Question:

Cognitive theories do not provide a full explanation of autism spectrum conditions. Discuss.

Answer:

Introduction

Many theories have been developed to explain autism spectrum conditions (ASC). During the 1970s and 1980s, psychologists started developing theories with a more cognitive focus (Bowler 2007, p. 184). Apart from the cognitive connection, researchers agree that the ASC has a genetic basis. The three main cognitive theories that have been developed to explain ASC include the Theory of Mind (ToM), Weak Central Coherence, and Executive Dysfunction (ED).

The main problem with these three cognitive theories is that they do not fully explain ASC. The theories have continued to dominate psychological approaches to autism research. In determining the suitability of the theories to explain autism, the main issues at hand relate to universality, uniqueness, and specificity (Rajendran & Mitchell 2007, p. 224). The issue of universality relates to whether the factors identified are found in every autism victim or just a majority of the sufferers. The uniqueness threshold is about whether the factor identified applies to other developmental disorders or it is unique to ASC. In terms of specificity, the issue under consideration relates to the choice between a domain-specific factor and multiple factors (Rajendran & Mitchell 2007, p. 225). This paper sets out to critically examine each of these three theories, pointing out the ways in which each of them fails to explain autism spectrum conditions. The theories have been developed after numerous research efforts, and subsequent studies have led to changes.

The Theory of Mind (ToM)

The theory of mind was developed by Premack & Woodruff (1978, p. 521). The theory addresses the ability to attribute mental states such as beliefs, thoughts, and desires to others as well as to explain their behavior. In the theory of mind, these mental states are representations of the world as opposed to reflections of it (Bowler, 1997, p. 58). In children who develop in the typical manner, an innate cognitive module is viewed to be developing during early childhood. Theorists within the ToM model suggest that this module is normally impaired in people with ASC, leading to several impairments.

In essence, ToM states that people with autism are normally unable to impute mental states to others as well as to themselves (Happé 1995, p. 853). In this regard, ASC victims are unable to, or fail, to take into account the mental states of other people. The theory is normally tested on the basis of the between false belief and true belief. In the testing process, participants suffering from ASC are normally expected to transfer false belief. However, the main shortcoming of this test is that not all children with ASC fail the test. In a test carried out by Baron-Cohen (1985, p. 41), 20% of the children with autism did not fail the test as the theorist had expected, meaning that the theory could not be used to account for ASC in all affected individuals. The 20% that was unaccounted for in Baron-Cohen’s (1985, p. 42) test made the mental deficit appear as if it was not universal.

Proponents of the Theory of Mind have sought to explain the lack of universality in results of tests on individuals with ASC in several ways. In the first, the theorists point out that multiple skills are required to pass the test of false belief; although some ASC individuals pass this test, they do this in a manner that is completely different from that of people who do not suffer from autisms. In essence, passing or failing a test is not considered very important; what is most important is the approach adopted by the participant in tackling the test. Another argument is that the test of false belief cannot be passed by an individual who does not have a representational theory of mind. The theorists oppose this argument because it makes the ToM hypothesis untenable because of its inability to account for the possibility that some people pass the test when autism is defined as impairment in the individual’s theory of mind (Klin 1989, p. 341). In light of these two arguments, one may conclude that although the ToM hypothesis explains well some of the cognitive impairments inherent in autism, it fails to explain the disorder convincingly in all its manifestations.

In light of these shortcomings of the ToM, Baron-Cohen (1989, p. 291) modified it by asserting that the problem with the theory of mind had to do with a delay and not a deficit. Baron-Cohen (1989, p. 291) illustrated this by designing a second test that was more difficult, though founded on the provisions of the previous one. None of the children with autism passed this test, and 90% of all the typically-developing children passed it. Baron-Cohen’s (1989, p. 292) conclusion was individuals with autism could not pass the second-order task because their theory of mind was not fully representational.

