11.4 Children's Understanding of Minds

To develop as fully functional humans, we must learn not just about the physical world but also about the social world around us. Most of us—adults and children alike—spend more time trying to understand other people than trying to understand inanimate objects, and we apply entirely different explanatory concepts to the two endeavors. In our explanatory frameworks, billiard balls move because they are hit by other balls or cue sticks, but people move because they want to get somewhere. We are all psychologists in our everyday lives, continually trying to account for people’s behavior in terms of their minds; psychologists sometimes refer to this as folk psychology. We attribute emotions, motives, feelings, desires, goals, perceptions, and beliefs to people, and we use those attributes to explain their actions. This is referred to as theory of mind—a person’s concept of mental activity; the ability to understand one’s thoughts, feelings, and behaviors and those of others. Theory of mind implies having some causal-explanatory framework to attribute intention to and to predict the behavior of others (Astington & Hughes, 2013; Wellman, 1990).

David Premack (1990) suggested that beginning at a very early age, humans automatically divide the world into two classes of entities—those that move on their own and those that don’t—and attribute psychological properties to the former but not the latter. When 3- to 5-year-olds saw videos of balls moving like billiard balls, only in response to physical impacts, they described the movements in purely physical terms; but when they saw videos of balls moving and changing direction on their own, they immediately regarded the balls as representing people or animals and described the movements in mental terms (Premack, 1990). A child described one sequence of movements as one ball trying to help another ball get out of a hole.

Even Very Young Children Explain Behavior in Mental Terms

By the time children have learned language sufficiently to offer verbal explanations—that is, by about 2 to 3 years of age—they already explain people’s behavior in terms of mental constructs, especially in terms of perceptions, emotions, and desires (Hickling & Wellman, 2001; Lillard & Flavell, 1990). They expect others to respond to objects that they (the others) can see but not to objects that they cannot see. They describe a crying person as sad. They say that a person filling a glass with water is thirsty and wants a drink. In one experiment (Repacholi & Gopnik, 1997), 2- year-olds demonstrated an understanding that another person’s desires could be different from their own. Having learned that a particular adult preferred broccoli to crackers as a snack, they gave that adult broccoli, even though their own preference was for crackers. Unlike the 2-year-olds, however, 14-month-olds gave crackers to the broccoli-loving adult.

In another experiment, researchers showed that even 12-month-olds can display a remarkable understanding of what is in another person’s mind (Tomasello & Haberl, 2003). In that experiment, each infant played with two adults and three new toys, one toy at a time. One of the two adults left the room while one of the three toys was being played with and therefore did not see or play with that toy. Then, at the end of the play session, all three toys were brought into the room on a tray and the adult who had missed playing with one of them looked in the direction of the three toys and said, “Wow! Cool! Can you give it to me?” In response, the majority of infants gave the adult the toy that that adult had not played with before, not one of the two toys with which the adult was already familiar. To perform in this way, the infants must have known which toy was new to the adult, even though it wasn’t new to them, and must also have known that people are more excited by new things than by familiar ones.

Delay in Understanding Beliefs, Especially False Beliefs

As adults, we explain people’s behavior not just in terms of their perceptions, emotions, and desires but also in terms of their beliefs, and we know that beliefs can be mistaken. For example, if we see a man carrying an umbrella on a sunny day, we might explain that he must have believed it was going to rain that day. Three-year-olds, however, rarely offer explanations in terms of beliefs (Colonnesi et al., 2008; Saxe et al., 2004), and tests indicate that they do not clearly understand that beliefs can differ from reality (Wellman et al., 2001).

Figure 11.10: A test of ability to understand false belief In this test, which is usually presented with the help of puppets that act out the sequence, most children under age 4 say that Maxi will look in the red cupboard.

(Adapted from Wellman et al., 2001.)

A typical test of false-belief understanding is the following (illustrated in Figure 11.10). The child is told a story, which is also acted out with puppets for clarity, in which Maxi puts his candy bar in a blue cupboard. Then Maxi leaves the room and his mother comes in, finds the candy bar, and moves it into the red cupboard. Then Maxi reenters the room to get his candy bar, and the child is asked: “In which cupboard will Maxi look first?” Most 4-year-olds answer, just as you would, “In the blue cupboard,” but most 3-year-olds insist that he’ll look in the red cupboard (Wellman et al., 2001). The problem isn’t poor memory because the 3-year-olds in these experiments have no difficulty reporting accurately on all the factual details of the story when questioned. Rather, they seem not to understand that someone can believe something that isn’t true.

