Abstract
Leo Kanner introduced the syndrome of early infantile autism to the scientific literature in 1943 with his paper, ‘Autistic disturbance of affective contact’ [1]. Kanner's paper described 11 children who seemed to relate better to objects than people. Kanner also noted that if language eventually emerged in these children, it was characterized by echolalia, pronoun reversal and concreteness. In the 1950s autism (like schizophrenia) was thought to be caused by ‘bad’ parenting, for example, the so-called ‘refrigerator parents’ [2]. Early opinion also held that autism was an early manifestation of schizophrenia, and thus was often referred to as ‘schizophrenic syndrome of childhood’ or ‘childhood psychosis’ [3] p.170].
One year after Kanner defined ‘autism’, Hans Asperger published a description of children with the condition ‘autistic psychopathology’. In Asperger's original description, he described a child who was precocious in learning to talk and often talked in a pedantic way about a topic of particular, circumscribed interest (Asperger, 1944 as cited in [4]). Asperger also observed that these children produced stilted and repetitive speech, which appeared to lack intonation. He noted that they were interested in social relationships, but lacked the ability to understand the rules of social behaviour. Asperger noted that his group of patients moved in a ‘clumsy’ way (Asperger, 1944, as cited in [5]). Despite the many similarities between Kanner's and Asperger's patient groups, Asperger disagreed that his disorder was a variant of Kanner's autism [6]. The condition described by Asperger was introduced to English speaking countries by Wing in 1981 [7]. Asperger's disorder did not appear as a separate disorder in standard diagnostic manuals until version IV of the DSM series [8], and version 10 of the ICD series [9].
The controversy about whether autistic disorder (which will be referred to as ‘autism’ henceforth) and Asperger's disorder merely exist on a symptom continuum, or are, in fact, distinct disorders with separate neurobiological underpinnings, is possibly as old as the diagnostic categories themselves. One contemporary view on the relationship between autism and Asperger's disorder holds that Asperger's disorder lies on a ‘disorder continuum’ with autism, at the less impaired end of the clinical spectrum [10–12]. This view has been challenged in recent years by an alternative view based on clinical and neurobehavioural evidence which suggests that autism and Asperger's disorder should be recognized as separate clinical entities as they are in DSM-IV [8] and ICD-10 [13], rather than as variants of a single disorder. The aim of this paper is to review current diagnostic criteria, epidemiological information, genetic and neurobehavioural evidence which supports the notion that autism and Asperger's disorder are separate clinical entities.
DSM-IV diagnostic criteria
Current DSM-IV criteria describes autism as a pervasive developmental disorder defined by impairments in social and communication function, and repetitive and stereotyped behavioural patterns. Areas of social disturbance include poor eye-contact, reduced ability to interpret emotional states, failure to develop peer relations and deficiencies in social-emotional reciprocity [8]. Core communication deficiencies include either a delay or total lack of expressive language, as well as a marked impairment in non-verbal behaviour. In cases where individuals develop language, they often have problems in initiating or sustaining conversations [8]. Developmentally inappropriate echolalia and pronoun reversal, reduced prosody and intonation, and impairments in the use of figurative language are also associated with autism [14]. Restricted, repetitive and stereotyped behaviours manifest as an intense preoccupation with a single subject or activity, adherence to non-functional routines or rituals, stereotypies and motor mannerisms, for example, hand or finger flapping, and body rocking [8]. Ornitz [15] describes hand flapping as ‘involving a rapid flexion and extension of the fingers or hand, or an alternating pronation and supination of the forearm’ (p.311). Disturbances in sensory modulation are also observed in individuals with autism, for example, under and over-reactivity to sensory stimuli [15].
For a diagnosis of autism, a child must, before 3 years of age, exhibit abnormal or delayed functioning in at least one of the following areas of social interaction: social use of language, symbolic play or imaginative play. Two subtypes of autism are generally referred to in the literature, namely, low-functioning and highfunctioning types (note that the low and high functioning autism distinctions are not made in DSM-IV). The lowfunctioning group is defined by the presence of intellectual disability (i.e. IQ below 70) and associated with an increased incidence of an acquired or genetically determined biological cause; for example, perinatal exposure to rubella, thalidomide, herpes encephalitis, tuberous sclerosis, fragile-X, Angelman or Cornelia de Lange syndromes [16]. Conversely, the high-functioning group (or idiopathic form) is associated with relatively intact cognitive functions (i.e. IQ above 70), and the absence of identifiable brain damage, neurological findings or biological markers [17]. Approximately 25% of individuals with autism are diagnosed with the high-functioning, idiopathic form of autism [18].
