Autism - Current Issues 47
This set of notes, drawn from recently published material, is largely concerned with the nature and aetiology of autism and ASD … putative causal factors or associated factors, and their interaction.
Reference is made to genetic loci, the link between the size of the amygdala-hippocampal areas of the brain and the patho-physiology of autism and ASD (and between amygdala volume and anxiety), parental age as a predictor of risk, and changes in the expression of autism and ASD over time and the interaction of the various areas of impairments.
The final section is concerned with early predictors of communication development, and with sex differences in the profile and severity of symptoms among young children with autism and ASD.
M.J.Connor April 2007
As a result of a large scale study of children with autism and their families, Satzmari et al (2007) reported that they have identified two further genetic links that contribute to the development of this condition.
This genome project collected genetic material from a sample of 1496 families
(nearly 8000 individuals altogether) in order to seek genetic similarities among those persons diagnosed with autism and ASD.
The research team highlighted a previously un-implicated region of chromosome 11 which they believe to contain a gene associated with the aetiology of autism; and they further reported that the neurexin 1 gene, known to be involved in the release of the neurotransmitter glutamate, and significant for early brain development, may also be implicated in the emergence of autism and ASD.
The latter piece of evidence links to existing findings of a link between anomalous glutamate function and the behaviours characteristic of autism. It is further noted that diagnoses of autism are common among individuals with either fragile-x syndrome or tuberous sclerosis, both of which are associated with disordered glutamate regulation and transmission.
The study will be continued with a view to a precise identification of the relevant gene on chromosome 11 … and the implication that one might draw from this study not only reinforces the already strong and converging evidence that genetic factors play a significant part in autism and ASD, but also that means will become increasingly available to make both diagnoses of autism and ASD which are both early and accurate with the opportunity afforded to initiate supportive and compensatory intervention at an equally early stage.
Further Indications of the Neural Basis for ASD (Amygdala and Hippocampus)
The research paper by Page et al (2006) begins with a recognition that ASD (autism, Asperger syndrome, and pervasive developmental disorders not otherwise specified) are highly genetic neuro-developmental conditions reflected in impaired communication and social reciprocity, and in repetitive stereotyped behaviours.
However, it is also recognised that there does not yet exist a clear understanding of the neural basis of ASD; and, while individuals diagnosed with ASD commonly show functional abnormalities in the limbic region involving the amygdala and the hippocampus, there is no causal explanation available for these abnormalities.
The authors continue by describing glutamate as the major excitatory neuro-transmitter and as critical in the early development of the brain … leading to an hypothesis that ASD may be caused by abnormalities in glutamate transmission.
However, they are able to refer to only one previous study (involving a very small sample of children with autism) which measured the concentration of glutamate or glutamine in those brain regions associated with the expression of ASD.
Magnetic resonance spectroscopy techniques are now available for quantifying brain metabolites, but the existing results have lost meaningfulness as a result of focusing upon children and of including learning disabled children with physical illnesses.
The study completed by Page et al
focused upon adults with ASD, along with a sample of typical adults as
controls, in order to investigate the amygdala and hippocampus regions.
The findings showed that the target individuals had a significantly higher concentration of glutamate/glutamine as well as creatine in these regions.
Glutamate is described as the most abundant central neurotransmitter and is critical for a range of neuro-developmental processes such as the creation of synapses; and the authors conclude by suggesting that neuro-developmental differences in ASD may be explained, at least to some extent, by differences in these glutamine functions … all of which is consistent with existing data indicating that metabolic anomalies in the limbic system are consistently associated (whether as cause or consequence) with autism and autistic spectrum disorders.
Still on the subject of the amygdala, Juranek et al (2006) explored, via magnetic resonance imaging, the postulated relationship between amygdala volume and anxiety level among children diagnosed with autism and ASD.
In their introduction, the authors describe how these conditions have traditionally been investigated via clinical assessments of particular aspects of behaviour, such as social deficits, delayed or disordered language, and restricted or repetitive behaviours.
However, technological developments over recent years have made it possible more directly to investigate neurobiological factors and the relationships between brain structure or function and observable behavioural traits.
The data emerging from such studies are taken by Juranek et al to provide strong support for the view of autism and ASD as having multiple causal origins. No single mechanism is likely to account for the observable features of autism; but the authors argue that a logical starting point for the quest for neurobiological bases is the establishment of subgroups of ASD.
