Back

The “Theory of Mind” Hypothesis: Explaining Autism

© 2002, Satyan Chari, B.O.T., G.C.Occ Thy, Msc. Occ.Thy., University of Queensland

Introduction
People with autism sometimes have difficulty comprehending when others do not know something. Autistic individuals can become extremely agitated when another person does not know the answer to a question they ask. By not understanding that other people think differently than themselves, many autistic individuals may have problems relating socially and communicating to other people. That is, they may not be able to anticipate what others will say or do in various situations. In addition, they may have difficulty understanding that their peers or classmates even have thoughts and emotions, and may therefore appear to be self-centered, or uncaring.

The term ‘theory of mind’ was coined by Premack and Woodruff, 1978 and is often used to refer to the ability to “attribute mental states and to use these invisible postulates to explain behaviour in everyday life. Premack and Woodruff defined theory of mind as the ability to ascribe mental states to oneself and to others. The ability to ascribe these mental states was called a ‘theory’ because mental states are not observable. Premack and Woodruff also used the term ‘theory’ as mental constructs of states of mind are used to predict behaviour.

The theory of mind concept was actually first proposed in research exploring primate psychology and primate social organisation. The concept was initially used to explain observations where the chimpanzees demonstrated behaviours that required them to know what another chimpanzee may be thinking, or was possibly about to do, in response to an occurrence in its environment. The ability to ascribe mental states to other chimpanzees was called a ‘theory of mind’ (ToM). This concept was subsequently identified by researchers in the field of autism as a possible absent ability that may be resulting in some of the manifest behaviours seen in children with autism. The definition of what a ‘theory of mind’ has however, changed as research has progressed and has different shades of meaning across different areas of research.

The Theory of Mind Hypothesis in Autism

A causal definition of autism at a biological level continues to remain elusive. But a fairly clear idea is emerging of a cause at a cognitive level. Specific discrete behavioural deficits are seen in autism such as impaired social functioning, impaired communication and lack of imaginative play. In fact, the triad of social impairment, reduced communication and decreased imagination characterizes the whole spectrum of autistic disorders (Wing, 1986). Baron-Cohen, Leslie and Frith (1988) suggested that this association between these three areas of functioning could be explained by a single cognitive deficit, that is, an inability to cognitively represent mental states, otherwise called a theory of mind. Representing mental states such as belief and desire is required to predict other people’s behaviour.

Frith, Morton and Leslie, (1991) described the difference between normal and autistic children as “ in a normally developing child, the computational capacity to represent mental states has an innate neurological basis. In the autistic child damage to the circumscribed system of the brain has occurred, and this prevents the normal operation of the critical cognitive mechanism”. The ability to represent mental states would appear to be crucial for normal reciprocal social interaction, for understanding a speaker’s intended meaning in communication, and for understanding contrast between real and imaginary events. Wimmer and Perner (1983) designed the first false belief tasks to test the hypothesis. The test required the ability to attribute a false belief to another person (Sally – Ann task). Wimmer and Perner found that almost all children over the age of 4 years passed the task. In contrast, Baron-Cohen, Leslie and Frith (1988) found that only 20% of a sample of autistic children were able to pass this task despite having mental ages over 4 years. It is interesting to note that in the same study, 85% of a sample of children with Down’s Syndrome passed the test even though they had significantly lower verbal ability than the age-matched group of children with autism. Subsequent studies have confirmed that a large proportion of children with autism have great difficulty in tests designed to tap into ‘Theory of Mind’ ability.

Levels of Theory of Mind

Two orders of theory of mind ability have been described in literature (Baron-Cohen, 1995). First-order ToM refers to inferring the thoughts of another person; an example of this could be a child (X) knowing what his friend (Y) might be going through after losing his favourite toy. Normal children pass first-order tests of ToM by the age of four years. Second-order ToM refers to reasoning, that is, what one person (other than the self) thinks about another person's thoughts. Extending the same example, this would refer possibly to a third child watching the two children and reasoning what child X would be thinking about child Y.

