This post is the first of two parts. Look for Part Two to run on Early Years soon.
Alison Gopnik is a scientist, but she’s also a philosopher. (In fact, her full title is professor of psychology and affiliate professor of philosophy at the University of California at Berkley.) Gopnik doesn’t find the two titles to be in conflict at all. Her research simply combines both areas of study. She takes big philosophical questions about things like how children come to know their own minds are separate from other minds and answers them scientifically.
Gopnik has published several immensely readable texts explaining both her work and the work of others in child development. My introduction to her work came early in my reporting focus on early education, when I read Scientist in the Crib, in an attempt to understand what all the preschool researchers, teachers and policy wonks were talking about. Gopnik has a new book coming out in August called The Gardener and the Carpenter, which is all about what is known scientifically about the relationship between parents and children.
I recently sat down with Gopnik to discuss her work, and the conversation was so interesting that we’re publishing it in two parts. The first part is about how children learn to navigate the physical world, and the second part will focus on how children learn to navigate the world of the mind.
This conversation has been lightly edited for clarity.
What is the simplest way to explain what you study?
I study how it’s possible for children, babies, and very young children to learn as much as they do about the world around them.
Can you give us an example?
If you had asked people even 30 or 40 years ago about babies and young children, people would have said [babies and young children] are irrational and illogical and they couldn’t differentiate cause and effect and their thinking was very linked to the immediate here and now. And people still have those thoughts about babies and young children.
What I have discovered is that, in fact, even very, very young children already have a lot of the same learning abilities that the smartest scientists have. And in the work that we’ve been doing most recently, we’ve actually shown in some detail what’s going on in their minds that allows them to learn as much as they do.
That’s fascinating, because when you spend time with a toddler you’re not struck by their logic.
Right. Well that’s one of the things that I think is most interesting that we’ve been discovering most recently. You know, if you ask a 3-year-old to tell you what they think, you will get a beautiful stream of consciousness poem about ponies and birthdays, but you won’t get anything that sounds like it makes a lot of sense.
But part of what we’ve figured out is to actually ask children to tell us what they think using their language instead of our language. If you actually show them real objects and get them to really do things with those real objects, they turn out to be remarkably logical and sensible.
So, for instance, what we do is have these little machines they call a blicket detector. The blicket detector is a box that lights up when you put some things on it and not others. And we show the children different kinds of ways that that machine can work, and then we ask them to actually make the machine work themselves.
Wait, let me just make sure I understand the machine. So if you put a spoon on it, it would light up, but if you put a cup on it it wouldn’t? And you would show them that’s how it works first?
Right. What we usually do is we have little blocks. So say you put a blue square block on it and a red, round block on it and it lights up. But then you put the blue block and the red block and another yellow block on it and it doesn’t light up. But then if you take the red one away it does light up. So that’s kind of the standard way that we do the experiments.
And then we’d say “Make the machine go.” And the question is what will they do to make the machine go?
We can also give children quite complicated patterns of evidence. So instead of just showing them, “OK, here’s the red block. It goes. Here’s the blue block. It doesn’t,” we can show them the red and blue blocks together make it go, but if you take the red block off then it stops going. And we can see if they can figure out from those more complicated patterns of data, can they figure out how the machine goes.
Instead of asking them hypothetical questions the way [Swiss developmental psychologist Jean] Piaget, for instance, would ask kids things like how do night and day work, and that’s the sort of question that gets you a lot of stream of consciousness poems, we showed them real things, real objects, that work even in very complicated ways and got the children to make judgments about things like what you have to do to make that machine go. And when we do that, we see that they’re actually incredibly sophisticated and logical.
How old are the children in these experiments?
They range in age, but the youngest that we’ve looked at is about 18 months old. So they range from about 18 months to about 5 years old. So these are mostly toddlers and preschoolers.
And do the younger kids have a harder time with the more complex patterns? Is that how it goes? Or can the young kids get those complex patterns?
It’s actually quite surprising. Even the youngest kids seem to be able to deal with pretty complicated patterns. So let me give you an example. An experiment that we just published we did with 24-month-olds, so that’s just barely toddlers. And we showed them one block that made the machine go four out of six times, and another block that made the machine go four out of 12 times. Basically what the kids are seeing is, sometimes you put the block on and it works, and sometimes you put the block on and it doesn’t work.
So then the question is, “Which block do you want to use to make the machine go?” Now, if you did the math and say which block actually has a higher probability of producing the effect, you’d say yeah, the four out of six block is actually a better choice than the four out of twelve block. But remember, you’ve seen both of these blocks do exactly the same thing.
And you’re 2 years old.
Yeah, and you’re 24-months. Barely 2 years old. Two-year-olds are getting this right. They’re consistently choosing the more-probable block to make the machine go. And we had originally done the experiment with 4-year-olds and we thought well, those are 4-year-olds. But 2-year-olds, not only do they do it, but they seem to be just as good as the 4-year-olds are.
And in fact one of the most striking things that we’ve discovered, just very recently, we just had a paper about this, is for some of these problems the 4-year-olds at least are actually better than grownups. So if you show them the box that works in a really unusual way—so for example you need to put a combination of two blocks on, no block makes it go by itself, but you need a particular combination—and you show them evidence that the block works that way, the 4-year-olds are actually better at figuring that out than Berkley undergraduates are.
That’s really interesting.
Oh, let me give you another example with the 24-month-olds. This is also a nice one ... and gives you a sense of how complicated this is. So in this experiment what we did is we showed the children the blocks on the machine, but this time what happens is two red blocks make the machine go and a red and blue block together don’t make it go, and then two square blocks make the machine go, and a square and a triangle together don’t make it go. And now the kids have a choice between two new round blocks or a round block and a square block. So in other words the thing that makes the machine go isn’t an individual block making it go or not; it’s whether or not the two blocks are the same are different. It’s actually the relationship between the blocks that’s the same that makes the machine go.
That seems so tricky.
Well, it does seem tricky. And in fact if you do this with chimps, where they get a banana if they get it right, they can go for hundreds of trials before they can—and still can’t learn this. But we discovered that 24-month-olds could do it in two trials. So you just show them what I told you. Then you give them a brand new set of blocks, two new ones that are the same, two new ones that are different. They’ve never seen them before. They’re the same on different dimensions than the ones they’ve already seen. And they get it right. They pick the right ones. This is in a paper [we published] in Psychological Science just last year.
Stay tuned for Part Two, which will run soon on Early Years.