Skip to main content

The Nature of Imagination

How children's imaginations can inform education and parenting
Girl reading book with sparkle concept

Children's imaginations are complicated and impressive, says Professor Paul Harris. Yet, oftentimes, when we watch at children pretending, we write it off as fantasy or child's play. What are educators and parents missing in those moments? How can adults be better informed about the nature of children's thinking? From how children mimic reality while pretending to why children develop fears to how they differentiate between make believe and reality, Harris' decades of research demystifies the complexities of children's thinking. 

“Once we move away from the preschool, where of course, pretend play is valued and into the classroom, the idea is that children could learn simply by exercising their imagination is rarely something that's deployed,” he says. “And yet, in many ways, I think it would be a source of pleasure for children to sit and think about the world.”

In this episode of the Harvard EdCast, Harris takes us on a journey through children's imaginations and contemplates how educators and parents can better use children's imaginations. 


Jill Anderson: I'm Jill Anderson. This is the Harvard EdCast.

When you watch children play pretend it's easy to dismiss their behavior as silly or just fantasy. Developmental psychologist Paul Harris has spent decades studying the nature of children's minds. He says for a long time we used to see children's imaginations as something negative -- not something that could actually inform everyday thinking or mature thinking.

It turns out there's a lot more going on in children's minds than you realize and there's ways to make better use of this in a classroom or at home. I was curious about Halloween or how something a child loves can turn into something they fear and how children differentiate make-believe from what adults tell them. First, I wanted to know whether we still harbor negative views of children's imaginations.

Paul Harris

We'd move too far because I think if you look at, particularly in the United States, there's a preoccupation with the possibility that young children's pretend play is a sort of harbinger for later creativity. And so there's a lot of anxiety about what's happening in preschools, the extent to which a more academic curriculum is constraining the time that children can spend in pretend play.

And at its most traumatic, the implication is that we're going to be raising a generation of dull children who lack imagination because their opportunities for play were so limited in early childhood. So that view, I tend to think of that as the so-called romantic view of the child's imagination. And equally, I think there are problems with that view as well as the negative view.

Jill Anderson: Is the thought that we've gotten too far away from children having that time to play due to all of the standards and requirements and things?

Paul Harris: Well, that's a conventional thought among middle class involve parents that they fret about the extent to which early childhood should be a time of freedom and exploration and worry about the intrusion of a more academic curriculum. What I'm hesitant about is the bigger message, which is that in some ways, young children are passing through a kind of critical period where we need to nurture their imagination. I think that's overstated.

In many ways young children are surprisingly uncreative. Their pretend play is very, very impressive and complicated when you start to think about how it operates. But if you give children various problems, such as making a simple tool, four and five year olds are pretty hopeless. I mean, they can deploy familiar tools, but you don't see them solving engineering problems by making a tool out of material, fashioning it and then using it.

Whereas, to use a dramatic example, we know that some birds can do that. A crow can take a piece of wire and bend it and stick it down a tube and extract a reward, whereas a four and five year old has a lot of difficulty doing that. So a sense of which we overestimate young children's creativity. They may well become creative, but it's very difficult to pick up early signs of that capacity in early childhood, if by creativity we mean surprising innovations.

Jill Anderson: Right, right. Yeah. You just mentioned how children's imaginations are more complex than we often realize, and we just kind of pass it off as, "Oh, they're just pretending, they're just playing make-believe," but there is actually a lot more happening when children are using their imagination. Can you talk a little bit about that and how reality informs imaginative play?

Paul Harris: Yes. I can give you a very simple working example. So let's say the child is around two years of age and perhaps playing with an older sibling or a caregiver and they're playing at tea parties. The partner, the older sibling or the caregiver picks up a teapot. Now, because this is a pretend tea party, there's no real tea in the teapot, but nevertheless, the teapot is tipped and there's a cup and the child is then handed the cup.