If a child meets the criteria for social and repetitive stereotyped behavioural disturbances, but does not exhibit disturbances in communication before 3 years of age, a diagnosis of Asperger's disorder is given (see [5] for a review of the DSM classification of autism Asperger's disorder criteria). Essentially, high-functioning autism is clinically distinguished from Asperger's disorder by the presence of clinically delayed language [8]. Diagnosis often occurs much earlier in autism (mean age at the time of diagnosis = 5.5 years) than for Asperger's disorder (mean age at the time of diagnosis = 11 years), although parents generally become aware of developmental problems much earlier, for example 18 months for autism, and 30 months for Asperger's disorder [19]. This difference may reflect differential times at which the two types of diagnostic behaviour (e.g. the onset of language development) would normally unfold.
While Asperger's disorder can be clearly differentiated from autism when the child has an intellectual disability, the clinical distinction between highfunctioning autism, when the child has an IQ in the normal range, and Asperger's disorder is much less clear. For example, in the absence of an extensive developmental history detailing language and social development, two normally intelligent young people, one with autism, the other with Asperger's disorder, both presenting with socialization problems and a history of repetitive, stereotyped behaviour, would be theoretically highly similar according to DSM-IV [8] criterion. This dilemma has fuelled the debate as to whether highfunctioning autism and Asperger's disorder are variants of a single disorder, or whether each has a distinct neurobiology and aetiology [20–22].
Reviewing the DSM-IV differentiation between highfunctioning autism and Asperger's disorder, Kugler [5] observed that some researchers and clinicians support the separation of autism and Asperger's disorder, while others argue that there is insufficient evidence currently available to support differential diagnosis. Kugler [5] makes the important point that while current diagnostic criteria (i.e. DSM-IV, ICD-10) have substantially increased the reliability of the diagnosis of pervasive developmental disorders (i.e. autism and Asperger's disorder), ‘evidence of its validity may need to await a much greater understanding of aetiologies and pathogenesis’ (p.23) (see [5] for a review of diagnostic reliability of high-functioning autism and Asperger's disorder, as well as similarities and differences in symptomatology).
Epidemiological comparison of autism and Asperger's disorder
Based on a meta-analysis of 23 epidemiological surveys of autism published in English between 1966 and 1998, Fombonne [23] calculated the median prevalence of autism to be 5.2: 10 000, and estimated a prevalence of 18.7: 10 000 for all forms of pervasive developmental disorder. Gillberg [21] reported an epidemiological comparison of autism (i.e. low & highfunctioning), high-functioning autism alone, and Asperger's disorder. This author reported that while autism, in general, occurs in 7–16 per 10 000 children, high-functioning autism was estimated to occur in less than 5 in 10 000 children in the general population. Asperger's disorder was found to be more prevalent than high-functioning autism; the prevalence was estimated at 36–71 in 10 000 children aged between 7 and 16 [21].
Gillberg [24] reported that the male: female ratio in autism is most likely 2–3: 1. Based on a population study, Ehlers and Gillberg [25] estimated the ratio of males to females was higher in Asperger's disorder, than autism, 4: 1.
Genetic comparison of autism and Asperger's disorder
There is sufficient evidence to suggest that genetic factors play a major role in the pathogenesis of autism and Asperger's disorder [26]. The finding that monozygotic twins or triplets show less than 100% concordance for autism/Asperger's disorder, and different patterns of severity [27, 28], suggests the involvement of multiple genes [29], and the additional presence of environmental factors [16].
Gillberg [30] notes that autism has been linked to anomalies in all but two chromosomes. Rutter [3] suggested that partial tetrasomy on chromosome 15 may emerge as a potentially significant factor in the pathogenesis of autism. Gillberg and Coleman [18] describes autism as reflecting extreme aspects of male behaviour and postulated a relative decrease in the influence of the female chromosomes (XX) and a corresponding increase in the influence of the male (Y) chromosome. Szatmari et al. [31] suggested that an X-linked serotonin transporter gene may be the mechanism for increasing susceptibility to autism.