Three-year-olds’ denials of false belief apply even to their own false beliefs. In one experiment, 3-year-olds were shown a crayon box and asked to say what they believed was inside (Atance & O’Neill, 2004). Each of them said, “Crayons.” Then the children (who were tested individually) were told that they could get some paper to draw on with the crayons if they wanted to. When the children returned with drawing paper, the box was opened, and it proved to have candles inside rather than crayons. When asked what they thought the box had in it when they first saw it, most said, “Candles.” When asked why they had gotten the paper if they thought the box contained candles, most had no plausible explanation.

Perhaps the concept of false belief is particularly difficult to grasp because of its inherent contradiction. False beliefs are both false and true at the same time. They are false in reality but true in the minds of the believers. In this way they differ from the products of make-believe.

Make-Believe as a Precursor to the Belief-Reality Distinction

Three-year-olds may have difficulty understanding false beliefs, but they have no difficulty understanding pretense. Toddlers who are 2 and 3 years old, as well as older children, engage in an enormous amount of pretend play, and researchers have found that even 1.5-year-old infants differentiate between make-believe and reality (Leslie, 1994; Rakoczy, 2008). An 18-month-old who turns a cup filled with imaginary water over a doll and says, “Oh oh, dolly all wet,” knows that the doll isn’t really wet.

These 3-year-olds may not understand false beliefs but they certainly understand pretense. Pretend play helps children acquire the belief-reality distinction and also helps them learn to reason hypothetically.

©Robert Brenner/PhotoEdit–All rights reserved.

Alan Leslie (1987, 1994) has suggested that children’s understanding of false beliefs emerges from their earlier understanding of pretense. Pretense is similar to false belief. Both, by definition, are mental conceptions that depart from reality. The only difference between the two is that pretenders know that their conception doesn’t match reality, whereas believers think that theirs does. Three-year-olds, who fail false-belief tests such as the one in which a crayon box actually holds candles, do not fail analogous tests in which they are asked to report what either they or another person had pretended was in the box before it was opened (Lillard & Flavell, 1992; Woolley, 1995).

Figure 11.11: Older siblings promote false-belief understanding Three- and four-year-old children with older siblings succeeded on a standard test of understanding of false belief at a much higher rate than did those who had no older siblings.

(Based on data from Ruffman et al., 1998.)

Children everywhere engage in pretend play, whether or not they are encouraged to do so (Carlson et al., 1998; Pellegrini & Bjorklund, 2004). Piaget (1962) regarded such play as an expression and exercise of the child’s ability to symbolize objects in their absence, but many developmental psychologists today ascribe even further significance to it. Leslie (1991) suggests that the brain mechanisms that enable and motivate pretend play came about in evolution because such play provides a foundation for understanding nonliteral mental states, including false beliefs. A child who understands that pretense differs from reality has the foundation for understanding that people’s beliefs (including the child’s own beliefs) can differ from reality and that people can fool others by manipulating their beliefs. Mark Nielsen (2012) goes even further, proposing that pretence and counterfactual thinking involved in fantasy play, along with imitation, are not only essential components of childhood but actually responsible for the emergence of the human mind. According to Nielsen (2012), “By pretending children thus develop a capacity to generate and reason with novel suppositions and imaginary scenarios, and in so doing may get to practice the creative process that underpins innovation in adulthood” (p. 176).

Evidence for the view that pretend play promotes false-belief understanding comes from research showing strong correlations between the two. Children who have engaged in lots of pretend role-play with other children pass false-belief tests at a higher rate than do children who have engaged in less (Jenkins & Astington, 1996). Other research has shown that children who have child-age siblings at home, especially older siblings, pass false-belief tests at a much higher rate than do same-age children who lack such siblings (McAlister & Peterson, 2007; Ruffman et al., 1998; see Figure 11.11). Children with siblings engage in much more role-play than do those without siblings, because their siblings are always-present potential playmates (Youngblade & Dunn, 1995). Social role-play (role-play with another child) may be more valuable for development of false-belief understanding than is solo role-play because in social role-play children must respond appropriately to the pretend statements of the other child, not just to their own pretend statements: I’m your mommy, and you must obey me, or Bang, I got you. They get used to the idea that other people can hold concepts in their heads that do not reflect reality.

Autism: A Disorder in Understanding Minds

Suppose you were oblivious to the minds of other people. You would not feel self-conscious or embarrassed in others’ presence because you would have no understanding of, or concern for, their thoughts about you. You would not ask others about their thoughts or inform them of yours because you would have no reason to. You would not look where others look, attend to their words, or in any way try to fathom their perceptions and beliefs. People would serve the same function to you as inanimate objects or machines. You might try to get a bigger person’s attention to help you get a cookie from the top shelf in the kitchen, but that person’s attention would have no value to you in and of itself.

