Celia]

I find that it’s much easier to be self-directed and independent if I’m having fun, which often comes from being curious. But I don’t always feel that way, and it’s a bummer. Is curiosity something that we can develop within ourselves?

[Dan]

The best way to characterize curiosity is, curiosity is a way, clearly, to prompt you to learn about your environment, but you’re not interested in just gathering facts about your environment. You’re really interested in understanding your environment.
You’re not curious about how many bricks there are in that building over there. That’s facts about your environment, but you’re probably not curious about it. We’re interested in learning new things that make us feel like we understand our environment, and understanding really means being able to predict it.
That’s a sensible hypothesis about the function of curiosity. If we can predict our environment, then we are better able to control it. We can anticipate what’s going to happen. That’s obviously going to have a survival function for us.
Controlling that feels improbable. How can I find this comedian more funny? They are or they aren’t. Similarly, curiosity is like, I’m curious or I’m not.
The way I’ve described it, you can, a little bit, try to get meta with it and say, “This is the kind of person I want to be.” I don’t want to immediately shut down—for example, someone invites you to a recital of a type of dance you’ve never experienced before. Candidly, I’m the kind of person who would say, “eh, I don’t know anything about that, I’m probably not going to get it, no.” You can get a little meta and say, "c’mon, Dan, this is the person you want to be." I don’t know how far that’s going to get you, honestly, but it’s probably worth a try.
But the other thing is to try to put yourself in environments that are information-rich. It feels like difficulty is an important variable here. If I know a lot about it, I already understand it. If I know absolutely nothing about it, then, it just hits me and I don’t know what’s going on. I’m not curious to learn more. That intermediate level is very useful.
What you can do is you can try to, more often, put yourself in situations where that kind of information is hitting you. You find websites that have stuff that’s a little meatier than just scrolling your Instagram or whatever it is. Or when you find particular authors—you read something and think, this really got me. I love this. Try to take note of that author and see if you can find other content by that author.
Those are some strategies, I think, in everyday life where you can try to feel more curious about things you want to be curious about. To me, that’s the big characteristic of the internet. I’m not really curious about TikTok. Why am I here for 45 minutes? This doesn’t make any sense to me.

[Celia]

More, after the break.

[AD]

[Celia]

You're listening to Wired for This.

[Celia]

I’m constantly asking myself the same thing. Switching gears a bit: there’s obviously a difference between how we’re taught to think in school, where most of our learning is very directed, versus how we need to think outside of school. Can you talk more about the difference between classroom learning and this more independent, real-world learning?

[Dan]

There’s a lot to say. There’s a bunch of differences.
We’ve already talked about this a bit, but you need to decide what it is you’re going to learn. Some of that may be related to work. There you have more direction. You have a sense of—if you want to learn something during my leisure time that would be useful at work, you probably know what it is.
But then other than that, it’s just stuff that I might find interesting. It really is the case that the world is your oyster. That’s a huge difference.
There are also characteristics of what is demanded of you cognitively outside of school that is quite different from what tends to happen inside of school. One thing is that outside of school, no one checks up on your work. Whereas in school you kind of know they will.
One of the things about cognition is that we’re pretty susceptible to hunches. We have intuitions about things. If you stop and think for a moment and check up on that intuition, you may recognize, “oh no, the first way I was thinking about that was not correct.”
When you’re out in the wild world, you know no one is checking you. You frequently don’t really feel the need to check your work in the way, at school, you might be a little bit more motivated to do it. The status of hunches is likely different.
There are other differences as well. A common characteristic of problems that we have to solve outside of school, real-world problems, is that they’re probabilistic. You need to decide, "Is Willingham going to make a good podcast guest?" There’s not an algorithm for that. You understand that the answer is not really yes or no. Even if you have a threshold, there’s some probability that Willingham is going to be okay, some probability that he’s going to be terrible, and so on.
People are not good at thinking about probabilities. Probability is something that humankind derived as a way of describing the world. It’s not a natural way to think. There are lots of ways we get probabilities wrong. Famous example is the gambler’s fallacy. There are several events that might happen. For example, I’m caught tossing a coin, I get three heads in a row, it feels more likely that the next toss is going to be tails. That’s of course not true. Every flip is independent. But the gambler’s fallacy is that these independent events don’t have independent probabilities.
This is just one example of the fact that people are not very good at thinking about probabilities. That would be a really good thing to be good at thinking about when you get out in the real world and you’re doing things like picking a health plan. You’re judging the probability that you or someone in your family is going to get sick in the next year.
To me, this is a pretty broad theme. You can find areas of cognition, basically types of problems that you commonly work in the real world, or I should say once you’re outside of school, that you just don’t get much exposure to when you’re in school.