Well, I mean, if the child were completely reality-oriented, so to speak, the child might say, "What am I supposed to do with this cup? There's nothing in it." The average two year old will not do that. They'll lift the cup to their lips and engage in pretend drinking. The fact that they do that, I think we take pretty much for granted. It seems very straightforward that they would do that.

But if you unpack it a little bit, you have to acknowledge that in order to do that, the child has realized that as where there's make-believe tea in this teapot, furthermore, that make-believe tea behaves in much the same way as real tea. So when you tip the container, the tea will pour out and it doesn't go sideways. This is not a world that is radically different from the real world. The tea is going to pour out of the spout and descend into the cup, and then the child also can presume that this make-believe liquid when you move that cup, the liquid is going to stay in the cup so the child can take it to its lips.

Well, I'm sort of laboring the point, but what I'm trying to underline is the fact that the child's interpretation of a joint piece of pretend play is underpinned by this ability to realize that it's to some extent regulated by a lot of understanding of the causal regularities in the real world.

And without that, it would be very difficult for play partners to come together and make sense of what they're doing. So that's why I say in some ways, children's pretend play is infused with their understanding of reality. If we also look at slightly more complicated pretend activities when children are playing with dolls or even a three and four year old might create an imaginary companion. Again, those dolls are infused with thoughts and feelings and perceptions and desires as is the imaginary companion.

And again, the child is borrowing from reality. It's not as if the child is confronted by this lifeless doll and doesn't know what on earth to do with it. The child has lots of knowledge about these imaginary creatures that they can draw on in order to act out various scenarios with them. Or in the case of an imaginary companion, sometimes to form a sustained relationship with this purely imaginary person.

Jill Anderson: You're talking about reality and then I'm thinking about a case like Halloween, which you might consider a time when a lot of people, even adults, are kind of leaning into their imagination of things that perhaps are not always reality-based.

Paul Harris: In some ways, I would say because it's an adult directed regulated ritual, it's perhaps a good deal, safer than it might otherwise be. In other words, the more frightening possibilities that we as adults can potentially associate with Halloween to do with ghosts and the afterlife and resurrection and so on and so forth.

I mean, the average five year old I think is fully insulated from those things. I mean, maybe there's a quick visit to a spooky house, but it's often with a secure base of the caregiver nearby. And of course, especially if I may say so, in the United States, Halloween has become a kind of candy fest.

Jill Anderson: Yeah.

Paul Harris: It's not much to do any deep existential issues. So to that extent is not necessarily an enormously, what can I say, liberating playground for the child's imagination. Having said that, of course, a lot of children derive enormous pleasure from thinking about their costume and the person that they're going to be.

That's a further manifestation of what we've already talked about, namely the ability of even young children to conjure up imaginary persons and in some ways, to occupy, to impersonate or to stand in and act out that person. And of course, in that case, a costume admired by friends and neighbors is going to be a boost.

Jill Anderson: Right. I want to bring up Little Red Riding Hood and share an experience that my daughter had with this. I suspect it's an experience a lot of parents or people who are working closely with children will understand.

My daughter, when she was about three, became very, very interested in Little Red Riding Hood. Loved the story, would want to watch anything about it and so this went on for a really long time, like a year. Halloween came and she wanted to be Little Red Riding Hood. We got her at a costume, we dressed up as a group in different characters, and then she developed a total fear of wolves.

She could no longer look at Little Red Riding Hood, no longer could look at a wolf on a TV. Just very, very terrified. So my question to you is about when something moves beyond just an imagining or playing healthy imagination or what we might perceive as a healthy imagination with a story or a character, and then it verges into this area of fear, what do you know about that?

Paul Harris: So it's always tricky to interpret one individual case, but if we think about more broadly, children's preoccupation with stories and many children, we're talking now about preschoolers, which I think your daughter was during this period, have what you might call an attachment to one or two particular stories. They want that story to be read to them sometimes on a somewhat demanding, regular, daily basis.

And for the most part, those stories can be a source of reassurance and pleasure and comfort rather than of anxiety. Having said that, I think it's possible that in the case of your daughter, maybe there was a shift in her assessment of the wolf, who's after all, one of the central characters in the Little Red Riding Hood story.

