Viewpoint Expert opinions on trends and controversies in autism research.
Opinion / Viewpoint

1985 paper on the theory of mind

by  /  9 May 2008
The Expert:


Tricky problem: Children with autism fail this False Belief Task.

And yet, what are we to do about this terribly significant business of other people… so ill-equipped are we all to envision one another’s interior workings and invisible aims?

―Philip Roth, American Pastoral


Most people are vastly more interested in the invisible aspects of other people’s actions than in the visible ones. What we generally want to know about others is their “interior workings and invisible aims” ― that is, their beliefs, desires and intentions.

When someone checks her watch, is she uncertain about the time, late for an appointment, or bored with the conversation? If someone shot his friend on a hunting trip, did he intend revenge or just mistake his friend for a deer? Why did Romeo kill himself, when Juliet was not really dead?

The mechanism people use to answer these questions, by inferring and reasoning about another person’s states of mind, is called the ‘theory of mind.’

The most common measure of theory of mind, in cognitive science, is the ‘false belief task.’ This task is effectively a stripped-down version of Romeo and Juliet.

A preschooler is presented with a story about two characters, Sally and Anne. He is told that Sally has a ball, that she has put the ball in a brown basket and gone outside; that Anne takes the ball from the basket and plays with it inside the house, and then puts it in a green box; and that Sally now is coming back inside to get her ball.

Where, he is asked, will Sally look for her ball?

We know that Sally will look for the ball in the brown basket: that is where she put it, and she thinks it is still there. Children at age 5 see it the same way: they breeze through the false belief task. But 3-year-old children do not.

The younger children consistently predict the opposite: they expect Sally to look for her ball in the green box, where the ball really is.

It’s as if the 3-year-olds cannot take Sally’s (false) belief about the ball into account in predicting her behavior ― or as if they would predict that Romeo will not kill himself, because Juliet is not really dead.

Developing theory:

These results suggest that children’s theory of mind is still developing between the ages of 3 and 5 years. The 3-year-olds don’t yet seem to fully understand the ways that other people’s actions depend on their own states of mind ― rather than on the state of the real world.

There are some striking exceptions to this developmental timeline, however.

Children and adolescents with autism or Asperger syndrome struggle to think about other minds in ways that most people find effortless. This difficulty with understanding other’s thoughts emerges clearly on the false belief task.

In 1985, Simon Baron-Cohen, Alan Leslie and Uta Frith reported for the first time that children with autism systematically fail the false belief task1.

In contrast, children with Down syndrome pass the task, in spite of having lower IQ scores than those with autism. Since then, this fundamental discovery has been replicated many times.

Based on their observations, the authors concluded that autism leads to a specific delay in the development of theory of mind.

The idea that autism is associated with social difficulties was not new. When Kanner first identified autism as a developmental disorder in the 1940s, the characteristic symptoms of autism included difficulties in communication and social interaction. But what was new in the 1985 paper was a proposal for the underlying cause of these difficulties.

Many possible explanations of reduced social interaction in autism have been offered over the years. Do children with autism have trouble perceiving the visible parts of other people’s actions, such as their movements and their facial expressions? Do they have atypical emotional reactions to other people?

The authors were the first to suggest a cognitive alternative: children with autism may be specifically struggling to understand the mental states behind other people’s observable shell.

Failure on the false belief task is not indisputable evidence of a specific delay in the theory of mind, however. Children with autism can fail the task for other reasons.

For example, passing the task requires sophisticated impulse control; children must inhibit the desire to point to the green box, where the ball really is. When the paper was first published, many scientists suggested that this could be why children with autism fail false belief tasks.

Difficult tasks:

The right way to test this alternative hypothesis is to think of another task that is just as difficult, but that doesn’t require imagining other people’s thoughts. If children with autism fail this ‘control’ task, then their failure on the false belief task might not be a symptom of a theory of mind deficit after all.

In 1990, Debbie Zaitchik introduced a control task that has the right properties: the “false photograph” story2. In this task, after Sally puts the ball in its original location, the brown basket, she takes a Polaroid photograph. Then Anne moves the ball from the basket to the second location, the green box. Before the child is allowed to see the picture, he is asked to predict: where will the ball be in the picture?

Baron-Cohen and Leslie were quick to use the false photograph task to test their theory of autism. In separate papers3,4, and with separate groups of children, both reported the same result: children with autism fail the false belief task, but pass the false photograph task. So their original conclusion was supported.