Baron-Cohen’s (1989, p. 289) idea of delay in mental development has been criticized a lot as well. Bowler (1992, p. 882), one of the critics, challenged this view by carrying out the second-order test using individuals with Asperger Syndrome. Seventy three percent of these individuals passed this more difficult false belief task. They brought to the fore the question of how autism could be considered an impairment in one’s theory of mind, yet some people with a clear ASC profile passed both the first- and second-order tests. Indeed, the ability by some individuals with autism to pass both tests has been used by many researchers as proof that a deficit in the theory of mind is not a universal trait among all individuals with autism.

Many cognitive psychologists have opted to support the argument on developmental delay. One such case of support has tended to involve the association between the performance in false belief task and in the verbal mental age in children with ASC. In experiments of this nature, children whose verbal mental age was 12 were certain that they would be able to pass. In terms of timing, this success was a representation of a huge delay in comparison to children whose development process was going on in a typical manner. For such children, success would be achieved by the time they were four years old (Wellman 2001, p. 662).

It should be borne in mind that there is a high degree of association between verbal mental age and an individual’s task performance. It is for this reason that belief attribution tasks fail to reveal the nature of impairments that well-performing individuals with higher-functioning cases of autism have. This problem has compelled researchers to design advanced tests aimed at testing the theory of mind among this well-performing population. Some of these tests include the Strange Stories test, the Faux Pas Recognition Test, and the Eyes Task.

The strange stories test was designed by Happé (1994, p. 129) and it comprised of 24 stories aimed at providing a challenge to autistic individuals from a very naturalistic perspective. This explains why the stories dwelt on simple events that take place on a daily basis. These events were about the things that motivate people to utter things that are not literally true. Happé’s (1994, p. 134) finding was that the degree of success among autistic individuals was not evenly distributed. The test greatly discriminated between autistic individuals who had previously failed tasks on the theory of mind, those who had passed the theory of mind tasks of the first-order, and those who had passed the tests of the second-order. Moreover, even the autistic individuals who passed the second-order tests on the theory of mind ended up providing incorrect justifications of mental states for some of the stories. This is unlike in the case of neuro-typical adult participants, for whom no such errors were recorded.

Indeed, the inconsistencies of the strange stories test are an indicator of the disagreements that exists among those who seek to explain autism from a cognitive perspective. In the face of such disagreements, it becomes difficult to explain ASC using cognitive theories. In the case of the strange stories test, a replication of Happé’s (1994, p. 137) test by Jolliffe & Baron-Cohen (1999a, p. 401) did not help matters. In the replication study, two adult groups comprising of autistic individuals were targeted. The individuals in one group had Asperger Syndrome (AS) while those in the other group were autistic individuals with higher-functioning level. In both studies, Jolliffe & Baron-Cohen (1999a, p. 402) found out that individuals in both groups were unable to use mental state terms that were context-appropriate when explaining strange stories. It is for this reason that they were seen to differ from comparison participants.

Controversy on the use of cognitive theories to explain autism extends even into the Eyes Task, which was the next most influential test since the development of the one on strange stories. In this task, individuals were required to infer the mental states of people represented in photographs with only their eye regions being shown. This task was carried out on three adult groups: one with Asperger Syndrome, another with Tourette Syndrome, and the third comprising of normal adults. The participants were required to respond by making a choice from four adjectives. Accurate responses from the adults with autism were fewer than for those who did not have autism.

Moreover, in the Emotion Task of judging photographs of people displaying different basic emotions, the performance of individuals with autism was such that experimenters had to rule out deficits in the ability by autism patients to perform the Eyes Task. This fundamental shortcoming of the Eye Task has called into question the ability by the cognitive approach to explain autism.