Research has shown that siblings of individuals with autism have a 3–8% chance of being diagnosed with the disorder (cf. 0.16% in the general population) [32]. Sib-pair studies indicate that 3–4 genes on the X chromosome may explain the 75-fold increase in risk that a child has of being autistic if a sibling has autism [33]. This pattern of transmission is also consistent with the involvement of multiple genes, rather than the involvement of a particular recessive or dominant gene.
Interestingly, relatives of individuals with autism often present with some features of autism (e.g. language disorders; social problems; repetitive, stereotyped behaviours), but do not meet the full criteria for autism, the so-called autism phenotype [34]. Folstein [35] examined cognitive performance, intelligence, reading, spelling and pragmatic language, in the parents and siblings of a large sample of children with autism, and in the parents and siblings of Down syndrome children. They found that relatives of individuals with autism reported a greater history of early language related cognitive difficulties, in comparison to Down syndrome relatives. Using a similar methodology, Fombonne [36] found lower mean verbal IQ scores in first-degree relatives of individuals with autism, and more specifically, reported that the siblings, affected with the autism phenotype, had significantly poorer literacy skills and lower IQ, than unaffected siblings.
Just as individuals with autism and Asperger's disorder tend to overlap with other neurodevelopmental disorders involving disruption within the basal-gangliathalamocortical (BGTC) circuitry (see [37] for a review), the families of individuals with these disorders also show a higher incidence of BGTC circuitry disorders such as obsessive–compulsive disorder, major depression and bipolar disorder [17]. Bradshaw [37] notes that high rates of OCD have been reported in the first-degree relatives of individuals with autism and Asperger's disorder. Bolton [38] found that motor tics, obsessive–compulsive and affective disorders (e.g. major depression) were the most commonly reported psychiatric disturbances in people with an autistic relative. These authors also found that relatives presenting with OCD symptomatology were more likely to exhibit autistic-like social and communication impairments in comparison to relatives presenting with other psychiatric disorders.
Past research has shown that individuals with Asperger's disorder perform better than matched controls on a visual-perceptual task, the Embedded Figures Test [39]. Using an adult version of this test, Baron-Cohen and Hammer [10] found a similar visualperceptual advantage in the parents of children with Asperger's disorder. This finding suggests that anomalies other than social and language deficiencies may also be inherited in the autism phenotype.
Gillberg [21, 40] observed that there were more ‘Asperger's-like’ features in the relatives of children with Asperger's disorder, than there were ‘autistic-like’ features in the relatives of children with autism, and suggested that a positive family history (of autism or Asperger's) is more typical for Asperger's disorder than high-functioning autism. Similar observations were made by Volkmar, Klin and Pauls [41] who suggested that genetic factors appear to play a more significant role in Asperger's disorder than in autism. Gillberg [21] argued that Asperger's disorder may be a predominantly genetic disorder, while autism may be caused by a combination of genetic and brain injury (i.e. environmental) factors. Rather than an additional brain insult occurring in the aetiology of autism, Szatmari [31] suggested that different gene combinations may distinguish high-functioning autism from Asperger's disorder. Both of these suggestions imply a different neurobiological aetiology for each of these disorders.
While neuroimaging and pathology studies have indicated that autism and Asperger's disorder are likely to involve multiple neural circuitry dysfunction, in particular, the basal-ganglia-thalamocortical circuitry, as well as the cerebellum, hippocampus, and limbic regions (see [37] for a review of the literature), there is little evidence to indicate either disorder is associated with gross neurological damage.
Clinical and neurobehavioural comparisons of autism and Asperger's disorder
Happe and Frith [42] suggest that individuals with Asperger's disorder tend to be less aloof, more socially interested and to have more pedantic language than individuals with autism. While individuals with autism have been described as living in a ‘world of their own’, as if others do not exist, individuals with Asperger's disorder have been described as living in ‘our world in their own way’ [5]. Bradshaw [37] p.204] notes that individuals with Asperger's disorder are ‘usually scrupulously law-abiding’, but ‘may be sensitive to perceived slights, disruptive, or even antisocial, especially if they think their special interests are at risk’.
The distinction between autism and Asperger's disorder is further supported by studies examining language [43], psychopathology [44, 45] and theory-of-mind ability [46]. These studies show that individuals with Asperger's disorder exhibit more pedantic speech patterns and have greater levels of psychopathology, but perform better on theory-of-mind tasks, in comparison to individuals with autism.