Their focus is upon anxiety because anxiety symptoms have consistently been reported to be common in both adults and children with ASD, but they are aware of no existing study which has looked into the possible link between brain structure and anxiety in this population.
However, given the social-emotional deficits or anomalies in ASD, the limbic system has been much studied although some inconsistencies exist in the findings in that some outcomes have highlighted decreases in the volume of the amygdala and hippocampus, others have indicated increases among high functioning individuals with autism, and others still have indicated no differences.
Such inconsistencies have been attributed to the widely heterogeneous nature of individuals with ASD, both from the clinical and neuro-imaging viewpoints, and to the individuality of inclusion or exclusion criteria when it comes to organising target and control groups for investigations.
The approach pursued by Juranek et al sought a more powerful analysis by examining particular traits within a sample of validly-diagnosed individuals with autism and ASD in order to evaluate a specific brain-behaviour association … ie the amygdala volumes in relation to symptoms of depression and anxiety within an ASD cohort as assessed by the relevant subscale of the Child Behaviour Checklist.
The participants had a mean age of around 8 years, with an age range of 4 to 14 years. Their mean full IQ according to the Wechsler Scales was 60.5, although the range of IQ figures recorded was very wide. There was a total of 42 children - 35 boys and 7 girls - whose results were subject to the analysis involving linear regression methods (with Pearson correlation coefficients) to assess the association between amygdala volume and observable levels of depression/anxiety.
The results, based upon the scores on the relevant scale of the Child Behaviour Checklist, supported a significant brain-behaviour relationship between amygdala volume and depression/anxiety.
This fits with existing evidence for amygdala enlargement in autism relative to typical developing children, and for the frequency of anxiety as a comorbid feature of autism. It also fits with animal studies where amygdala lesions in primates bring about altered responses to arousing stimuli, suggesting that abnormal functioning of the amygdala contributes to the observed higher levels of anxiety.
The authors compared their
findings with those of Sparks et al (2002) who also worked with a broad range
of participants and who reported a larger volume in the right and left amygdala
among children with autism compared to children with less severe (pervasive)
While there is still scope for investigating whether there is a specific mapping of depression and of anxiety onto left and right amygdale areas, it was concluded that generally increased amygdala volume could be associated with the symptoms of depression and of anxiety among children with autism and ASD.
Parental Age and Risk for Autistic Spectrum Disorder in Children.
According to recent evidence from the study by Croen et al (2007), advanced maternal and paternal ages are independently linked to higher risk for ASD among children.
These authors monitored all singleton children born in one region of California between the start of 1995 and the end of 1999, and they were able to identify nearly 600 children who were diagnosed with ASD. A comparison was made between the circumstances of this subgroup and the remaining majority (132,000 + children).
It was calculated that the risk for ASD increased significantly for each 10-year increase in maternal age and paternal age. While sex differences did not reach a statistically significant level, there was an observable trend for associations with paternal age to be somewhat stronger for girls than for boys.
The authors argued that the finding of heightened ASD risk independently linked with maternal and paternal ages could be seen as valid in the light of the large population-based sample and of the prospective collection of data concerning ASD diagnoses and associated factors …. ie avoiding possible biases that can be linked with parental recall.
The authors concluded that, if the relationship between parental age and ASD is causal, the proportion in the current sample attributable to parental age above 35 years could be as much as 13% …. and they indicated the need for further studies to identify both genetic and environmental factors that correlate with advanced parental age.
Anomalous Functioning : Changes over Time
Pellicano (2007) begins her thoughts on autism as a developmental disorder by noting the contrast between the length of time since autism was first isolated and the still limited level of understanding of its nature and causation.
Her particular interest has been in the underlying cognitive abnormalities and how their manifestation changes over time.
Autism is seen as a complex neuro-developmental disorder whose symptoms of anomalous social interactions, delayed or disordered communication, and restrictive/stereotyped behaviours, are observable from very early childhood … although there is much variability in the particular profile of strengths and weaknesses and in the severity of the impairments, hence the tendency to talk in terms of autistic spectrum disorder.
The implication behind this observed variability is for multiple perspectives and means of understanding the disorder, including genetic and biological, cognitive, and behavioural areas. Further, anomalies and impairments within the three areas will interact with each other so that, for example, early cognitive deficits may have significant effects upon ongoing psychological and neurological development.