In essence, cognitive theorists argue in support of the existence of an innate cognitive module among all individuals (Nadel 2000, p. 138). This cognitive module develops gradually during childhood in all typically growing children. The theorists suggest that the impairment of this module in individuals with ASC causes several impairments. In the case of the theory of mind, the seemingly biggest challenge is proving through experimentation that certain cognitive weaknesses apply universally to all individuals with ASC. The theory’s inadequacy manifests itself in the fact that not all individuals with autism fail the false-belief tasks of the ToM. Nevertheless, proponents of the theory insist that those who pass these tasks sometimes have difficulties in engaging in ‘mind-reading’ on an everyday basis. There is also a gaping weakness in the tendency by children and adults without autism to fail ToM tasks.

The Weak Central Coherence (WCC) theory

The WCC theory explains autism from the perspective of a domain-general process. In this regard, social and non-social features of autism are described as being characterized by attention to acute detail as opposed to the overall meaning (Happé, F, 1996, p. 875). Proponents of this theory argue that for people with autism, the drive for global coherence is either very weak or virtually absent (Rajendran & Mitchell 2007, p. 232). Such people are more concerned with the constituent parts as opposed to the bigger picture.

Researchers interested in the WCC theory have expressed a lot of interest in research on the individuals’ perceptual abilities (Lawson 2011, p. 60). Such research has led to several refinements of the theory. However, the theory continues to suffer from numerous shortcomings because of variations of interpretations. In many tests aimed at determining their perceptual abilities, children with autism have been seen to achieve above-average performance. A case in point is the Children’s Embedded Figures Test (CEFT). In this text, individuals are requested to locate a small shape within a drawing of a larger conventional shape made using confusing lines (Klin 1991, p. 34). For typically developing children, the larger everyday shapes become so captivating that the embedded shapes are not easily detectable. For individuals with autism, these embedded shapes are more noticeable owing to their preoccupation with the constituent parts at the expense of the bigger picture. The explanation for the impressive performance is that the individuals lack the necessary cognitive drive to conceptualize the global form.

However, it is in the analysis of visual illusions that variations in interpretations exist with regard to the WCC theory. In visual illusions, the test was inspired by suspected variations in perceptual and visual-spatial functioning (Rajendran & Mitchell 2007, p. 235). Happé (1996, p. 83), for instance, predicted that individuals with autism are not susceptible to visual illusions. According to Happé’s (1996, p. 83) argument, people normally succumb to visual illusions simply because of their tendency to process all the components of an illusion in an integrative manner. For people with autism, the approach is characterized by a piecemeal perspective with no any element of integration or comparison of various features. However, in a contradictory twist, Ropar (1999, p. 1285) found out that individuals with autism were equally susceptible to both different visual illusions and controls.

To support Ropar’s (1999, p. 1286) findings, Milne (2002, p. 261) points out that the impression created is that global processing in autism is under attentional control, and therefore care should always be taken before making a conclusion in favor of low-level perceptual deficit among autistic individuals. In a different explanation, higher-order processing is considered an optional element among autistic individuals; for people without autism, it is mandatory. This understanding leads researchers to conclude that the nature of perceptual processing among autistic individuals depends on how the question has been worded.

Another contrasting finding relating to visual illusions was arrived at by Brosnan (2004, p. 461), who reported that people with autism could succumb to visual illusions just in the same way as individuals without autism. Brosnan gave the example of a Muller-Lyer illusion, where individuals were asked ‘which line looks longer’. When the question was changed to ‘which line is longer’, autistic individuals did not fall into the trap of illusion. Again, an additional proof was provided that the wording of the question was the key determining factor on whether an autistic individual could become susceptible to visual illusion or not.

The inadequacy of the cognitive approach is also evident in the tendency by researcher to switch from one conceptual dimension to the other. A case in point is the suggestion to replace the WCC theory with the Reduced Generalization theory (Plaisted 2001, p. 156). According to Plaisted (2001, p. 156), superior autistic performance in the Embedded Figures and Block Design (the two main tests used in WCC theory) is better explained as a reduction in the way similarities between stimuli and situations are processed.