Theory-of-mind describes the ability to ‘attribute independent mental states to self and others in order to explain and predict behaviour’ [47] p.116]. Volkmar [48] notes that theory-of-mind fails to account for the clinical phenomenology of autism because the social deficits characteristic of autism appear at a point in development even before normally developing children demonstrate the acquisition of a theory-of-mind. The notion that theory-ofmind ability is not central to the behavioural deficiencies of autism/Asperger's disorder is also suggested by Dahlgren and Trillingsgaard's [49] study which showed that children with autism and Asperger's disorder, who have normal intellectual functioning, perform similarly to normal, healthy controls and to each other, on theory-ofmind tasks (although see [50–52] for conflicting results). Muller et al. [53] have suggested that, although neuroimaging studies have revealed atypical fronto-striatal activation when individuals with autism/Asperger's disorder perform theory-of-mind tasks (see [50, 54]), these findings may be secondary to more fundamental irregularities in neural connectivity. Volkmar [48] argued that, rather than examining the inability to interact and process at a social level, perhaps a greater understanding can be gained from examining the non-social, cognitive-motor, and perceptual deficiencies associated with these disorders. Indeed, such neurobehavioural studies, (e.g. studies of executivefunction, lateralization, visual-perceptual and motor ability) have identified dissociations between individuals diagnosed with high-functioning autism, and those diagnosed with Asperger's disorder.
Executive functioning
Manijiviona and Prior [11] compared a large sample of children with autism and Asperger's disorder using a neuropsychological battery designed to assess intelligence, brain lateralization, and executive function. Executive function tests included the Tower of London test and the Controlled Oral Word Association test. These authors concluded that individuals with autism could not be distinguished from those with Asperger's disorder on any of the executive function or laterality measures. The most significant finding was that the Asperger's disorder group had a higher IQ than the autism group.
Szatmari et al. [55] used a battery of tests which included: an intelligence test, auditory comprehension and memory test, verbal problem solving test, facial recognition test, visual-motor functioning test, a test of manual speed and dexterity and the Wisconsin Card Sort Test (WCST). Only the WCST test discriminated between the two groups indicating that autism, but not Asperger's disorder, was associated with set-shifting deficits. However, Ozonoff, Rogers and Pennington [56] found no difference in set-shifting performance between autism and Asperger's disorder using the WCST.
Our recent series of experiments have indicated that executive functioning, in particular, inhibitory deficiencies, are quantitatively and qualitatively different in autism and Asperger's disorder [57, 58]. Rinehart et al. [58] used a visual-spatial Stroop task where inhibition was invoked by manipulating either: spatial congruency, (respond to the side in which the arrow is pointing and inhibit the more automatic response of responding to the side in which the arrow appears); or conditionality, (respond to the same side in which the arrow appears if an ‘equals sign’ appears, respond to the opposite side of the arrow if a ‘cross’ appears); or both (i.e. manipulations of spatial congruency and conditionality). Individuals with autism had difficulty inhibiting cognitivemotor responses at increasing levels of task complexity. The pattern of cognitive responsiveness observed in the autism group is similar to that seen in Tourette's disorder [59], another neurodevelopmental disorder with frontostriatal involvement. In contrast, individuals with Asperger's disorder performed similarly to age and IQ matched controls. It was noted that a combination of inhibitory and set-shifting deficits may have accounted for performance deterioration in the autism group. Interestingly, past researchers have emphasized only setshifting, but not inhibitory, deficiencies in autism [60]. A dissociation in executive ability between autism and Asperger's disorder was also uncovered by Rinehart et al. [57] using a local-global set-shifting paradigm; while the autism group was significantly slower at shifting attention from the local to the global features of a numerical configuration, the Asperger's disorder participants had no such difficulty [57]. Deficiencies in set-shifting and inhibitory function are consistent with anomalies within the dorsolateral prefrontal and lateral orbitofrontal circuitry. Indeed, damage to both of these frontostriatal circuits can result in clinical features associated with autism; for example, patients with acquired lesions within the dorsolateral prefrontal circuitry exhibit problems with generating unique ideas and coping in novel situations, while lesions within the orbitolateral circuitry are associated with deficiencies in inhibiting previous output, social withdrawal, and obsessive–compulsive symptoms [61, 62].