It is also recognised that multiple and interacting genes are involved in the aetiology of the disorder and that the neurological abnormalities are pervasive across a number of brain regions.
Cognitive research has been dominated by three “domains” …. theory of mind (the ability to recognise the thoughts and feelings and perspectives of other people); executive functioning (planning actions and using experiences); and central coherence (the capacity to use context and to make sense of information coming from different sources).
However, no amount of research in any one of these three domains has been able to provide an explanation of the nature of autism and ASD but can only provide some insight into part of the main triad of symptoms. For example, limited theory of mind might be a plausible hypothesis to explain problems with shared attention, or with pretend play, or with empathy, but it can hardly account for the repetitive and restricted interests which are more associated with poor executive functioning. Meanwhile, certain autistic strengths (such as a focus upon detail without being distracted by the whole stimulus array, or visuo-spatial capacities as in Block Design puzzles or in figure-ground discrimination) could be explicable in terms of a weak tendency towards central coherence.
Pellicano’s own work has sought to determine whether young children with autism are characterised by some overall and single cognitive abnormality or by a number of coexisting abnormalities.
Her participants were 40 children diagnosed with ASD or with pervasive developmental disorder , aged between 4 and 7 years, and within the higher functioning range (all with an IQ above 70).
All the children were assessed on a range of tasks which tap skills in the three cognitive areas listed … for example, false belief tasks to determine theory of mind capacity; tests of planning and of cognitive flexibility to tap executive functioning; and various tasks associated with local processing (ie weak central coherence).
The findings provided support for the multiple-deficit theory since the target children performed more poorly than typically-developing controls on the false belief tasks and tasks of higher order planning and flexibility. However, they performed more efficiently on tasks involving part by part processing instead of global processing.
The conclusion was that children with ASD possess a particular profile of cognitive weaknesses and strengths so that there is no single deficit but several underlying deficits along with certain strengths.
Further, the expectation that children who perform most poorly on cognitive tasks would prove to be the ones with the more severe symptoms was not confirmed. Instead, severity of autistic symptoms was unrelated to performance in any of these cognitive domains.
The questions raised include how the cognitive strengths and weakness do map onto the behavioural features of ASD; whether the pattern of strengths and weaknesses observed in young children persists over time; what underlies the profile of strengths and weaknesses; and whether one area of cognitive impairment is primary or whether each is independent of the others.
Pellicano then describes the “matching” approach common in research with children whereby children with ASD are compared across a number of skill areas with children of a similar mental age but with some alternative developmental difficulty such as a learning disability. The information obtained from such matching gives some insight into the level of performance observed, and whether it is in line with expectations for a given developmental level.
The problem is, however, that typical research methodology tends to examine particular capacities or behaviours at a single point in time, and can only offer a “snapshot” of the children and has little information about developmental processes, when observers have recognised that behavioural symptoms of ASD vary in their severity over the course of time. For example, poor level of speech at five years is not a strong predictor of the level of speech at ten years (although the quantity of speech may still fall short of the quality of speech and its effectiveness for true communication and for establishing/maintaining positive social interaction).
The question remains, therefore, about the way ASD manifests itself over time, especially with regard to cognitive capacities.
What limited evidence there is, according to Pellicano’s review, of the stability (or otherwise) of performance and abilities and behaviours, suggests that autistic symptomatology is indeed liable to change over time. Social and communicative abilities tend to improve, but there is reported to be little change in the area of restricted and repetitive behaviours. One possible implication (according to Charman et al 2005) is that is that there are different developmental trajectories for the different symptom areas.
Meanwhile, consistent evidence points to the stability/continuity of the clinical diagnosis so that most children diagnosed with ASD during the preschool period would still be so-classified during later childhood, adolescence, and adulthood.
Her review also indicates that there can be significant improvement in performance on a variety of theory of mind tasks, which seems to be linked to the children’s language abilities ….. (and also, one might speculate, to the opportunities to practise such tasks and to receive direct guidance and teaching about this kind of social skill) …..; but planning abilities and the ability to be flexible, as in a willingness to shift attention, undergo little change over time.
The existing questions have,
therefore, been refined to include the nature of the relationships between
cognitive abilities and cognitive deficits ; the nature of early predictors of
later cognitive functioning; the links between cognition and behaviour; and the
role (and predictive significance) of language skills.