Disagreements in the way autism can be explained from a cognitive perspective exist even in the realm of WCC-related hypotheses. For instance, the hypothesis that people with autism are able to process unique features of visual stimuli in a better manner and the everyday features poorly yields two key complementary predictions. The first one is that they ought to be better at all extremely difficult discrimination tasks where the stimuli in question hold numerous elements in common and only a few unique elements. The second one is that they ought to be much poorer in tasks that require two sets of stimuli to be categorized. Plaisted’s theory portends that this is because autistic individuals are better at processing differences between different categories as compared to shared category features.

Owing to the inadequacy of the WCC theory, Mottron & Burack (2001, p. 135) suggested an alternative model known as hierarchization. This model is founded on finding that the way autistic individuals process hierarchical stimuli differs from the way non-autistic individuals do. The frequent shift from one model to the other is a manifestation of the inadequacy of the cognitive perspective in explaining autistic spectrum conditions. Subsequent research studies have only yielded mixed support for the hierarchization model.

Moreover, the other variant of the WCC theory, known as Enhanced Perceptual Processing (EPP), is founded on the weaknesses of the WCC theory. The EPP model postulates that ASC does not have elements of poor integration of Gestalt information, but rather by an enhancement in the way all the individual elements are discriminated. In the EPP account, local processing is viewed as being enhanced; the autistic individuals use a bottom-up approach to process information but they do not do so at the expense of top-down (global) processing (Noens 2008, p. 216).

Since the debate between EPP and WCC is still ongoing, it is not possible to rely on either model to explain ASC. This is especially true considering that so far, both accounts have presented data that support their respective theses. Moreover, the evidence that has been gathered so far in both models has targeted only the high-functioning autistic individuals. The lack of studies on low-performing autistic individuals from the perspective of the WCC theory is a further indicator that cognitive theories cannot be relied upon in explaining ASC. Whenever mention is given to the lower-functioning autistic individuals, the aim does not normally entail settling the debate on EPP and WCC. All that the proponents of these two models can hope for is that further studies can shed light on different aspects of the ASC debate (Noens 2008, p. 221).

The Executive Dysfunction (ED) theory

Executive function encompasses impulse control, mental flexibility, planning, monitoring of action, inhibition, and working memory (Hill 2004, p. 26). There is an ongoing debate on the relevance of the issue of executive dysfunction among autistic individuals. This debate is also of great relevance to other disorders that related to the frontal lobe function. However, cognitive behavioral studies have not quite created a coherent ground for explaining autism spectrum conditions. The studies merely highlight the need for more detailed research on the analysis of the execution function among people with ASC. Conventionally, the best way to do this would be to assess all the relevant executive functions as well as determine their neuro-anatomical correlates across an individual’s lifespan (Hill 2004, p. 26).

There is normally a need for the various aspects of the executive function to disengage from an individual’s immediate environment in the process of guiding actions. For people who have had their frontal lobes impaired, their executive functions get damaged. This damage is associated with such clinical disorders as obsessive compulsive disorder, attention deficit hyperactivity disorder (ADHD), schizophrenia, Tourette Syndrome, and autism spectrum conditions. However, as a way of explaining autism spectrum conditions, the executive dysfunction approach suffers from the fundamental weakness in that sometimes, executive dysfunction can also manifest itself among individuals for whom damage has affected non-frontal brain areas (LeDoux 2002, p. 73).

The account of executive function (EF) which is domain-general arose not from neuro-typical research, but from researchers who observed that certain symptoms of autism exhibited resemblance to those relating to injury to specific parts of the brain. A conventional observation was the difficulty in explaining the symptoms from the perspective of the theory of mind. Some of these symptoms include difficulties in switching attention, an obsession with elements of sameness, inability to control impulse, and the tendency to perseverate.