Although children with autism appear to have the logical and cognitive resources necessary to pass the false photograph task, they still fail the false belief task, which hinges specifically on the theory of mind.

Of course, there are still many unanswered questions. Not all people with autism fail false belief tasks: with training, or even just with time, many learn to make the correct action predictions.

Does this mean that a deficit in the theory of mind is only a temporary feature of autism? Or does it mean that some people with autism have found ways of compensating for their weaknesses?

More generally, if difficulties with social interaction are explained by a deficit in theory of mind, what explains that deficit? Autism is a neurological disorder, with a genetic basis. The basis ― either in the brain or in the genome ― of the social difficulties of children with autism are yet to be elucidated.

Finally, there appear to be many different causes of autism, including many different kinds of genetic bases. But why do all of these different biological causes lead to a common effect on the theory of mind?

These questions are the legacy of Baron-Cohen, Leslie and Frith’s landmark papers.


  1. Baron-Cohen S. et al. Cognition 21, 37-46 (1985) PubMed 
  2. Zaitchik D. Cognition 35, 41-68 (1990) PubMed 
  3. Charman T. and S. Baron-Cohen J. Child Psychol. Psychiatry 33, 1105-1112 (1992) PubMed 
  4. Leslie A.M. and L. Thaiss Cognition 43, 225-251 (1992) PubMed 


6 responses to “1985 paper on the theory of mind”

  1. Anonymous says:

    Why does Baron-Cohen not study also the paternal age effect as a major cause of non-familial autism and help reduce the number of autistic children born?

  2. Anonymous says:

    This comment has got little to do with the ‘theory of mind’ hypothesis of autism as far as i can see. The paternal age effect has not yet been widely replicated. Even if confirmed, it is not clear if this operating through a biological mechanism or if it is just an index of the broader autism phenotype. Lastly, the idea that we need to reduce the number of children with autism who are born is ethically contentious, to say the least. We need to be careful with such comments because it can easily be translated into some kind of eugenics. Noone would argue with the idea that we need to reduce troubling symptoms (such as epilepsy or self-injury or gut pain or severe language disorder or low IQ) but these are not autism-specific, and the idea that we should eradicate autism per se overlooks that some aspects of autism are not disabling and can even include areas of talent or strength (such as remarkable attention to detail, and a facility with ‘systemizing’). Interventions or treatments need to target areas of difficulty whilst leaving areas of strength either untouched or even free to blossom.

  3. usethebrainsgodgiveyou says:

    The blind man cannot see…let me help you, look at me!

  4. Ellie Kesselman says:

    I don’t understand. The so-called false belief behavior is not what a rational child or adult would infer.

    Sally puts the ball in the basket. She leaves. She returns. She has no additional information. She looks first where she left her ball. She does not assume that Anne is deceitful or immoral. She behaves as though Anne doesn’t exist. Yet that is NOT the autistic person’s response? The author of this article claims that an autistic person’s response is to make subtle assumptions about Anne’s behavior. This is not intuitive to me, not at all.

    No, I am NOT autistic! I just don’t understand how this is meaningful. It seems to be implying that normal individuals are logical, while autistic individuals act based on assumptions and intuition!

  5. Adri says:

    Usually, puppets are used during this test. And the puppet Sally who is supposed to “go for a walk” just hides beneath the table…and thus “hears” everything. She is indeed present while the “Anne” puppet places the marble in the box. And, after all, they are puppets. They don’t “know” anything. It’s rather a sign of mental health and intelligence not to develop a tom towards a puppet.

    The autistic kid sees a strange adult playing with puppets (strange enough!) and gets strange questions. I believe this test should only be done by a person of trust to get a reliable result. This test is just not applicable to autistic kids, even more since they aren’t really into these make-believe-games (and have a healthy mistrust towards strange behaving adults…)

    This ToM theory only harmed autistic people (false belief as lack of empathy, what on earth could have been worse…!) Getting such a stigma is terrible. Cohen’s hypothesis caused tremendous pain among those he actually should have helped.

  6. Sheogorath says:

    I have an image (a copy of which I sent to the mman himself) which shows Ali G in a yellow tracksuit. The text on the photo says, “Is it ‘cos of my cousin that peeps think I is an idiot?” Yep, Simon Baron-Cohen is the cousin of a man famous for playing the fool, hence the highly similar surnames!

Leave a Reply

Your email address will not be published. Required fields are marked *