Higher functions in the Theory of Mind and Asperger's Syndrome

Children and adults with Asperger’s Syndrome have also been shown to suffer from deficits in ToM though they are more likely to pass tests of ToM relative to the performance of children and adults with autism (Baron-Cohen, 1995). Since many children and adults with Asperger's disorder do pass second-order ToM tasks, such tasks cannot be interpreted as conclusive evidence of intact ToM. However, a person’s performance on such tasks does provide evidence for the level of impairment.

For those who do pass such tasks due to a high level of functioning, more advanced tasks can measure ability to read nonverbal cues such as facial and eye expressions. As a group, children and adults with Autism or Asperger's disorder are less able to interpret nonverbal language and cues than normal persons are. We often use facial expressions to determine what a person may be thinking or what message he/she may be giving us. A deficit in this ability puts anyone at a major disadvantage in almost every situation, especially at social situations when a person is required to interpret such cues from multiple persons simultaneously. These cues let us know when it is all right to interrupt, when others are interested in what we are saying, when to stop talking, when others are lost by tangential conversation, etc. Without this ability, a person is likely to feel overwhelmed in such situations and may prefer to retreat from others.

Biological causes of Theory of mind impairments

The biological causes of autism vary greatly. It is three to four times more frequent in male children. This, along with the fact that increased concordance rates are noted in monozygotic twins points towards a probable genetic association. There has been a lot of research done on people with autism in the past two decades to identify the affects autism has on the brain. Every lobe of the brain, with the exception of the occipital lobe has been implicated in the pathophysiology of autism. However, specific brain regions seem to be more involved in studies where subjects were tested on theory of mind tasks.

Brain studies of normal subjects engaged in tasks designed to tap into the theory of mind have shown increased activation of a range of frontal regions. Individuals with frontal lobe lesions have also shown to display abnormalities of social and emotional functioning (Saver and Damasio, 1991). Adults with right hemisphere damage appear to have social and communicative deficits that resemble those seen in high-functioning individuals with autism (Happe, Brownell and Winner, 1999).
Baron-Cohen, Ring, Moriarty, Schmitz and Costa (1994) and Fletcher, Happe, Frith, Baker, Dolan, Frackowiak, and Frith (1995) found frontal brain regions to be specifically more active during theory of mind tasks than during control tasks but there is relatively little agreement on the exact sub-regions involved. In the study by Baron-Cohen et. al. (1994), increased activation was noted in the right orbito-frontal regions. Fletcher et al. (1995) used stories requiring mental state attribution, a skill difficult for even high-function autistic children and found through PET imaging that left medial frontal regions were significantly more active during the task.

Gallagher, Happe, Brunswick, Fletcher, Frith and Frith (2000) found activity increases in the medial frontal and paracingulate regions through MRI studies. They also identified increased activity in the temporal-parietal junction and temporal poles bilaterally when subjects were engaged in theory of mind tasks.

In a study by Happe, Brownell and Winner (1999), left hemisphere damaged patients showed no evidence of ToM impairments, but it needs to be noted that none of these patients had lesions to the key frontal areas implicated in imaging studies of ‘theory of mind’. While left frontal regions may play a key role in the neural circuits underlying ToM, the most probable explanation would indicate a neural circuit that has important centres in other regions of the brain (primarily in the right frontal lobe) which would cause similar deficits if affected.

Current Theoretical Models explaining Theory of Mind Deficits:

Evidence from brain imaging studies and the peculiar association between specific deficits in children with autism, has provided researchers with information to build models to explain theory of mind deficits. Current models that attempt to explain the theory of mind concept (and impairments in it) can be organized into two distinct categories. One group of models is based on the theory of modularity proposed by Fodor (1983), the second group of models can be classified as non-modular.

Modular Theories

Those in favour of a dedicated cognitive mechanism, or module, underlying
The theory of mind have pointed to evidence of dissociation in developmental
disorders. Theory of mind ability appears to be dissociable from cognitive ability in other domains. For example, children with autism may lack a theory of mind but may be intelligent in other respects (Baron-Cohen, Tager-Flusberg, Cohen, 1993), and children with William’s syndrome, can show intact theory of mind despite delayed mental development in other domains (Karmiloff-Smith, Klima, Bellugi, Grant and Baron-Cohen 1995).