There is abundance of evidence regarding the important role that the prefrontal cortex plays in executive control. An interesting hypothesis is that the root cause of many behavioral patterns among individuals with autism may be in abnormalities in the prefrontal cortex. The hypothesis is an interesting one because although there is much progress in different areas of ASC research, no common ground is yet to be reached regarding the neural basis of autism. In terms of reliability, though, the explanation that is based on executive function is inadequate since in most cases, not all aspects of executive processes become impaired, and variations in the impact of deformity in the prefrontal cortex cannot be easily tested empirically.

Experimentation has shown that there are two main challenges with regard to the use of the cognitive tenets of the ED theory to explain autism. The first challenge relates to cognitive control and cognitive flexibility. Cognitive control is the ability by humans to enact a specific behavior even when there is a distracting or competing response. Cognitive flexibility, on the other hand, is the human ability to adjust cognitive control fluently as contingencies continue changing. Researchers like using the WCST (Wisconsin Card Sort Test) and Stroop task to measure cognitive control. For measuring cognitive flexibility, the WCST test is used. Autistic individuals normally exhibit poor performance in the WCST test. However, on the Stroop task, they do not show more interference than health controls. This dichotomy presents a key challenge to the idea that autism is the outcome of global impairment in an individual’s executive processes, mediated perhaps by abnormalities in the frontal lobe.

The second challenge has to do with the issue of executive deficits in ASC as viewed from a developmental perspective. Among young autistic children, executive abilities are normally seen to be intact when in comparison with controls for verbal ability and age. This outcome calls into question the role that ED plays in the manifestation and development of autism and related symptoms. The lack of any observable deficits among children with autism is a major blow in the ED theorists’ efforts to explain autism spectrum conditions.

Indeed, the ED approach is one of the most complex approaches used to explain autism from a cognitive perspective. The complexity manifests itself particularly with regard to the investigation of executive functions of autistic individuals. For instance, researchers have to contend with the possible influence of intelligence quotient (IQ) on the executive performance among autistic groups. At the same time, the researchers have to address the possibility of an overlap between test performances on executive function as far as other neuro-developmental disorders involving frontal lobe congenital deficits are concerned. Two key disorders where an overlap may have far-reaching effect on ED research include Tourette’s syndrome and attention deficit hyperactivity disorder.

On the issue of planning, one of the tasks traditionally associated with EF, there are many unclear areas. Careful consideration of these areas indicates that it may be more useful for researchers to focus on the way learning disability influences planning performance. Such focus is necessary before any reliable conclusions can be reached regarding the nature of executive dysfunction in autism. There is also a further difficulty regarding the choice of tasks that can be relied upon in assessing executive function. The so-called Tower tasks, by their very nature, are extremely complex, and many other processes have to be considered over and above the planning activity per se. The individual undertaking the tasks has to engage the working memory as well as seemingly inefficient sub-goal moves. This calls for the need for control tests to be put in place in future studies before the ED approach can be considered sufficient in the explanation of autism. Moreover, planning-related tests tend to have very low test and retest reliability, particularly within the normal population. The tests are not an accurate tool for indexing the maturation or integrity of children’s underlying neurological systems.

Recent work on the theory of mind

In recent times, there have been increased efforts to develop the theory of mind into a more acceptable cognitive model of explaining autism. These efforts have yielded two new theories: the enactive mind theory (Klin et al 2003, p. 345) and the empathizing-systemizing theory (Baron-Cohen 2004, p. 119). The development of these two theories, which drift away considerably from the tenets of the theory of mind, has triggered concerns that it may not be possible to explain the range of symptoms in autistic individuals using a single cognitive theory.