So there's some interesting older work done by someone called Arthur Jersild in Columbia in the 1930s who became interested in children's fears. And what he found was that as children became a little bit older, so we're now talking about four, five, six or so, whereas two and three year olds didn't often show any fear of frightening creatures, monsters, for example. That tendency to become afraid of frightening creatures increased as children entered, as it were, the kind of elementary school years.

Now, one plausible interpretation of that is that in some ways, perhaps at first your daughter was thinking of this as purely make-believe, but as she learned more about the world the wolf in the story could be connected to what she knows about wildlife in general, that there really are dangerous creatures out there. And so it's possible that that was what was happening in her case, that the status of the wolf as a purely make belief character slowly shifted over time. I'm not sure if that's the correct interpretation.

Jill Anderson: Right. But I feel like a lot of parents and just guardians experience that type of thing where their child can be really interested in something and then suddenly it shifts and becomes a challenge as an adult to respond. Is there a good way to respond to something like that when there's a fear and a phobia that helps and supports them?

Paul Harris: In the case of the wolf, I suppose it's somewhat difficult because there really are wolves.

Jill Anderson: Right.

Paul Harris: I mean, one suggestion I would have, and here I'm stealing from Arthur Jersild, who I mentioned a moment ago was that he encouraged parents whose children were, for example, having trouble settling down to sleep at night to actually deploy the child's imagination. So you encourage the child to think that they had a magic sword or a magic shield, which would protect them from predators of all descriptions.

And according to Jersild, that was relatively effective so you fought fire with fire, so to speak. So the child who's got an over rich imagination and is imagining that this wolf is going to be at the door, so to speak of their own cottage, you equip the child with, as I say, some kind of protective but imaginary device.

Jill Anderson: I have heard that just telling a child something doesn't exist, it doesn't really help them very much in those scenarios.

Paul Harris: Yeah. And in some ways it comes back to something we observe with adults. If we go to a horror movie, we know when we walk in the movie theater that there's going to be something frightening. But we also know that it's pure make-believe, but that doesn't stop us from becoming... Our heart beats faster at certain junctures during this horror movie. So the child's and the adults' emotional system is not very well attuned or calibrated in terms of the reality of the situation that's being thought about.

Jill Anderson: I wanted to talk to you about when children are told something versus what they actually believe. And I know you've done a lot of work in this area looking at when we tell a child about something like the tooth fairy or we talk to a child about religion, obviously these are two different things, but how are these different for children?

Paul Harris: Well, now you're putting your finger on a fairly thorny deep question. Even though we think of children as hands-on learners, that underestimates them dramatically because there are many things that they can't put their hands on, They can't put their hands on the shape of the earth, and yet by six or seven, they know that it's round.

In the past three years, we've seen children preoccupied by, worried by the coronavirus, but none of them have ever seen a coronavirus. So in many ways, children deploy their imagination to think about aspects of reality which are ordinarily hidden from view, and that includes long-term historical processes, which they can't watch. It's not as if they can go back and see how the Civil War unfolded in the United States, but we tell them about it. It also applies to many aspects of science, which are invisible. Children learn about germs and viruses and atoms and oxygen.

And so in many ways what we see among children, but we also of course see this among adults, is what I call trust in testimony. So the next step is to ask, and you raise this issue yourself. How does that trust in testimony stand vis-a-vis children's exposure to claims about the afterlife or the existence of God or the existence of the ancestors?

And I'm tempted to say, actually it's very, very hard to draw conceptual distinction between those domains. Of course, as adults, we intuitively think there's a big difference between, well several things. We think there's a big difference between the existence of viruses, the possible existence of witches or witch crime and the existence or possible existence of God.

But I think from the point of view of a child, drawing those conceptual distinctions is more challenging because in all cases, the child is being told about these entities primarily by adults, is hearing narratives, stories, biblical or otherwise about these entities and there's very little firsthand experience that the child can draw on to make some of the conceptual distinctions, which we adults are tempted to make.