I. Brother’s Social Brain Theory

Brothers, Ring and Kling (1990) suggested proposed that social intelligence should be differentiated from other kinds of intelligence (hence a modular theory). He suggested that social intelligence might have its own neural circuitry (amygdala, orbitofrontal cortex and superior temporal sulcus). He proposed that these neural structures together could be called the “social brain”. Studies on selective social impairments seen in brain damaged patients provided considerable evidence for this theory.

II. Leslie’s Modularity Theory

Leslie (1991) suggests that a specific area in the brain that is selectively responsible for understanding and interpreting mental states is present in normal cases and impaired in individuals with autism. He coined the term ‘Theory of Mind Mechanism’ to describe what he considers to be a ‘module’ responsible for this function.

III. Baron-Cohen’s Innate Minimalist Modularity Theory

Baron-Cohen (1995) proposed a theoretical model (based on his background in evolutionary biology and developmental psychology) that consisted of four discrete sub-systems or mechanisms, which together comprise the human ‘mind-reading’ system. The first mechanism is the ‘Intentionality Detector’ (ID). The ID is proposed to be a perceptual device that interprets ‘primitive volitional’ mental states such as goal and desire (want). These are seen as basic mental states required for making basic sense of the movements of all organisms in the environment. For example, an infant seeing an adult move across the room would (through the ID mechanism) interpret that as ‘the adult wants to go there” or “the adult does not want to stay here”. The ID is considered to be an innate mechanism that infants possess for reading mental states.

The second innate mechanism is the Eye Direction Detector (EDD). Baron-Cohen suggests that the EDD has three basic functions of detecting the presence of eye-like stimuli, computing whether eyes are directed towards it or towards another direction and inferring from the observation that the organism’s eyes are directed at something else that it actually sees something. In simpler words, the third function interprets gaze as “seeing”. The last function is considered especially important to mind-reading as it allows the infant to attribute a specific perceptual state to the organism (“the monkey sees the banana” or “mummy sees me”). Both these mechanisms allow the infant to interpret observed behaviour as a small number of mental states. Another defining similarity between these two mechanisms is that they allow the infant to interpret behaviour on a dyadic level, that is, between the organism/agent and the infant or the organism/agent and the object it sees. This is considered by Baron-Cohen to be suggestive of the ‘autistic’ child.

The third mechanism is the Shared Attention Mechanism (SAM). The SAM is considered to be a higher order skill that allows the individual to form what is called ‘triadic representation’. Triadic representations conceptualize relations between an Agent, the Self and an Object (which can be another agent). The SAM builds triadic representations by perceiving the perceptual state of another agent and computes shared attention by comparing another agent’s perceptual state with the self’s current perceptual state.
The last mechanism described is the Theory of Mind Mechanism (ToMM). The ToMM is considered to be a system for inferring the complete range of mental states observed from human and animal behaviour. The ToMM achieves this by representing the set of epistemic mental states (imagining, dreaming, knowing, deceiving, believing etc.) and combining the different mental state concepts (volitional, perceptual and epistemic) in to a rational understanding of how mental states relate to actions and each other.

IV. Johnson and Morton’s Minimalist Innate Modularity Theory

The modularity theory proposed by Johnson and Morton (1991) is a truly minimalist theory. The authors of this theory postulate that only two innate mechanisms are involved in the steps leading to acquisition of a theory of mind. The first mechanism is termed CONSPEC which takes hold of the infants attention to look at a face-like stimuli and CONLERN which directs the infant’s attentional system to learn about faces. Therefore according to this theory, mechanisms such as ID, EDD and ToMM would in turn be acquired modules.

Non-Modular Theories
General Learning Mechanism Theory

Extrapolating the argument of minimalism to an even more simplistic level, Baron-Cohen (1998) proposes a general learning mechanism theory, which encapsulates the view held by certain learning theorists. This theory argues that there is nothing innate in cognition and that everything learnt from the environment. There are fundamental deficits in this kind of a model in that it cannot account for certain kinds of conditions. An example of this would be a case of a person with Asperger’s syndrome who have good general intelligence but have difficulties in the social domain. This would contradict a singular all-encompassing learning theory as good general intelligence would indicate a good general learning mechanism and could not therefore explain social deficits.