The enactive mind theory focuses on normative-IQ autistic individuals for whom there is a discrepancy between their capability to solve explicit socio-cognitive problems and the ability to meet the demands of day-to-day social situations. The enactive mind approach facilitates the documentation of this discrepancy through newer techniques, for example eye tracking. These new techniques allow researchers to measure and determine how autistic individuals search for meaning whenever they find themselves in naturalistic social scenes. The enactive mind (EM) approach is seen to be critical in the analysis of patho-physiology of autism (Mundy 2000, p. 158). The EM has its foundations in ‘embodied cognitive science’ (Klin 2000, 83). Embodied cognitive science is a neuroscience framework in which cognition is understood in terms of bodily experiences arising out of the adaptive actions of an organism in response to the immediate environment (Newell 1991, p. 269).

Proponents of the EM model, though, err by making the radical assumption that mental representations are simply proxies for the actions that generate them and for which they always stand. In this regard, social cognition is understood as an equivalent of social action. The implication of this assumption is that whenever an individual upholds and manipulates symbols in his mind, he is also evoking an intricate network of experiences arising out of a life history of various actions that have an association with that symbol. Such an assumption is complex, and it makes it difficult for differences to be made between the cognition of individuals with autism and those without it.

On the other hand, the empathizing-systemizing approach, as expounded by Baron-Cohen (2004, p. 167), makes a comparison between the brain functioning among males and females in the general population, vis-à-vis those with autism. Baron-Cohen (2004, p. 167) states that on average, females in the general population have a stronger drive towards emphasizing while males possess a stronger drive to systematize. As for people with autism spectrum conditions, the drive to empathize is below average while they possess an above-average drive to systematize. As such, Baron-Cohen (2004, p. 167) conceptualizes such individuals to possess the extreme version of the typical male brain. From this description, it is clear that while the EM model is overly complex, the empathizing-systemizing approach is too simplistic. This makes them both not a helpful way of explaining ASC.

Conclusion

In summary, this paper has discussed the notion that cognitive theories do not provide a full explanation of autism spectrum conditions. Indeed, there is abundance of literature on cognitive theories of autism, with the three main ones being the Theory of Mind (ToM), Weak Central Coherence, and Executive Dysfunction (ED). It is also clear that a systematic trend is yet to arise with regard to the modification of the existing cognitive theories. Much of recent literature, notably the one relating to the EM and empathizing-systematizing approaches borrows a lot from the theory of mind. By and large, two strands emerge: domain-specific and domain-general approaches, each with varying hypotheses and assumptions. This makes cognitive theories insufficient as ways of explaining autism spectrum conditions.

 

References

Baron-Cohen, S, 1985, ‘Does the autistic-child have a theory of mind’, Cognition, Vol. 21, No. 1, pp. 37–46.

Baron-Cohen, S, 1989, ‘The autistic child’s theory of mind—a case of specific developmental delay’, Journal of Child Psychology and Psychiatry and Allied Disciplines, Vol. 30, No. 2, pp. 285–297.

Baron-Cohen, S, 2004, Mind reading: The interactive guide to emotions, Jessica Kingsley Limited, London.

Bowler, D, 1992, ‘Theory of mind in aspergers syndrome’, Journal of Child Psychology and Psychiatry and Allied Disciplines, Vol. 33, No. 5, pp. 877–893.

Bowler, D, 1997, ‘Asperger’s syndrome and memory: similarity to autism but not amnesia’, Neuropsychologia, Vol. 35, No. 1, pp. 65–70.

Bowler, D, 2007, Autism spectrum disorders: psychological theory and research, John Wiley & Sons, London.

Brosnan, M, 2004, ‘Gestalt processing in autism: failure to process perceptual relationships and the implications for contextual understanding’, Journal Of Child Psychology And Psychiatry, Vol. 45, No. 3, pp. 459–469.

Happé, F, 1994, ‘An advanced test of theory of mind—understanding of story characters thoughts and feelings by able autistic, mentally—handicapped, and normal—children and adults’,  Journal of Autism and Developmental Disorders, Vol. 24, No. 2, pp. 129–154.

Happé, F, 1995, ‘The role of age and verbal-ability in the theory of mind task- performance of subjects with autism, Child Development, Vol. 66, No. 3, pp. 843–855.