Lateralization
Fein et al. [63] suggested that subtypes of autism may display characteristic modes of left hemisphere impairment. Unlike autism, Asperger's disorder is not associated with early language delays and has often been imputed to right, rather than left hemispheric dysfunction. While there is neurobehavioural evidence to support a right hemisphere deficit in Asperger's disorder [64], neuroimaging has uncovered evidence to support both left and right hemisphere abnormalities [54, 65]. Dissociations in hemispheric functioning in autism and Asperger's disorder may point to neurobiological differences between these closely related disorders. Rinehart et al. [66] recently examined laterality in a group of children with high-functioning autism, and a group with Asperger's disorder. Two laterality tasks were employed; the first involved controlling movement skilfully along an unpredictable light path in a serial choice reaction-time task, and the second involved responding in a simple reaction-time paradigm to laterally located stimuli in a Posner-type task. While the serial choice reaction-time task was used to determine whether participants were slower or made more errors when required to sequence rightwards movements, the Posner-type task [67] was used to examine whether individuals with autism and Asperger's disorder tended to ‘neglect’ visual stimuli (i.e. cues) presented in the right hemispace. This was measured by examining the number of times participants responded to cues before targets appeared in each hemispace. It was predicted that individuals with autism would ‘neglect’ cues presented in right-hemispace, and would thus be less likely to execute an anticipation response on this side. Compared to controls, the autism group displayed deficiencies in righthemispace and thus in left hemisphere performance on both of these tasks. The Asperger's disorder group performed similarly to their matched controls on all three tasks, thus showing laterality patterns typical for their age and IQ. On the basis that Asperger's disorder is, by clinical definition, not associated with the clinically impaired language delays that characterize autism, it was not entirely surprising to find that this group did not show left hemispheric impairment. Rinehart et al. [66] proposed that the period where dominance shifts from the right to the left hemisphere is important in whatever process might dictate the emergence of either autism or Asperger's disorder.
Visual-perceptual processing
Manijiviona and Prior [11] reported that the cognitive profiles of both groups feature a relative strength in block design (a visual-spatial task). Ropar and Mitchell [68] reported similar performance ability on a visual illusion tasks in autism and Asperger's disorder, with both groups performing similarly to controls (although see [69]). Autistic savantism, which is thought to result from overly focused attention to detail, has also been reported in both groups [70–72]. Jolliffe and Baron-Cohen [39] reported that individuals with autism and Asperger's disorder performed similarly on Embedded Figures tests; for example, both groups were better at disembedding figures (local detail) from within a complex global shape.
A superior ability on Embedded Figures tests has been found among the parents of children with Asperger's disorder [10]; superior spatial abilities have also been demonstrated in the siblings of children with autism [73]. Together these findings indicate that visualperceptual anomalies may be part of the autism/ Asperger's disorder phenotype [10]. Indeed, Rinehart et al. [74] found that individuals with high-functioning autism and Asperger's disorder exhibit a similar ‘lack of global precedence’ when responding to a large or ‘global’ number composed of incongruent, smaller or ‘local’ numbers. It was suggested that a ‘lack of global precedence’ may be underpinned by a more fundamental inhibitory deficit in these disorder groups [74]. To illustrate, while in normal, healthy individuals the semantic and physical processing of an object (i.e. local processing) is usually terminated once it has been interpreted globally [75], this may not be the case in individuals affected by autism and Asperger's disorder; for example, these individuals may have difficulty inhibiting the processing of redundant ‘detail’ even though the global image has been identified [74]. Interestingly, Rinehart et al. [74] found that the autism and Asperger's disorder groups were dissociable on this visual-perceptual task on the basis that the autism group committed more global response errors [74]. It was suggested that this difference in neurobehavioural profile may be due to a more impaired global processing system in individuals with autism compared to Asperger's disorder [74]. It is possible, that while both autism and Asperger's disorder may be associated with a problem inhibiting ‘detail’ (hence lacking global precedence), this may lead to a functional global processing impairment only in autism – possibly because autism is associated with greater executive impairment in general (see ‘Executive functioning’).