The ‘Fish out of Water Theory’

The 'fish out of water' theory is that children with autism have failed to develop mammalian behavioural response patterns, so they have to rely on an earlier system of orientations. Griffin, J., (1999) advanced this theory based on certain phylogenetic observations of fetal development, and stress-responses of normal infants and children with autism. He proposed that the ontogenetic mirroring that is seen in human development may include certain mechanisms in cognitive development. A lack of appropriate ‘mammalian’ responses could be attributed to a developmental deficit according to this theory. The research support for this theory is however, minimal.

Implications to Practice

The theory of mind hypothesis is increasingly being recognised as an explanatory cause for a large majority of deficits seen in Autism Spectrum Disorders. Theory of mind refers to the notion that many autistic individuals do not understand that other people have their own plans, thoughts, and points of view. Furthermore, it appears that they have difficulty understanding other people's beliefs, attitudes, and emotions.

Many of the tasks used to test this theory have been given to non-autistic children as well as children with mental retardation, and the theory of mind phenomenon appears to be unique to those with autism. Acquired theory of mind deficits are also seen in patients with brain injuries, those who have undergone neuro-surgical procedures, schizophrenia and William’s syndrome. Evidence is growing that there might be dedicated neural circuitry that involves a number of areas in the right frontal lobe, certain portions of the left orbitofrontal cortex and amygdala that is responsible for theory of mind abilities among other functions. Current theories are evolving into constructs that are still relatively modular in nature but with less apparent delineation of functions. However, a purely neurolobiological explanation will not be able to completely explain the deficits seen in theory of mind abilities as autism has definite genetic and biochemical influences.

The clinical implications of recent findings about theory of mind deficits are large. An understanding of the specific mechanisms underlying the manifest behaviour in children with autism will help clinicians devise different levels to intervene on. For example, a large number of available interventions in dealing with communication in children having autism focus on end-product abilities (PECS, Makaton etc.). Social skills training programs also aim at teaching children with autism appropriate skills that would help them respond in an adaptive way when functioning in social settings. These skills, although beneficial, will not be easy to generalise for an individual with autism as he/she would lack the insight into the ‘theory of mind’ of others around him/her. Therefore the only alternative available for autistic individuals would be to memorise a large enough repertoire of interactional skills such that he/she would be prepared for most social eventualities. The easier and more effective option would be to train individuals with autism to develop even a basic insight into theory of mind concepts. However, there is very little literature documenting the use of such programs and there is even less literature documenting the efficacy of and methods that can be utilised.
From a clinical perspective, theoretical evidence is sufficient in support for the ‘theory of mind’ model to start developing and testing specific interventions based on it. Many standardised and reliable variations of false-belief task tests and mental state attribution tests are currently available. These tests can be incorporated in to clinical studies on theory of mind interventions to provide significantly objective results.

Conclusion

For social interaction, the ability to understand other people’s beliefs, motivations, and goals is crucial. ‘Mentalizing’ (Frith and Frith, 1999) or Theory of Mind (Premack and Woodruff, 1978), is at the core of successful interaction with other human beings. Theory of Mind (ToM) is necessary for understanding and predicting other people’s behavior and, in turn, for reacting adaptively to the changing environment. There is a large body of knowledge about the development of this ability in childhood (Wellman, 1993), and it has been shown that autistic children have specific deficits in ToM processing (Baron-Cohen, Leslie and Frith 1985). Recently, a number of neuropsychological patient studies and neuroimaging studies have attempted to further understand the neuroanatomical basis of ToM abilities. Cortical regions shown to be particularly important for ToM processes are the frontomedian cortex and the amygdala. Increased understanding has brought research to a point where theoretical evidence can be translated into specific interventions. In terms of a clinical perspective, future research should aim at designing appropriate intervention programs for children with autism based on the ‘theory of mind’ model.

References:

Baron-Cohen, S., (1995). Mindblindness: an essay on autism and theory of mind / Simon Baron-Cohen. Publisher: Cambridge, Mass.: MIT Press.