Pellicano believes that answers will only emerge as a result of closely monitoring children from early infancy (and initial diagnosis) through adolescence and into adulthood along with the integration of information from research into genetics, neurobiology, cognition, and behaviour.
Predictors of Communication Development
To continue the theme of developmental change and its prediction, one notes the work of Toth et al (2006) who looked at joint attention and imitation and toy play as possible foci by which to gain early predictions about later communication capacities.
They begin by describing the great variability among outcomes in autism so that some individuals consistently maintain poor levels of functioning while others go on to achieve adequate levels of scholastic, social, and occupational functioning.
They argue, on the basis of a review of studies, that a major predictor of positive long term outcomes is the acquisition of spoken language … meaningful speech by around 5 or 6 years of age has been associated with both social and academic competence at a later stage.
The observable capacities associated with speech and language development in both children with ASD and control children include joint attention, imitation, and toy play.
The first indications of joint attention, such as following another person’s direction of gaze, can typically be observed by 9-12 months of age, and can be as early as 6 months. Through this kind of interaction, the infant is able to link words and phrases with objects and actions; and it is within the context of shared attention that the infants begin to demonstrate intention through their sounds and gestures, such as pointing to, or reaching out for, some desired object.
However, children with autism and ASD show impairments in these joint attention skills; and the impairments are more marked in “protodeclarative” sounds or gestures (sharing attention for purely social reasons) than in “protoimperative” sounds and gestures (requesting access to something).
It is further speculated that joint attention skills provide a foundation not only for the development of language, but also for other complex abilities such as pretend play and theory of mind.
Motor imitation skills have also been linked to the emergence of social and language skills; and it appears that imitation of actions by the infant serve to provide shared social experiences and a sense of connection/communication with other people.
One argument has it that a failure to engage in early imitative play could interfere with the development of joint attention, social reciprocity, and theory of mind abilities.
While imitative abilities can be observed virtually from birth in the case of typically-developing children, the children with autism and ASD show significant impairments in imitation of facial and body movements … and such weaknesses have been correlated with early language performance.
Play is a further skill area associated with language and social communication ability. Again in contrast with typically-developing children, children with autism and ASD show specific impairments in symbolic play, and this kind of impairment can be detected as early as 18 months; and even when symbolic play skills are developed, their level of sophistication tends to remain below that of language level.
The authors introduce their own study by noting that joint attention, imitation and play have all been linked to the development of language and communication, but their research set out to examine the simultaneous contributions of all three towards language skills in a large sample of preschool children with autism, with the opportunity afforded to examine the relationship among these early skills and the rate of development of communication skills during the preschool and early school years.
The participants included 42 children with ASD and 18 with pervasive developmental disorder, ranging in age from 34 to 52 months, and followed until 65 to 78 months of age. The ASD diagnosis had been based upon the Autistic Diagnostic Interview and the Autism Diagnostic Observation Schedule; and, over time, the children were assessed regularly on the Mullen Scales of Early Learning, the Early Social Communication Scales, and the Vineland Adaptive Behaviour Scales.
Observation sessions focused upon joint attention skills, imitation, and toy play.
The results indicated that initiating protodeclarative joint attention and immediate imitation abilities were most strongly associated with language skills in 3-4 year old children.
Examining growth curves showed that toy play and deferred imitation were associated with higher rates of acquisition of communication skills between 4 and 6-5 years … ie children with ASD who had better toy play and deferred imitation skills had communication acquisition rates that were comparable to those of typically developing children (although their lower basal level meant that language skills were still below the age norm at outcome).
The general pattern of findings suggested that all three cognitive areas – joint attention, imitation, and toy play – are important for laying a foundation for language development among children with autism and ASD, but it is the combination of deferred imitation and toy play skills that is associated with the continued expansion of communication skills over the longer term (through pre-school age and the early years of schooling).
The authors speculate that joint attention and immediate imitation are important as starter skills that form the basis for social and communicative exchanges in which language can develop. Once this stage is set, representational skill become the more important with toy play and deferred imitation abilities providing the context for the development of higher level cognitive skills socially and linguistically.
Such skills require an active interest in other people and/or objects … the capturing of the child’s attention; representational thinking … forming and storing mental images; and intact recall memory (calling up that mental image at some later time); and planning skills. In other words, through toy play and imitation, the child not only comes to an understanding of his world but can also show the understanding.
The practical implication is for the opportunity to practise joint attention, imitation, and toy play in early intervention programmes as a significant precursor to the development of other skill areas, notably expressive language.