I think what happens in due course is that not only our children, as it were, listening to adult testimony, but they also become alert to certain subtle signals in that testimony. So if the parent is telling the child a story about mermaids, there will be certain clues in the narrative to tell the child, so to speak, that this is make-believe. This is not real. It's not as if the child is warned, "Well, look, when you go swimming on that beach or in that lake, beware, you might run into a mermaid."

That same parent is talking about viruses. The child may well be warned about washing their hands regularly wearing a mask. So in other words, some of these invisible unobservable entities, by virtue of the way that the adult talks about them become part of the sort real world furniture, others are in this more imaginative or make-believe space.

Now, that still leaves the question of how parents talk about religious figures because in some ways religious figures are clearly different from mermaids. Well, what we find is of course, especially if children are being raised in a family that regularly goes to church, the child ends up with of course, a belief in the existence of God.

But we also have discovered, and this is something that we find both in the United States and in other countries, the child has a belief in scientific entities and religious entities, but if anything, they express a little bit more confidence in the existence of the scientific entities, than the religious entities.

So if you ask them how sure they are that there are germs, they will say, "Oh, I'm really sure about that. They certainly exist." "Have you seen one?" "No, I've never seen one." "Do you know what they look like?" "No, I don't know what they look like." "But you're sure they exist?" "Yes, I'm sure they exist."

On the other hand, if you pose those same questions about religious phenomena, be it the soul or God or others, the children are fairly confident that these things exist but not as confident as they are about the scientific entities. And our interpretation of this is that in subtle ways, parents signal to their children that now we're moving into this different realm, we're moving into the realm of religion. And they talk about it, for example, with slightly less certainty, sometimes underlining the fact by saying, "Well, our family believes such and such."

But a parent wouldn't say in talking about atoms or viruses, the parent wouldn't say, "Well, in our family we believe such and such." So children pick up fairly early on, on this distinction between, as it were, well established scientific phenomena and religious phenomena, which some people, of course would want to say are also well established. But nonetheless, what our data suggest is that in the minds of adults, there's this qualification or uncertainty and they convey that to their children.

Jill Anderson: And it's so fascinating because the children can pick up on that.

Paul Harris: Right. Children have very good antenna with respect to the confidence and certainty with which they're being told something.

Jill Anderson: It's kind of interesting. It's almost like they don't just outright believe what they're told necessarily. There's some skepticism.

Paul Harris: Right. We see that in other ways. So children are very good at figuring out who's a more reliable informant versus a less reliable informant. Even with a very brief exposure, they can pick up on somebody who says things that are turnout not to be true, and they'll remember that for some time afterwards. So this has been an important line of work that I and others have been pursuing, what we call it selective trust.

So the idea that young children are credulous with respect to all comers and all messages is just plain wrong. Children are surprisingly astute in discriminating among different informants, and to some extent, as I was saying earlier, discriminating among different types of message.

Jill Anderson: I'm wondering where we go from here with that kind of knowledge. Is there a way to take some of this and make use of this information in the classroom or as a parent?

Paul Harris: Well, one area where I think there might be hope and possibility for the world of education is this notion of thought experiments. So just to elaborate on that a little bit. So as I mentioned earlier when I was talking even about two year olds, the two year old can imagine easily what happens when the empty teapot is tilted. The child can visualize, so to speak, the tea entering the cup and goes ahead and pretend drinks the tea.

So in the world of early childhood education, there's been traditionally a lot of emphasis on the way in which children learn from hands-on experience. They're supposed to touch and observe, and this will teach them the way that the world works. But I would argue that that underestimates the possibility of children learning simply from using their imagination and thinking about what will happen. So let me give you a concrete example of where we've done a mini intervention.

So if you take children around the age of three and you present them with a couple of tubes, which are crisscrossed in a sort of X-type fashion, if you drop a ball at the top of one of the tubes, we as adults realize that it's going to fall in a diagonal path and it's going to come out on the other side, so to speak. But if you ask the child to try to catch the ball, they will put their hand directly underneath the point of entry.