Baron-Cohen, S., Jolliffe, T., Mortimore, C., Robertson, M. (1997). Another advanced test of theory of mind: Evidence from very high functioning adults with autism or Asperger syndrome. Journal of Child Psychology and Psychiatry, 38 (7), pp. 813-822.

Baron-Cohen, S., Leslie, A.M., Frith, U. (1985). Does the autistic child have a "theory of mind"? Cognition, 21, pp. 37-46.

Baron-Cohen, S., Ring, H., Moriarty, J., Schmitz, B., Costa, D., Ell, P. (1994). The brain basis of theory of mind: the role of the orbitofrontal region. British Journal of Psychiatry, 165, pp. 640–649.

Baron-Cohen, S., Tager-Flusberg, H., Cohen, D.J. (Eds.), (1993). Understanding Other Minds: Perspectives From Autism. Oxford University Press, Oxford.

Brothers, L., Ring, B., Kling, A., (1990). Response of neurons in the macaque amygdala to complex social stimuli. Behavioral Brain Research. 21, 41(3), pp. 199-213.

Fletcher P.C., Happe, F., Frith, U., Baker, S.C., Dolan, R.J., Frack-owiak, R.S.J., Frith, C.D. (1995). Other minds in the brain: a functional imaging study of ‘theory of mind’ in story comprehension. Cognition 57, pp. 109–28.

Frith, C. D., Frith, U. (1999). Interacting minds, a biological basis. Science, 286, pp. 1692–1695.

Frith, U., (1987). Metarepresentation and autism: how not to lose one's marbles. Cognition, 27(3): pp. 291-294.

Frith, U. (1989). In C. Gillberg's Diagnosis and Treatment of Autism (Ed.) Plenum Press; New York.

Fodor, J.A. (1983). The Modularity of Mind. Cambridge, MA: MIT Press

Gallagher, H.L., Happe, F., Brunswick, N., Fletcher, P.C., Frith, U., Frith, C.D. (2000) Reading the mind in cartoons and stories: an fMRI study of ‘theory of mind’ in verbal and nonverbal tasks. Neuropsychologia 38, pp. 11–21.

Griffin, J. (1999) Autism: A Sea Change. The New Therapist 6 (4) retrieved on 22 November, 2002 from http://www.radicalpsychologytv.org/autism.html (electronic source).

Happe, F., Brownell, H., Winner, E. (1999). Acquired ‘theory of mind’ impairments following stroke. Cognition 70, pp. 211–240.

Karmiloff-Smith, A., Klima, E., Bellugi, U., Grant, J., Baron-Cohen, S., 1995. Is there a social module? Language, face processing and theory of mind in individuals with Williams Syndrome. Journal of Cognitive Neuroscience, 7, pp. 196–208.

Leslie, A.M., Frith, U., (1990). Prospects for a cognitive neuropsychology of autism: Hobson's choice. Psychological Review, 97(1): pp. 122-131.
,
Premack, D., Woodruff, G., (1978). Chimpanzee problem-solving: a test for comprehension. Science, 202 (4367): pp. 532-535.

Premack, D., & Woodruff, G. (1978). Does the chimpanzee have a "theory of mind?" Behavioural and Brain Sciences, 4, pp. 515-526.

Saver, J.L., Damasio, A.R., (1991). Preserved access and processing of social knowledge in a patient with acquired sociopathy due to ventromedial frontal damage. Neuropsychologia 29, pp.1241–1249.

Wellman, H. M. (1993). Early understanding of mind: The normal case. In Understanding Other Minds: Perspectives from Autism (S. Baron-Cohen, H. Tager-Flusberg, and D. J. Cohen, Eds.), pp. 10– 39. Oxford Univ. Press, Oxford.

Wimmer, H., Perner, J., (1983) Beliefs about beliefs: representation and constraining function of wrong beliefs in young children's understanding of deception. Cognition, 13(1), pp.103-108.

Wing, L., (1986). Clarification on Asperger's syndrome. Journal of Autism and Developmental Disorders, 16(4), pp. 513-515.