[Celia]

In both in and outside of school, we’re often incentivized to present ourselves as smarter and better at stuff than we actually are. Generative AI makes it increasingly easy to look competent while secretly lacking understanding. So, how do you think people, both in education and elsewhere, should handle tools that can do some of our critical thinking for us?

[Dan]

I try not to engage in a lot of crystal ball-gazing because the history of education technology is that—if you look back on the predictions people have made, it’s a joke. In 2011, people were saying with a straight face that in five years there would be 10 universities in the United States. Everybody would go out of business because you could take college courses online and they would be better. Twenty years ago, everyone was saying smart whiteboards were going to completely change the face of education. Anyway, you keep hearing all of this. None of it ever comes true.
That said, I am deeply concerned about student use of artificial intelligence, because it’s just too easy. It makes it too easy to replace what is difficult work of thinking. When you write things out, you go through a different cognitive process. You force yourself to be more explicit, to make your arguments clearer, to think about how your thoughts are organized, what parts are not supported.
Obviously, if you’re using a large language model (LLM), you don’t have to do any of that. It’s very different if you already have a lot of expertise and you’re using ChatGPT, for example, just as an editor. That’s no different, I think, from me sending you an essay and saying, "Celia, would you look at this and help me tweak it?" I have no problem with that, if I’ve already done most of the thinking and then ChatGPT is being used to do a bit of wordsmithing.
But I’m not alone. I don’t know very many educators who aren’t deeply worried about this.

[Celia]

Putting maybe a more constructive spin on it, what skills do you think students will need in the near future to become and practice being independent critical thinkers in a world where technology might let everybody get away with not thinking that hard?

[Dan]

That question has, I think, one foot in labor economics. Why do we think? Most people would say, I think because I need to think for my job. Some people would say, I like thinking. But the things we like thinking about are pretty selective. ChatGPT is not a threat to that. If you like going to the track and figuring out which horse is going to be the best horse, you’ll still do that.
Part of your question is really about, how is this going to change the workforce? People are very much still figuring that out. When you ask what skills are still going to be necessary, and how are schools going to impart those skills—I’m tempted to think that no one knows the answer to that. For my children, I want my children to know how to read and how to write and how to do mathematics and so on. In other words, my answer to that question is not any different today than it would have been 10 years ago. That’s not because I don’t think anything is likely to change, but because I’m playing the odds.

[Celia]

What other questions about independent thinking and self-directed learning have you found yourself grappling with lately?

[Dan]

