‘Theory of mind’ is a catch-phrase in contemporary psychology, referring to a universal human tendency to attribute mental states (feelings, beliefs, intentions, attitudes) to oneself and to others. ‘Mentalizing’ has much the same meaning. Mental states are used to explain and predict other people's behaviour intuitively — as if guided by a theory about the nature of mind.
Imagine you are watching a movie. ‘Why did the detective duck into the doorway? Because he thought he was being followed.’ This explanation holds even if the detective was not actually followed by anybody. What counts is that he believed he was being followed, and this determined his behaviour. Belief is more significant than reality. It is vitally important not to confuse mental states and physical states. To achieve this, the human brain is equipped with a mechanism that represents mental states in a special way. The underlying mechanism is likely to have an innate basis, which explains why social learning early in life is both rapid and universal. One early function of the mechanism may be to direct infants' attention to other human agents, who are likely to affect their lives.
There is increasing interest in the biological basis of mentalizing among neuroscientists and primatologists. It is uncertain whether any non-human primates possess the mechanism, but if so, it is probably limited to apes (gorillas and chimpanzees). Monkeys are thought not to possess it at all. Clues to the physiological basis of mentalizing come from autism, a developmental disorder, which is characterized by severe impairments of social communication. These impairments can be explained by a deficit in the ability to attribute mental states to self and others, suggesting a failure of the mentalizing mechanism. In contrast to children with autism, normally-developing children show evidence of mentalizing during the first year of life. For example, from about 8-12 months of age they spontaneously establish ‘joint attention’ with adults. They follow with their gaze to where an adult looks or points and begin themselves to point at things to show objects of interest to others. This is not so as to obtain things there and then, but to share mental states and to build communicative relationships to other people in the long term. Some time in the second year, children begin to understand pretend play, which relies on the ability to represent mental states and not to confuse them with reality. From the age of 4 or 5, children in all cultures show a reliable and explicit appreciation that mental states may differ from reality, and that one person may think something different from another, and different from what they know to be the truth.
A standard test to demonstrate mentalizing ability requires the child to track a character's false belief. This test can be done using stories, cartoons, people, or, as illustrated in the figure, a puppet play, which the child watches. In this play, one puppet, called, Sally, leaves her ball in her basket, then goes out to play. While she is out, naughty Anne moves the ball to her own box. Sally returns and wants to play with her ball. The child watching the puppet play is asked where Sally will look for her ball (where does Sally think it is?). Young children aged around 4 and above recognize that Sally will look in the basket, where she (wrongly) thinks the ball is.
Mentalizing is pervasive in everyday life, in communication and co-operation, in pedagogy, in play-acting, but also in deceiving, cheating, and outwitting.
Autism — a specific disorder of mentalizing
Autism is a life-long developmental disorder that affects 0.1-0.5% of the population. It is characterized by qualitative impairments in social interaction and communication and the presence of restricted and repetitive interests and activities. Autism is associated with epilepsy (one in three cases) and mental retardation (three in four), and affects at least three times as many males as females. Studies of families and of identical twins suggest a major genetic contribution to autism. Various types of brain abnormality have been found, although the specific and defining features of autistic brains have not yet been identified. This is in part because of the heterogeneity of the condition, which makes it problematic to correlate behavioural observations and physiological data. Autism is now conceptualized as a spectrum of clinical signs — a range of manifestations varying with age and ability, but showing core impairments in social, communicative, and imaginative abilities. One extreme of the spectrum is the severely impaired child of low IQ, who may be silent, aloof, and locked into repetitive motor behaviour (stereotypy). But at the other extreme is the high-functioning individual, who may be pedantic and verbose, active but odd in social approach, with an obsessive pursuit of narrow interests (e.g. collecting registration numbers on lamp posts). This latter picture conforms to the new diagnostic category of Asperger syndrome.
Among the first reliable signs of autism in the young child is an absence of joint attention activities, e.g. a lack of spontaneous pointing or pretend play. These early difficulties signal a failure to attend to other minds, which is later manifest in failure on simple false-belief tests, such as the Sally-Anne task (see figure). Children with autism, even high-functioning ones, seem to be unable to pass this test at the appropriate mental and chronological age. This is in contrast to other children with learning disabilities, such as those with Down syndrome or Williams syndrome. Furthermore, this result holds when alternative reasons for task failure, such as problems of motivation, language, or memory, have been ruled out through the use of closely matched control tests.
A deficit in mentalizing explains the particular social communication difficulties that are typical of children and adults with autism, who nevertheless display a range of other apparently normal social and emotional behaviours. For instance, children with autism are able to use ‘instrumental’ gestures (to affect behaviour directly) much better than expressive gestures (to affect inner states). They are able to understand and engage in sabotage (affecting behaviour directly) better than in deception (to alter somene's belief). They are able to remember messages verbatim, to give factual and honest answers to questions, and to deliver speeches. But they are curiously unable to respond to hints, to engage in gossip, to keep secrets, and to make confidences.
In individuals with Asperger syndrome, verbal ability and social adaptation tend to be higher than in autism. Thus, compensatory learning of social communication skills can occur even in the absence of a start-up mechanism early in life. A proportion of sufferers do gradually acquire mentalizing abilities with practice, and through the application of logical inference. And in some, mentalizing may be merely delayed. However, late-acquired mentalizing appears to lack an intuitive basis, and tends to be fragile and effortful.
What is the neural substrate of ‘theory of mind’? To date there are only a handful of relevant studies, but the new imaging techniques, which allow activity to be detected in the living human brain, are producing interesting results. In such experiments, the brains of volunteers are scanned while they perform tasks in which they must make inferences about mental states in contrast to physical states. During mentalizing, a number of brain areas become active, most notably a region called the anterior cingulate cortex (lying on the medial surface of the frontal lobe, hidden from view in the split between the two hemispheres), a region at the junction of the temporal and parietal lobes, and part of the limbic system called the amygdala. Each of these areas of the cerebral cortex appear to have analogous regions in the brains of non-human primates, and even non-primate mammals. For example, some nerve cells in the equivalent of temporo-parietal region in monkeys become active when the monkey watches humans or other monkeys walking. In the same general area are found nerve cells that are similar to the ‘mirror neurons’ described sby the Italian neurophysiologist, Giacomo Rizzolatti, in a part of the monkey's frontal lobe. These cells respond when the monkey looks at particular actions (e.g. grasping) being carried out either with his own hands or by the hands of another monkey. The other main component of the mentalizing system, the anterior cingulate cortex, is involved in the ‘monitoring’ of action. In humans it is particularly active when we monitor our own thoughts and feelings.
To speculate, the brain's mentalizing system might have evolved from a system concerned with analysing the movements of other creatures and from a system that processes information about our own inner states.
— Uta Frith
Bibliography
- Baron-Cohen, S., Tager-Flusberg, H., and Cohen, D. (ed.) (2000). Understanding other minds: perspectives from developmental neuroscience, (
2nd edn ). Oxford University Press. - Happé, F. and Frith, U. (1996). The neuropsychology of autism. Brain,
119 , 1377-400
See also learning disabilities.