A further piece of practical advice concerns the wide variability of language skills among children with autism and ASD so that true capacities and their change over time are not easily measured by any single language test, and the authors recommend the use of the Vineland Adaptive Behaviour Scales as a means of assessing language acquisition over time.
Sex Differences in Preschool Children with Autism
The starting point for the study completed by Carter et al (2007) was their recognising the lack of studies concerning sex differences in developmental functioning or clinical manifestation in respect of ASD … although there is much evidence that the prevalence of such disorders is greater among boys than among girls (with estimates ranging from 2.5:1 to 4:1).
In any case, what research findings do exist on sex differences are not easy to summarise or interpret given that studies differ widely in their methodologies, and cover a long time scale during which ASD diagnostic criteria have changed. Thus, for example, early studies are likely to have included the more severely affected children who would not have had the opportunity to benefit from the advances in intervention approaches.
There is, in addition, the problem whereby studies vary in their use of group formation with some seeking to match groups according to mental age or chronological age, and it is now recognised that IQ matching among participating groups is important so that any observed sex differences can be attributed to the degree of autism rather than to differences in cognitive levels.
One existing and converging piece of evidence is that the ratio of males to females within the ASD population is moderated by IQ … ie the majority of females diagnosed with autism and ASD tend to be in the lower end of the IQ distribution, and the greatest boy : girl ratio differences are observed among samples of high functioning individuals with autism and ASD.
( The present writer – MJC – would simply ponder whether there might still be something artifactual about these ratios. It is recognised that boys may stand out more readily than girls if autistic behaviours are more discrepant from what is perceived as typical of boys than what is typical of girls, so that one might more readily monitor and assess boys who tend not to join active play, for example. Girls may be identified less readily as showing some special needs or atypical behaviours unless the observed symptoms and limited performance are marked. In other words, while the greater awareness of ASD needs and symptomatology now would probably ensure appropriate diagnoses, it may just be possible that there has been hitherto some under-identification of those girls with ASD who are towards the higher functioning end of the autistic spectrum. [ Alternatively, it could be that the current trend to be a little generous in giving informal labels of ASD to children could still involve a ratio in favour of boys given that boys’ maladaptive behaviour is such – disturbing as much as disturbed – that there may tend to be more boys than girls available for this kind of informal classification ] ??)
In any event, Carter et al go on to report that findings in other areas have been less consistent, although there is some evidence that males more frequently than females have unusual visual interests , and less appropriate (more stereotypic) styles of play even after controlling for the effects of IQ.
Further, parents recalling their children’s development before the age of 5 years tended to rate boys as more impaired in social communication and interaction. However, when describing their children’s social behaviour at a later stage, parents often found their daughters to be more impaired in respect of making and maintaining friendships.
Among community samples of individuals with autism, Posserud et al (2006) found that parental reports of 7-9 year old children usually rated boys as showing more features of autism than girls; although self reports about empathy and general social functioning suggest better performance among females.
The aim of the study by Carter et al was to investigate sex differences in toddlers with ASD with regard to profiles of verbal, nonverbal, and motor functioning, and general symptomatology of autism including social performance, communication, and repetitive/restricted behaviours. It was claimed that this study would be innovative in including very young children; and it was hypothesised that girls compared to boys would demonstrate lower developmental functioning.
The target sample of children comprised 22 girls and 68 boys, all diagnosed with ASD, and aged between 18 and 33 months.
Diagnosis was confirmed with the ADI and the ADOS; and level of functioning was measured by the Vineland Adaptive Behaviour Scales, and the Infant-Toddler Social and Emotional Assessment where parental reports are gained concerning the child’s behaviour, competences, and problems.
The Mullen Scales were used to assess cognitive and developmental functioning in the areas of expressive and receptive language, fine and gross motor skills, and visual reception (with the opportunity to calculate an early learning composite score).
Home visits were arranged to conduct interviews and to assess the children, and parents were sent the rating scales to complete.
The results indicated that, consistent with existing evidence, boys and girls showed relative strengths in visual reception and fine motor skills, while gross motor skills and language skills (expressive and receptive) were poorer. Language was the weakest area of functioning.
The hypothesis that girls would show poorer performance than boys in all the aspects of developmental functioning was not supported.