I think we’ve already talked about it, which is self-management in the face of digital technology. This is something that educators are very concerned about. The research is lagging, but starting to catch up.
The feeling that, especially post-pandemic, students are not motivated the way they used to be. This is not teachers being old fogeys and saying, back in the good old days students really wanted to work. This is something new. Teachers feel like they’re seeing something new. The thing we do have very good data on is students just not showing up for school and record chronic absenteeism.
Part of this, the sub-area of this question on which there’s a lot of research, is whether digital technology destroys students’ ability to focus attention. I think it’s pretty clear at this point that the answer is no. It’s not the case that you just can’t focus attention. Nevertheless, teachers are not wrong in what they think they’re seeing. There’s a distinction between being able to pay attention and choosing to pay attention. What we’re seeing is students more frequently deciding that whatever is happening in the classroom is just not that interesting.
There are several possibilities about why this might be. It could be that the way students value reward has changed over the last generation. It could be temporal discounting has changed. The extent to which I’m willing to put up with doing something that is not very fun in order to get a reward later, that changes over a lifetime. Kids are not willing to do that at all. For a three-year-old, ice cream two hours from now is meaningless. If it’s not in my mouth immediately then the ice cream just doesn’t count.
And so, that changes not only over a lifetime, but it could conceivably change across generations. That could be different in today’s kids, and that’s why they’re less willing to focus attention. This is, to me, a very urgent question, because it gets right at the heart of what has been a theme of this conversation, which is, serious cognition is hard. Learning is hard. Thinking is hard. People will avoid it if they feel like it won’t cost them too much to avoid it. If we want students to do that at school, we need to know ways we can make that more palatable.

[Celia]

Maybe on the flip side, you’ve been teaching for a while. You’ve seen many years of college students come through your university. What, today, gives you the most hope about students’ capacity for independent critical thinking? Have you noticed anything evolving over time in a positive direction?

[Dan]

I think I’ve always been quite positive about what I see in college students. I, of course, recognize that I’m seeing a pretty select group. I’m seeing—these are the students who’ve been really successful in kindergarten through twelfth grade (K-12). They’ve not only graduated high school, but they’ve enjoyed it enough and they’re invested enough, they have enough of their self-image as a student, that they want to continue on and keep going.
I recognize that, that I’m seeing some really good students. But that said, they’re really good. They’re curious about stuff. They’re irreverent at the right times. They’ll call professors out on things they say that aren’t backed up. And I haven’t seen a big change in that. But I’ve always been very optimistic about the state of 18 to 22 year olds in terms of education because they’re smart and resourceful and they’re interested in stuff. It’s very fun to have a job where you get to hang around with that age group in particular and just talk about interesting ideas.

[Celia]

Oh, that's so interesting and maybe leads to a good place to wrap up because a lot of our listeners are scientists. Many of them also teach. What would you tell someone who wants to start fostering a better classroom environment where they can encourage their students to push themselves to think independently and take academic risks? What’s one concrete change that people could make?

[Dan]

I’ll say two things about that. First, I would say—I wouldn’t do that in an introductory class. Students will do it if you ask them to, but you’ll probably be a little disappointed. You will feel a little weird, because they’re going to say a lot of stuff that doesn’t make any sense. I would save it for where they at least have an intro course under their belt. And then, the thing I would do is I would just make it a very explicit assignment. Say something like, "Here’s a situation. Tell me the typical way of thinking about it. Then tell me an unusual way of thinking about it, or a way that raises questions."
You need to give them some tools, because they’re not going to know starting points. There are standard ways of trying to provoke creativity, like thinking in opposites: Normally, we would fully expect this to get larger, this value to get larger. Let me think of a way, no matter how crackpot, that it gets smaller. Even if that doesn’t end up being productive in and of itself, that’s going to free my mind a little bit. I’m going to maybe come up with an interesting idea.
That’s what I would do. I would just make it an explicit assignment. Again, you know your students best, but a lot of students will feel uncomfortable with this, and anxious about it, because they haven’t done it before. Make it a relatively low-stakes thing. If you think they probably won’t do it if it’s low-stakes, make them do it in class.

[Celia]

That’s great. Thank you so much for sharing that advice, and thank you so much for joining us today.

[Dan]

It’s been a real pleasure, thanks.

[Music: Wandering by Nat Keefe FADE IN]

[Celia]

Thanks again to Dan Willingham for joining in on this episode of Wired for This. You can find links to our sources in the episode description.
You’ve been listening to a podcast by American Scientist, published by Sigma Xi, the Scientific Research Honor Society.
Wired for This is produced and edited by Nwabata Nnani and hosted by me, Celia Ford.
Thanks for listening.