The Metaphors of Learning

The importance of talking about learning & How to do that

16 minute read 09 January 2020 reflections

We are all learning experts. We have been learning since we were born, to survive, to grow, to adapt. It’s a consequence of or a motivation for all our actions. We are sent to school for years, we are asked to learn from our mistakes, and to experience new things to learn more. But as much as we are very good learners, do we have such a good understanding of how it works?

A few weeks back, i asked the internet to draw what learning means to them. I wanted to know more about how people represent learning, and specifically what is important to them in learning.

If you didn’t participate, i recommend you pause now and try to do this yourself: how would you represent learning? To you: when and how does learning happen?


Ok, back to it! The first thing i noticed is how difficult it can be for people to answer these questions: not only by drawing, but also by just talking. When i asked why, people often answered “learning is too abstract”. We discussed a bit more about this and realised that abstractness might not be the only problem, because we talk a lot about many abstract things: think about love for example, or taste, or art, or what it means to be human. But many of us don’t talk about learning all that much. We ask kids if they learnt something at school today. We complain after a bad presentation that we didn’t learn anything. But we don’t usually go much deeper than that. I remember that in high school, when a teacher told us we would have a presentation about how to learn, we all suspiciously looked at him: but we know how to learn, we do it everyday, what more could there be to it?

But as you grow, and discover more about your learning process, you realise that some topics are easy for you to learn, but some others are much more difficult. Does it mean that the teacher sucks? Does it mean that you are not made for these topics? Or is there something about the way you attempt to learn, or the way it’s taught, that could have an influence? If you are trying to build a house, but the ground is unstable and the scaffolding is inadequate, you might struggle a bit more than necessary.

deus ex machina

Deus Ex Machina - Guy Billout

> the extra bit_

The truth is: talking about learning is difficult. Understanding what learning is is difficult. In learning sciences, just like in physics, our goal is to build models to be able to predict certain outcomes. For example, in physics, we would like to be able to say things like “If i apply that force here, this object will accelerate that much”, so we can launch rockets to the moon and what not. In learning sciences, we want to be able to say things like “If i give a person that exercise, this person will learn that much”, so we can enable people to send rockets to the moon.

The thing is, in physics, the stuff we care about is often measurable: we can measure how much an object moves when a force is applied to it. In learning sciences, it’s not all that easy. We can’t reliably measure how much someone learnt. We can evaluate learning with tests, but if you ever had an exam in your life, you know how it’s possible to have a good grade and be totally unable to reuse what you learnt. We can also use brain imaging to see how the brain appears transformed by a learning experience, but we are very far from having measurements precise enough to deeply understand anything. Moreover, in learning sciences, human beings are involved, and that means we need to be careful regarding the influence of our measurements as it can affect the person being evaluated. But the fact that these things are difficult to measure does not mean that we shouldn’t care about them!

knowledge of how people learn in a graph

Knowledge of how people learn. Reproduced from [Bransford 2000].

Ok, so to untangle this mess, we need to start somewhere. In research, there are two main metaphors of learning, that means: two ways of talking about learning [Sfard 1998]. First, the acquisition metaphor: this idea that knowledge is something to acquire and own within oneself. This is what we use when we say things like “Did you get it?”. This metaphor is really centered on the individual: you possess knowledge, you get knowledge, you build yourself as the owner of many knowledge items. These knowledge items are incomplete, ill-shaped, but they are yours.

acquisition metaphor

Acquisition Metaphor

Second, the participation metaphor: the idea here is that knowledge is something bigger than the individual and that we all contribute to it. For example, this metaphor accepts that no-one gets the full picture of mathematics, and no-one ever will. But, as a community, we can grow an understanding of it. When you are learning mathematics, you are actually joining the mathematics community and contributing to its quest.

participation metaphor

Participation Metaphor

Why does this matter? Well, imagine the influence of such vocabulary in schools for example. If we go for a more acquisition perspective, the idea is, through the help of others, to build and grow our own knowledge. Everybody has their own little copy of The Maths™, including variations resulting from experiences with the topic. Going to school means growing some kind of toolbox in your head and body. Now, if we look at this from a participation perspective, going to school means connecting with a community, for example the community of mathematicians. The teacher is an experienced member of the community, and through learning, one gets a chance to grow the knowledge of the community and advance the field. The motivation is really different! In the first case, we grow knowledge to make ourselves more powerful. In the second one, we grow the community’s knowledge of the topic to increase its reach.

In particular, this change of metaphors has consequences on how we perceive the main elements of learning, as listed in the table below.

metaphorical mappings

The Metaphorical Mappings. Reproduced from [Sfard 1998].

Personally, the participation metaphor often makes more sense for me. Thinking this way really helps me learn better. But in research, both these metaphors are still useful, as they explain different phases and effects of learning more or less easily. In that sense, they complement each other. [1]

> the extra bit_over_

Now, what did you folks think about learning? First of all, many of you did not explicitly mention a teacher (in the common definition of it), nor included a school. I find that interesting, because although these are designed to help us learn, we acknowledge that they are definitely not necessary. So what’s left? What is important to learning?

végé’s learning

Végé’s learning

For Végé and Fraisier, learning is like a video game. You gain experience, and using this experience, you can acquire new skills, or reach higher levels. It’s about progression: a continuous process. But it’s also about unlocking new abilities: a jump or a break towards a higher step. I think that’s quite nice, maybe you even felt it yourself: you learn more and more about a topic, and suddenly, someday, POOF, you feel like you can actually use this knowledge to do actual things in life. Just like how leveling up in a video game gives you access to new spells, new weapons: basically new ways to interact with the world.

fraisier's learning

Fraisier's learning

Now, if we stay a bit longer in the context of video games, i would say that playing a game is a learning experience. You start with very little skills, facing levels that seem impossible sometimes, but through practice, you finally manage to solve them, and even end up finding them easy. I am always amazed about how focused and resilient we can become when playing video games.