Happé, F, 1996, ‘Studying weak central coherence at low levels: children with autism do not succumb to visual illusions: A research note, Journal of Child Psychology and Psychiatry and Allied Disciplines, Vol. 37, No. 7, pp. 873–877.

Hill, E, 2004, ‘Evaluating the theory of executive dysfunction in autism’, Developmental Review, Vol. 24, No. 2, pp. 189-233.

Hill, E, 2004, ‘Executive dysfunction in autism’, Trends in Cognitive Sciences, Vol. 8, No. 1, pp. 26-32.

Jolliffe, T, & Baron-Cohen, S, 1999a, ‘The strange stories test: a replication with high functioning adults with autism or asperger syndrome’, Journal of Autism and Developmental Disorders, Vol. 29, No. 5, pp. 395–406.

Klin, A, 1989, ‘Understanding early infantile autism: an application of G. H. Mead’s theory of the emergence of mind’, London School Economics, Vol. 3, pp. 336–356.

Klin, A, 1991, ‘Young autistic children’s listening preferences in regard to speech: a possible characterization of the symptom of social withdrawal’, Journal of Autism Developmental Disorders, Vol. 21, pp. 29–42.

Klin, A, 2000, Theory of mind in action: developmental perspectives on social neuroscience, Oxford University Press, Oxford.

Klin, A, Jones, W, Schultz, R, & Volkmar, F, 2003, ‘The enactive mind, or from actions to cognition: lessons from autism’, The Royal Society, Vol. 358, pp. 345–360.

Lawson, W, 2011, The Passionate Mind: How People with Autism Learn, Jessica Kingsley Publishers, London.

LeDoux, J, 2002, Synaptic self: how our brains become who we are, Viking Penguin, New York.

Milne, E, 2002, ‘High motion coherence thresholds in children with autism’, Journal of Child Psychology and Psychiatry and Allied Disciplines, Vol. 43, No. 2, pp. 255–263.

Mottron, L, & Burack, J, 2001, ‘Enhanced perceptual functioning in the development of autism’, in J, Burack, (Ed), The development of autism: Perspectives from theory and research, Lawrence Erlbaum Associates, Mahwah, NJ.

Mundy, P, 2000, ‘Neural plasticity, joint attention and autistic developmental pathology’, in L. Glidden (Ed.) International review of research in mental retardation, Vol. 23, pp. 141–168, Academic Press, New York.

Nadel, J, 2000, ‘Do children with autism have expectancies about the social behavior of unfamiliar people? A pilot study using the still face paradigm’, Autism, Vol. 4, pp. 133–146.

Newell, A, 1991, Unified theories of cognition, Harvard University Press, Cambridge, MA.

Noens, I, 2008, ‘The central coherence account of autism revisited: Evidence from the ComFor study’, Research in Autism Spectrum Disorders, Vol. 2, No. 3, pp. 209–222.

Plaisted, K, 2001, ‘Reduced generalization in autism: an alternative to weak central coherence’, in J Burack, (Ed), The development of autism: Perspectives from theory and research, Lawrence Erlbaum Associates, Mahwah, NJ.

Premack, D, & Woodruf, G, 1978, ‘Does the chimpanzee have a theory of mind?’, Behavioural and Brain Sciences, Vol. 4, No. 2, pp. 515–526.

Rajendran, G, & Mitchell, P, 2007, ‘Cognitive theories of autism’, Developmental Review, Vol. 27, pp. 224–260.

Ropar, D, 1999, ‘Are individuals with autism and asperger’s syndrome susceptible to visual illusions?’ Journal of Child Psychology and Psychiatry and Allied Disciplines, Vol. 40, No. 8, pp. 1283–1293.

Wellman, H, 2001, ‘Meta-analysis of theory-of-mind development: the truth about false belief’, Child Development, Vol. 72, No. 3, pp. 655–684.