If you had these characteristics, you would almost certainly be diagnosed as having autism, or autism spectrum disorder, a disorder which, as mentioned in Chapters 2 and 4, is characterized by severe deficits in social interaction, severe deficits in acquiring language, a tendency toward repetitive actions, and a narrow focus of interest (American Psychiatric Association, 2013). Among the earliest signs of autism in infants are failure to engage in prolonged eye contact, failure to synchronize emotional expressions with those of another person, and failure to follow another person’s gaze (Baron-Cohen, 1995; Mundy et al., 1990). The deficit in language seems to be secondary to the lack of interest in communication. Unlike children who fail to learn language because of deafness, children with autism rarely use gestures as an alternative form of communication, and when they do, it is almost always for instrumental purposes (for example, to get someone to help them reach a cookie). Those who learn language at all learn it late, almost invariably with the help of deliberate teaching, and their language always contains peculiarities that seem to reflect a lack of sensitivity to other people’s minds and perspectives.

Figure 11.12: Performance of individuals with autism and typically developing 4-year-olds on false-belief and false-picture tests As shown here, individuals with autism were much more likely to understand that a picture could misrepresent current reality than to understand that a person’s belief could misrepresent current reality. In contrast, typically developing 4-year-olds performed better on the false-belief tests than on the false-picture tests.

(Based on data from Leslie & Thaiss, 1992.)

As you might expect, people with autism perform poorly on false-belief tests and on tests of ability to either deceive or detect deception (Tager-Flusberg, 2007; Yirmiya et al., 1996). In fact, Simon Baron-Cohen (1995) proposed that the primary deficit of these children is an inability to read minds, or what he calls mindblindness. Stated another way, people with autism lack a fully developed theory of mind—the ability to understand our own thoughts, feelings, and behaviors and those of others. In one experiment (Leslie & Thaiss, 1992), relatively high-functioning children with autism and adolescents whose verbal abilities were equivalent to those of typically developing 6-year-olds were compared with typically developing 4-year-olds on two false-belief tests and two “false-picture” tests. The false-belief tests were versions of the changed-location and container tests described previously, and the false-picture tests were constructed to assess the understanding that a photograph, rather than a belief, might misrepresent reality. In one false-picture test, for example, the child saw a photograph being taken of an object at one location. Then the object was moved to a new location, and the child was asked where the object would be in the photograph when it was viewed. The results of the experiment are shown in Figure 11.12. As you can see, the individuals with autism performed much worse than the typically developing 4-year-olds on the false-belief tests, but much better than they did on the false-picture tests. This experiment not only demonstrates the specificity of the intellectual impairment in autism but also suggests that the human capacity to understand mental representations (beliefs) is distinct from the capacity to understand physical representations (pictures).

In line with Leslie’s theory that make-believe play is a developmental precursor to understanding false beliefs and other nonliteral mental representations, children with autism have consistently been found to lack such play (Mastrangelo, 2009; Wulff, 1985). Children with autism explore the real physical properties of objects, as do typically developing children, but they do not make one object stand for another or pretend that an object has properties different from those it actually has. In contrast, children with developmental disorders such as Down syndrome, including those who have less understanding of the physical world than children with autism, do engage in pretend play (Hill & McCune-Nicolich, 1981) and eventually develop a much better understanding of false beliefs and deception than do children with autism (Baron-Cohen et al., 1985; Yirmiya et al., 1996).

Do Chimpanzees Have a Theory of Mind?

Some scientists believe that our close genetic relatives, chimpanzees, may be described as having not a theory of mind but rather a well-developed theory of behavior. As we noted earlier, chimpanzees do not engage in shared attention, and although they live in complex social groups and display high levels of social learning, they are not able to pass false-beliefs tasks that most 4-year-old human children do (Herrmann et al., 2007; Krachun et al., 2009). In some circumstances, however, they seem to realize that if another chimpanzee or person is looking at something, that individual is aware of the object (that is, the “eyes have knowledge”) (Hare et al., 2001), but not in other situations (Povinelli & Eddy, 1996). Chimpanzees thus seem to have the rudiments of theory of mind, but not much more, suggesting that “mindreading” may be an ability that is unique to humans. However, chimpanzees do seem to have a limited understanding of the psychological states of others, which suggests that our common ancestor with chimpanzees also likely possessed the social-cognitive abilities that would one day lead to human theory of mind (Tomasello, 2009).