In contrast, animals do not imitate spontaneously as an attempt to initiate or engage in a social exchange. They may have the capacity for mentalizing, but invariably this is limited to situations that satisfy self-serving needs. For example, amorous male apes and monkeys will manoeuvre female partners out of the line of sight of dominant males in order to copulate surreptitiously.54 Many animals will steal food if they believe that others cannot see the theft. All of these abilities of perspective taking are heightened when there is potential danger from a competitor. However, it is not clear that these evasive actions really involve mentalizing. I know that I can avoid the strike of a snake if I approach from out of its line of sight in much the same way that I can avoid a tumbling boulder if I coordinate my actions correctly. In neither situation do I attribute mental states. I simply observe the actions and reason about what is relevant information. To establish mentalizing, there needs to be evidence of the attribution of beliefs – states of mind that individuals hold to be true about the world in the absence of any direct evidence. If I think you have a belief, then I assume that you hold certain expectations about the world to be true.
Even then, one could attribute belief to others simply by putting yourself in their shoes. For example, we can both separately enter and exit a hotel room and I can describe what I believe you saw based on my own experience. I would reason that because we both went into the same room, you must have seen what I saw. However, that need not be true. You might have had your eyes shut or something in the room might have changed, in which case I would be mistaken. For true mentalizing ability, you need to be able to understand that someone else might hold a different view from yours and indeed be completely wrong about the true state of the world. In other words, the litmus test for real mentalizing is the understanding that someone can hold a false belief.
Consider the following test. If I were to show you a confectionery box with ‘M&Ms’ written on it and ask you what is inside, then in all likelihood you would answer ‘M&Ms’. However, if I open it up to reveal pencils, then you should be a little surprised and possibly a little annoyed because you expected a chocolate treat. If I ask you what you originally thought was in the box, you would say ‘M&Ms’ because you understand that you had a false belief. This may seem trivially easy, but most three-year-olds give the wrong answer and claim that they thought there were pencils in the box.55 It’s as if they have completely rewritten history to fit with what they now know to be true. They do not understand that they held a false belief. Understanding that someone can be mistaken is part of a capability called theory of mind and children operate with an increasingly complex set of assumptions about the minds of others.
If three-year-olds do not understand that they were mistaken, then it is not too surprising that they are unable to attribute false beliefs to others. If I ask you what someone else will answer when posed the same question about what’s in the box, then you understand that they, too, should answer ‘M&Ms’. You can see things from their perspective and understand that they will also have a false belief. Again, three-year-olds give the wrong answer and say pencils. It’s as if they cannot easily take another person’s perspective.
When young children act in this self-centred view they are said to be egocentric because they view the world exclusively from their own perspective. If you show young children a model layout on a tabletop of a mountain range with different landmarks and buildings and then ask them to select a photograph that corresponds to the view they can see, three-year-olds correctly choose the one that matches their own perspective. However, when asked to choose the picture that corresponds to the view that someone else standing on the opposite side of the table can see, they typically choose again the photograph that matches their own.56
Young children cannot easily formulate a mental picture of what it is like to see the world from someone else’s viewpoint. The classic demonstration of this false belief perspective involves two dolls, Sally and Anne.57 In the Sally–Anne task, Sally has a marble that she puts in a toy chest before saying goodbye to Anne and leaving the house. Whilst she is out, Anne moves the marble from the toy chest and places it in the cupboard under the sink in the kitchen. The child is asked where Sally will look for her marble. Adults easily know that Sally will look in the original location. After all, she does not know that Anne moved the marble and she isn’t psychic! Again, three-year-olds fail the test and say Sally will look in the cupboard in the kitchen, under the sink.
Why does it take so long for young children to understand that others can be mistaken? After all, infants understand that adults behave purposefully when watching their actions. One explanation is that young children do not yet understand that others have minds that can harbour false beliefs. Another explanation is that these tests require individuals to make a response that runs counter to what they know to be true. They have to actively ignore the true state of the world. If the task requirements are changed so that the need to respond is taken away, then a different picture emerges. One study examining the looking behaviour of infants reveals that they will look longer when Sally, who should hold a false belief, goes to the correct location as if she knew that her marble had been moved to a new location.58 Sally’s psychic ability creates a violation of expectancy in the infants, so that they are surprised.
Appreciating that others can have false beliefs appears to be uniquely human, as there is no compelling evidence that other animals can acquire this aspect of theory of mind. As noted earlier, they can consider another’s perspective, which is how animals learn to deceive or pay attention to potential competitors; but they do not reliably pass tasks that require understanding that another holds a false belief. When tested on a similar non-verbal version of the Sally–Anne task, apes fail when required to make a choice by looking for food in one of two locations; but like human infants, they seem to register some indecision by looking longer or backwards and forwards between locations when there has been a surreptitious switch of target from one hiding place to the next.59 Together, the looking measure suggests that there is some rudimentary knowledge about mentalizing present in both apes and young infants. However, only in humans does that understanding develop into a full theory of mind that we observe in typical four-year-olds.
Working out what others know is not always as trivially easy as the Sally–Anne task. Consider more complicated plots with more characters and more changes of events. When someone says ‘I know that she knows that he knows’, then they are applying multiple theories of mind. Keeping track of who knows what can easily become more difficult with each layer of plot added. Even then, you have to pay attention because if you miss a key step or forget who did what, you get it wrong.
Add to this the trouble with knowledge. When we know something is true, it is harder to ignore the content of our own minds when attributing a false belief to others.60 We have to actively suppress our own knowledge in order to correctly identify the state of mind in another. As we shall see later, in Chapter 4, deciding not to do something requires actively doing something, which may be compromised in young children and absent in most other animals. So even adults who pass Sally–Anne tasks take longer to correctly attribute false beliefs to others. They are also much slower to solve false-belief situations when you give them a second task that occupies their own minds. It takes effort to think carefully about what others are thinking. Also, it is not clear that adults always employ a theory of mind during most social interactions.61 When you open the door for someone, do you really try to work out what his or her intentions are or do you mindlessly execute an action out of habit? Just because we can generate a theory of mind does not mean that we always do.