Introduction

Please imagine a pizza, in your “mind’s eye.”

What does that actually look like? How detailed is your pizza? How much information is “baked” into your image? Is it truly an image, or just a thought? If someone were to record the activity of neurons in your brain, would they be able to decode that you were thinking of a pizza?

If we want to understand how we think, understanding how we represent ideas is a good first step. It’s also grounds for lively debate! Below, you’ll see competing viewpoints about the accuracy of our “mind’s eye,” and how these different perspectives affect how we study and understand the mind.

Imagination

Let’s address the question of the accuracy of a mental image with an exercise.

• Close your eyes and imagine a cat next to a mouse.
• Open your eyes.

Now, try this one:

• Close your eyes again and imagine the same cat next to an elephant.
• Open your eyes.

Now, consider what you imagined! When you imagined the cat next to the mouse, did you imagine them to scale, like in Figure 3.1?

Or did you imagine them around the same size, like in Figure 3.2?

Odds are that you likely imagined the animals more like Figure 3.1. That is, you imagined the cat larger than the mouse, and the elephant larger than both of them. Additionally, you may have had to “zoom out” your perspective to fit the elephant into your mental image. Why? You don’t have to do any of that; it’s your imagination! Things don’t have to have a realistic scale because a mental image has no physical size.

When you assign a realistic physical size to things you imagine, it’s called pictorial representation. When you don’t, it’s called abstract representation. Do both types of representation happen in most peoples’ brains? Almost certainly. Have cognitive scientists argued for years about whether pictorial or abstract representation is “correct”? Absolutely.

Mental Travel Time

One way to put some nice, concrete numbers to this topic is to measure what we call mental travel time. In a mental navigation task, you are asked to follow a path along a map inside your head. This is how you do it:

• Consider a location you are familiar with: a city, a neighborhood, a campus, or even a home.
• Select two points within that area.
• Imagine going from one of those points to the other.

There, you’ve done it! That may have been pretty simple, but the way you did it is of great interest to cognitive scientists. Did you imagine every step of the way, experiencing your surroundings as if you were actually there? Did you imagine a blurrier, skimmed-over version of the real thing? Or did you forgo details entirely and simply imagine teleporting from one spot to the other? As cognitive scientists considered these questions, a debate emerged. Some believed firmly that our mental images are faithful, proportional representations of the world (like the animals you imagined earlier). The other side argued that mental images are abstract, and don’t have to follow the rules of real three-dimensional space. You’ve already seen an anecdotal argument for the “realistic” side; let’s now examine some experimental data.

A 1978 experiment¹ made a clear argument for the realistic side when it asked subjects to memorize a map like the one in Figure 3.3, and then to imagine a small black dot travelling from one point to another.

Figure 3.4 shows what they found: a linear relationship between real distance and mental travel time. The farther away two shapes were on the physical map, the longer it took for subjects to imagine moving from one to the other.

This is a fairly straightforward argument in support of the idea that our brains imagine things to the scale that they are in the real world. There is no actual “distance” between the square and the triangle in your brain; why should it take any longer to imagine travelling between the square and the triangle versus the square and the circle? Are we automatically acting out events in our brains as if they obeyed real physics?

This question contains the word automatically for good reason. It’s possible that the to-scale imaginations of this experiment’s subjects were not entirely automatic. In this study, subjects were explicitly asked to imagine a small black dot moving from one location to another. At face value, this seems like a minor thing to point out, but consider the consequences: If I ask you to mentally travel from wherever you are right now to Paris, France, are you going to imagine packing your things, getting into a car, train, boat, and/or airplane, and journeying there? This might take you a few minutes. Will you imagine a map, with you represented as a little dot that zooms from one place to another? This will take several seconds. Or will you simply imagine being in Paris – with no time passing at all? In the experiment described above, subjects were explicitly asked to use a method of imagining that would not only take a measurable amount of time, but also one that would scale with real-life distance. In other words, the experimenters designed an experiment that would very likely prove their point.

For the other side of the argument, let’s examine a different experiment. We’ll now look at some data from 2003² that achieved quite different results by changing the instructions. Instead of asking subjects to imagine a black dot moving from one map location to another, this group asked subjects to visualize a light turning off at one point of the map and turning on at another point. In other words, they asked subjects to visualize instantaneous travel. Unsurprisingly, when asking subjects to imagine something that is instantaneous, they found no difference in reaction time, regardless of how far apart the two areas of the map were.

These experimenters also presented a different approach: imagine the relative orientation of two objects to each other. For example, think of the map in Figure X. Where is the pentagon in relation to the square – above, right, below, or left? They found a similar result in these experiments – the time it takes for a person to visualize the orientation of the objects is about the same, regardless of how far away the objects are from each other. This suggests that we have the “data” for the map saved in our minds, like how a computer might store coordinates in a table, or how you can represent a circle with a mathematical formula.

As mentioned earlier, there has been a lot of debate about which is the “correct” way to describe how our brains represent the world, and as is usually the case with psychology debates, both sides do a good job at describing different things. There is not one simple way in which we represent images in our minds. However, thinking about these questions helps us step forward into more exciting questions, like…where’s my cheese?

Mental Maps

Consider a rat. Let’s name him Tolly (after a scientist we’ll mention shortly). Tolly has to navigate a maze to get to a nice wedge of cheese. As you can see in Figure 3.6, the cheese is to the left of Tolly’s starting position. However, to get to the cheese, Tolly first must turn right. Suppose Tolly runs through this maze several times. Does he understand where the cheese is in space? Or does he just remember the memorized steps – right, left, left?

Good news! There is a way to test this! After Tolly gets used to the maze in Figure 3.6, you can give him the maze in Figure 3.7.

If Tolly has simply memorized the set of directions he needs to follow to get to his cheese, he will go down Path B, because it is to the right. That will not lead him to his cheese. However, if Tolly understands where the cheese is in actual space – that is, if he has a mental map of his world, he will take Path A, because he knows that the cheese is ultimately to his left.

Well? What do rats actually do? In 1948, a scientist named Edward Tolman (hence: Tolly) wrote a paper called Cognitive Maps in Rats and Men³, which I mention by name here because it’s quite a pleasant read. In this paper, Tolman details several arguments for the idea that rats (and humans) have a cognitive map in their minds. In the experiments coming from Tolman’s lab, rats do the equivalent of selecting the left path in Figure 3.7. In other words, they know where the cheese is in the world, which means they must have some mental picture of what the world looks like.

It’s a nice way of putting together the ideas of pictorial and abstract imagery. To have a mental map of a place the way that rats (and humans) do, you have to have an understanding of the true spatial relationships of things. But you can also access that information quickly and efficiently, without having to “explore” the map inch-by-inch in your mind. What, then, does this mental data actually look like? This is what we will explore in the next chapter.