Schizophrenia and theory of mind

We suspect that people have an everyday theory of mind because they explain and frequently talk about the behaviour of others and themselves in terms of beliefs and desires. Having a theory of mind means that we believe that other people have minds like ours and that we understand the behaviour of these others in terms of the contents of their minds: their knowledge, beliefs and desires. But how can we demonstrate experimentally that people are using their theory of mind to predict the behaviour of others. This problem is particularly acute in the case of animals or young human children when they do not have language. Dennett (1978) discussing Premack & Woodruff’s (1978) seminal paper ‘Does the chimpanzee have a theory of mind?’, suggested that the use of false beliefs to explain behaviour would provide convincing evidence. When their belief is true (i.e. corresponds to the actual state of the world) we can explain peoples’ behaviour on the basis of the state of the world without needing to know about their beliefs. This ambiguity does not arise when the belief is false. The first experiment to use this approach was published by Wimmer & Perner (1983). They showed that at around 4 years of age a child knows that Maxi will look for his chocolates where Maxi believes them to be, even though the child knows that this belief is false because he has seen Maxi’s mother moving the chocolates. In the English-speaking world the task involving Maxi and the chocolates has become the Sally-Anne task (see this issue, Lee et al. 2004). Since Wimmer & Perner’s publication the development of theory of mind abilities in children has been studied intensively (e.g. Baron-Cohen et al. 2000; Wellman et al. 2001). Theory of mind abilities have also been studied in various non-human species including apes, monkeys, dolphins and birds, but it remains a matter of controversy as to whether any creatures other than humans have such abilities (see Povinelli & Vonk, 2003; Tomasello et al. 2003). It is important not to equate theory of mind with social cognition. Social cognition includes many abilities for which a theory of mind is not necessary. This is clearly the case for monkeys. It is generally agreed that monkeys do not have any knowledge of the mental state of others (Cheney & Seyfarth, 1990), but at the same time they have very sophisticated social abilities, being able to recognize facial expressions, use a variety of calls and keep track of alliances among their peers (Seyfarth & Cheney, 2000). Interest in the study of theory of mind received a considerable boost from the finding that children with autism have specific difficulties with theory of mind tasks (Baron-Cohen et al. 1985). The observation that performance of theory of mind tasks could be impaired while other abilities remained intact suggested that a dedicated neural system might underlie theory of mind abilities. A number of brain imaging studies have confirmed that a rather restricted group of brain regions are reliably activated when volunteers perform a variety of theory of mind tasks. In contrast there has been a surprising lack of reports that theory of mind problems can be acquired as a result of brain damage, but in last few years a few such studies have appeared. A useful review of the neural basis of theory of mind may be found in Lee et al. (2004). Much work still needs to be done before we can specify precisely what the brain regions associated with theory of mind tasks actually do (for some speculation see Frith & Frith, 2003). Does having a theory of mind depend upon being able to perform special kinds of computation, or does it depend on being able to represent a special kind of material (i.e. mental states)? There is some evidence that the material is critical. Autistic children can understand the concept of a false photograph, but not a false belief (Leslie & Thaiss, 1992) and