Rather, girls and boys show different cognitive and developmental profiles. The boys showed stronger verbal and motor skills, and were seen as showing more advanced social skills, while the girls showed stronger skills in visual reception or in nonverbal problem solving.
Such findings are in contrast with previous literature concerning older children and adults with ASD which found consistently higher performance among males in all domains of cognitive functioning. Further, current results indicated that nonverbal skills (as reflected in scores of visual reception) are very significant for early development and are a significant predictor of performance on composite measures of language, motor, or social performance.
The more impaired gross motor skills observed in the girls was seen as consistent with existing findings about the better physical skills, as observed/assessed, among boys.
However, there were no indications in this sample of sex differences in repetitive or stereotypic movements … possibly attributable to the very young age of the children concerned.
In contrast to findings among typically-developing children where parents rate girls as having generally higher social-emotional competence than boys, the current sample of parents of children with ASD rated girls as having lower competence, especially with regard to empathy. This was not reflected in the ADOS ratings, and the authors speculated that the results were associated with higher expectations held by mothers concerning girls’ social-emotional competence relative to that of boys such that comparable behaviours among boys and girls are rated more negatively among the girls.
More advanced motor skills were reported for boys across both rating scales and direct assessment; but for the social and language domains, significant sex differences were observed only in parental reports and not in direct assessments.
The authors acknowledge some weaknesses in this study, notably the lack of opportunity to determine whether the current findings, which show some discrepancies from previous findings, may be a function of the young age of the sample of children, or are attributable to something unusual about this particular sample.
There was also acknowledged to be the possible confound of exposure to various forms of early intervention on the part of many of the children.
The authors summarise their findings in terms of their expectation, based upon the preponderance of boys relative to girls among individuals with autism, and the suggestion that girls tend to be more impaired than boys, that girls require a higher threshold of genetic vulnerability to result in observable and significant symptomatology … ie that girls would be affected less frequently, but more severely.
This model is not supported by the current findings (nor by large epidemiological studies such as that of Szatmari et al 2000).
The possible implication is that sex differences observed may reflect different underlying genetic mechanisms for boys and for girls, (or subgroups thereof), or that hormonal influences or differences in the development of male and female brains moderate shared genetic influences.
* * * * * *
M.J.Connor April 2007
Carter A., Black D., Tewani S., Connolly C., Kadlec M., and Tager-Flusberg H. 2007 Sex differences in toddlers with autistic spectrum disorders. Journal of Autism and Developmental Disorders 37 86-97
Charman T., Taylor E., Drew A., Cockerill H., Brown J., and Baird G. 2005 Outcome at 7 years of children diagnosed with autism at age 2. Journal of Child Psychology and Psychiatry 46 500-513
Croen L. et al 2007 Archives of Pediatric and Adolescent Medicine 161 334-340. Reported by Barclay L. and Nghiem H. “ Advanced parental age predicts risk for autistic spectrum disorder in children. ” Medscape Medical News : April 5th 2007
Juranek J., Filipek P., Berenji G., Modahl C., Osann K., and Spence A. 2006 Association between amygdala volume and anxiety level. Journal of Child Neurology 21(12) 1051-1058
Page L., Daly E., Schmitz N. et al 2006 In vivo H-magnetic resonance spectroscopy study of amygdala-hippocampal and parietal regions in autism. American Journal of Psychiatry (December) 163(12) 2189-2192
Pellicano E. 2007 Autism as a developmental disorder : tracking changes over time. The Psychologist 20(4) 216-219
Posserud M., Lundervold A. and Gillberg C. 2006 Autistic features in a total population of 7-9 year old children assessed by the ASSQ. Journal of Child Psychology and Psychiatry 47(2) 167-175
Szatmari P., McLean J., Jones M., et al 2000 The familial aggregation of the lesser variant in biological and non-biological relatives of PDD probands. Journal of Child Psychology and Psychiatry 41(5) 579-586
Satzmari P. et al (The Autism Genome Project Consortium) 2007 Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nature Genetics 39 319-328
Sparks B., Friedman S., Shaw D., et al 2002 Brain structural abnormalities in young children with autism spectrum disorder. Neurology 59 184-192
Toth K., Munson J., Meltzoff A., and Dawson G. 2006 Early predictors of communication development in young children with autism : joint attention, imitation, and toy play. Journal of Autism and Developmental Disorders. 36 993-1005
© Mike Connor 2007.
Back to NAS Surrey Branch Welcome Page