And a thing that video games got right is the importance of failure. For Nulk and Gwen, learning is a process that includes failure, either as a drawback on the way, or as a necessary step to efficiently learn.

nulk's learning

Nulk’s learning, reproduced and translated

Well, intuitively, this makes sense. Imagine someone gives you a game to play with. Let’s say they tell you “This is how the game is solved” and then show you the solution. You would know how to solve the game now, but would you still be able to solve it months later? Now imagine they just give you the game and let you figure out the solution on your own, through try and fail cycles, with help when necessary. If you were to play the game again later, how do you think this would go? Much better, i believe.

gwen's learning

Gwen’s learning

The curious mind could wonder: is it important that i am the one who fails? For example, what if i see someone trying something out, and failing. Would it have the same impact as trying things out myself and failing? I’ll let you think about it, and we will come back to it in an article on failure!

waffle's learning

Waffle’s learning, edited

A lot of you also mentioned the importance of splitting the problem into pieces. For example, when L34 sees something she doesn't understand, she decomposes it into smaller chunks, and learns by experimenting with them. She might not understand everything, but at least she has a better idea of how it works.

l34's learning

L34's learning

For Šaljivac, learning is about iterating, until finding the optimal path to a solution, similar to the salesperson problem in Computer Science. Many paths lead to the destination, but one is optimal. Not taking this optimal path does not mean that you are wrong, but that there is still space for learning.

saljivac's learning

Šaljivac's learning

What does this tell us? Learning is something that can be decomposed, iterated, experimented with. I think this is an important thing to remember. We don’t have to (and generally won’t) find an optimal solution on the first try. And it’s perfectly ok. I would even say it’s important: every time we take a new path in our problem space, we have a chance to find a new approach, a new understanding, of the problem.

ken kong's learning

Ken Kong's learning

Ok, but this is all very internal, this is about how one can learn by themselves. What if we look at the bigger picture? What external factors make learning possible? If you look at Ortie’s drawing, you can see that learning is supported by research, specialists, but also an environment. All of these are very important, but i want to focus on the environment for a bit, as it particularly crucial. For example, imagine you are trying to learn about tree species. Where do you think you will learn better? In a classroom or in a forest? Or let’s say that you want to learn aerial silk. Would you rather start above a mattress or above the ground?

ortie's learning

Ortie's learning

Following up on that, with this very sweet drawing, Zmaj explains how dwarf mongoose parents teach their kids how to hunt: they start by hunting a prey themselves, and injure the prey a bit, by hurting a leg and such. Then they give the injured prey to the kids, so they can learn to hunt on a easier target. Ok, now that i wrote this with words, i am not sure anymore if it’s sweet or disturbing, but well, here it is.

zmaj's learning

Zmaj's learning

And finally, from Bouh’s perspective, learning is just like putting some new glasses on. I like this very much: learning transforms your view of the world.

bouh's learning

Bouh's learning

Now, i have to play the game too! And tell you about how i would represent my learning. When i was a kid, i thought of learning as building a heap. I had these blocks that i added in my head, a new block building on top of the previous ones. The blocks had different shapes and rigidity, according to the topic. If not stimulated for a long time, the blocks could fade. This was really scary for me, because i thought everything would just fall apart if they did.

The Stacking Perspective

The Diverter’s learning: The Stacking Perspective

When i grew up, things changed. I started to see knowledge as opportunities for connections. In this perspective, knowing something just means i can connect to some topics more easily [2]. Forgetting something means the connection loses strength, but not necessarily that everything else gets destroyed. I find this visualisation much closer to my experience, and also much less stressful.

The Connecting Perspective

The Diverter’s learning: The Connecting Perspective

All these representations echo very interesting topics and questions of Learning Sciences. I will go deeper into them with following articles, to help you improve your learning, but also your teaching. And now, please, tell me: did you ever have a meaningful learning experience? A thing that you understood and transformed your perception of the world?


The Diverter

Much to learn you still have

PS: I would like to thank all the people who sent their contributions for this article, including @Extraits_deVie and @meresauvage. I was very happy to discover all your drawings, and i think they add up to a quite nice coverage of the field!

To go further:

The Diamond Age - Neal Stephenson.

This book really triggered my curiosity for learning when i was a teenager. I always loved to teach, but never really thought of investigating the theory behind it until then. The Diamond Age is also exploring the role technology can play in learning, and how it can impact different people.

The Diamond Age's cover

The Diamond Age - Neal Stephenson

I also add here some other drawings i received. I did not have space to discuss all of them in the article, but i think they are also interesting to look at! You can scroll through them to see the variety or perspectives on the topic.

brandebouc's learning

Brandebouc's learning

Big J's learning

Big J's learning

Melon's learning

Melon's learning

Mandarine's learning

Mandarine's learning

Thunder's learning

Thunder's learning


[1] For those out there who like to code, i like to see this as a Java vs Haskell kind of relationship for example. Some problems are going to be trivial to solve with Haskell, but overly complex with Java. And also, sometimes, the other way around.

[2] A bit like how intermolecular forces can make two molecules connect with one another.


[Bransford 2000] Bransford, J. D., Brown, A. L., & Cocking, R. R. (2000). How people learn (Vol. 11). Washington, DC: National academy press.

[Sfard 1998] Sfard, A. (1998). On two metaphors for learning and the dangers of choosing just one. Educational researcher, 27(2), 4-13.