Motor functioning
Historically, motor functioning in autism has been studied in the context of gait [76–80]. More recent studies examining motor functioning in autism have focused on motor clumsiness. The focus on this aspect of movement comes from clinical observations that suggest clumsiness is a feature which might distinguish Asperger's disorder from autism [81]. Motor clumsiness is referred to in the ICD-10 [9] as a symptom often found in Asperger's disorder; however, no mention of motor clumsiness is made in the clinical description of autism. No mention of clumsiness is made in the DSM-IV for autism or Asperger's disorder [8]. In reviewing the literature on Asperger's disorder, Ghaziuddin, Tsai, and Ghaziuddin [82] found that approximately 50% of publications referred to clumsy, uncoordinated movement patterns in either single case studies or group studies of children with Asperger's disorder. Gillberg [40] p.528] observed that individuals with Asperger's disorder ‘appeared to be generally clumsy’, had a ‘stiff or awkward way of walking (often without arm-swing)’ and were ‘uncoordinated in posture and gesture’. It is interesting to note that this clinical description of Asperger's is similar to Maurer and Damasio's [79] p.198] earlier ‘parkinsonian’ observations of movement in autism, for example, ‘odd posture when standing’, and ‘diminished associated movements of the arms (during locomotion)’. Indeed, problems with preparing movement, another ‘parkinsonian’ feature [83], have also been observed in children with autism and Asperger's disorder while engaged in a motor re-programming task [84]. Rinehart et al. ′ s [84] study was important in demonstrating that the nature of motor preparation anomalies differs between groups; for example, while an atypical deficit in motor preparation was found in Asperger's disorder (e.g. being slower at preparing for a predictable movement after an unpredictable movement had been successfully executed), in the autism group movement preparation was characterized by a ‘lack of anticipation’ (e.g. not showing a motor preparation time advantage when movements were predictable) [84]. The differences in movement preparation profiles in these disorders was suggested to reflect differential involvement of the fronto-striatal region, in particular the supplementary motor area and anterior cingulate.
A number of research studies have challenged the view that autism and Asperger's disorder can be distinguished on the basis of motor impairment, and contend that both disorders are associated with clinically significant deficiencies in fine and gross motor function (see [85, 86]). Neuroimaging evidence critical to this debate is beginning to emerge in the literature. Recent FMRI data has revealed that individuals with autism exhibit less pronounced activation in the primary motor cortex and supplementary-motor-area (SMA) during a simple finger tapping task [53]; this ‘parkinsonian’ pattern of decreased supplementary-motor-area activity (see [87]) has also been observed in our recent movement-related-potential (MRP) study with autistic children [Rinehart, Tonge, Bradshaw, Iansek, Enticott, in preparation]. Using the MRP experimental technique described by Cunnington [87], MRP's associated with sequential movements were investigated in 16 children with high-functioning autism and 4 age, sex and IQ matched normally developing children. The MRP's for the control children revealed similar SMA activation patterns for internally (non-cued) and externally (cued) determined movements to Cunnington et al.'s [87] larger young control group. In the children with autism, however, the SMA area was only involved in movements that must be internally determined, but not externally cued movements. The pattern of results observed for autism parallel those reported for Parkinson disease [87]. Cunnington et al. [87] interpreted this pattern of results as indicating impaired internal control mechanisms, operating via the SMA. We are currently conducting comparative MRP studies with children diagnosed with Asperger's disorder.
Conclusions
While current DSM-IV [8] criteria define autism and Asperger's disorder as separate clinical entities, individuals diagnosed with Asperger's disorder continue to be conceptualized as having a ‘milder’ variant of autism; furthermore, the terms Asperger's disorder and highfunctioning autism are frequently used interchangeably. There is some epidemiological and genetic evidence that supports the uniqueness of these disorder groups. In the absence of identifiable and uniquely defined neurological damage, neurobehavioural studies have provided an important source of information about the potential neurobiological dissociation that may exist between autism and Asperger's disorder. While there are broad overlaps in the clinical (e.g. social dysfunction) and neuropsychological (e.g. visual-perceptual processing anomalies) features associated with high-functioning autism and Asperger's disorder, the accumulating research that identifies different disturbances in executive functioning, lateralization and motor ability would suggest that it is premature to rule out the possibility that autism and Asperger's disorder may be clinically, and possibly neurobiologically, separate. As Kugler [5] points out, if only the broad overlaps between high-functioning autism and Asperger's disorder are considered, and other clinical and neurobehavioural differences ignored, we may run the risk of obscuring ‘differences that might have very real implications for causation, course, response to intervention and outcome’ (p.23).
Footnotes
Acknowledgements
This research was funded through Cure Autism Now's Young Investigator fellowship programme, with resources provided in part by the Autism Coalition for Research and Education. This research was also supported by the Australian Research Council, Soroptimists International, and The Queens Trust for Young Australians.