So rather than going to the following the diagonal, so to speak, they put their hand underneath the point where you put the ball in. Now, why do they do that? Well, nine times out of 10, when you drop something, gravity ensures that it drops vertically. It doesn't drop diagonally. So what children are doing in this little setup is bringing their understanding of the world again to bear.
But of course, in this case, their understanding of reality is a bit circumscribed because they're failing to take into account the fact that the trajectory of this ball inside the tube is constrained by the walls of the tube. However, it is also true that especially by the age of two and three, children are very good at realizing that one solid object cannot actually pass through another. It will be blocked. A ball that hits the wall doesn't travel through the wall, it's constrained by the wall.

So now putting all of this together, what we have done is to ask children to, as we are putting the ball, as it were, just at the top of the tube before we're releasing it, we can ask the child to stop and think, to try to imagine what's going to happen to the ball. And what we find is that once we do that, the child is less likely to make this gravity error, are more likely to realize that the ball is going to travel diagonally down the tube in which you drop it. So you need to put your hand out on the left hand side or the right hand side, depending upon which diagonally you've chosen.

So this is a very basic and simple illustration of the way in which the child's potential for using their imagination can teach them something about the way that the world works, namely in this case that whilst gravity is a very potent force, it can be constrained by another force that they already know about, which is the solidity of the tube within which the ball is traveling.
Now I'm emphasizing thought experiments because if you be looked back at the history of science or the history of philosophy, thought experiments, thinking about what might happen if you do such and such, and realizing that something you hadn't really thought carefully about would happen so that you really need to revise your conceptual apparatus, that's been a very potent engine in the development of science and in the development of philosophy.

Galileo famously asked, [inaudible 00:26:05] you to think about two different objects of different weights that are yolked together and how rapidly they will fall and use this as a compelling argument with respect to the impact of gravity.

Now, in the educational system, the idea that children could be little Galileos is never entertained, whereas I'm saying no, their imagination is good enough, is reality bound enough for it to help them think through situations which perhaps on at first site draw false conclusions about. But if they were to engage in some more sustained, careful, imaginative reflection, they would realize that they ought to revise their thinking about it. And indeed, that's what we found with this tube study that I just described. So there's a quick example of the potential for the child's imagination to be deployed in the classroom.

Jill Anderson: We were talking a little bit about how we've... It was considered a negative thing, the imagination, and then it went kind of in another direction. We've leaned a lot into the pretend play and imaginary play, but not into this area that you're just talking about, which is the thought experiment.

Paul Harris: Yes.

Jill Anderson: And we can really be using imagination in a different way as parents, as educators to help children learn about things.

Paul Harris: I couldn't have put it better myself.

Jill Anderson: Why would you say ultimately it's so important to do that, even at home?

Paul Harris: Of course, in some ways, there's a practical benefit. We can get children to think about the ways that the world works in more accurate ways. And so in some ways it's a teaching tool, but I also would say insofar as we see children engaging in pretend play, it's something for the most part, they enjoy. And yet, once we move away from the preschool, where of course, pretend play is valued and into the classroom, the idea is that children could learn simply by exercising their imagination is rarely something that's deployed.

And yet, in many ways, I think it would be a source of pleasure for children to sit and think about the world. I've used an example, which is based on, as it were, naive physics, But we can also imagine children being invited to think about growing up in a different place, being born with a different family, belonging to a different community, and thinking out the implications of what their value system would be like had they been born in that different place or time or nation or whatever.

So I think this tool is not just one that would be confined to the physical or inanimate world. It's also a tool that could be exploited or used in the child's thinking about the social human world as well.

Jill Anderson: Paul Harris is a developmental psychologist and professor at the Harvard Graduate School of Education. His most recent book is Child Psychology and 12 Questions. I'm Jill Anderson. This is the Harvard EdCast produced by the Harvard Graduate School of Education. Thanks for listening.


An education podcast that keeps the focus simple: what makes a difference for learners, educators, parents, and communities

Related Articles