Jane Shields
Rosemary Varley
Paul Broks
Adrian Simpson

Developmental Medicine and Child Neurology, 1996, 38, 487-495
©Mac Keith Press


Autism is a developmental condition involving a triad of social impairments: in social relationships, communication, and imagination (Wing and Gould, 1979). Rapin and Allen (1987) commented that one subtype of developmental language disorder - 'semantic-pragmatic syndrome' - was frequently seen, in a severe form, in verbal children with autism. In children with 'semantic-pragmatic' disorder, the onset of language use is delayed, and when language emerges it is accompanied by echolalia, jargon and auditory inattention. When they are older, such children use superficially complex language with clear articulation but have difficulty with the use and understanding of language, interpreting over-literally and using language inappropriately in conversation. There has been debate as to whether semantic-pragmatic language disorder forms part of the disorders of the autistic spectrum.

Language has canonical components such as phonology, syntax and lexical semantics, but other non-componential, non-literal and context-bound aspects of language are involved in its use in social discourse. The exercise of these pragmatic language skills relies upon the integrity of broader cognitive functions concerned with the processing of social information. Language use also involves the broader skills of attention and drawing inferences.

The study of social cognition has been concerned both with the nature of social knowledge and with the nature of cognitive processes involved in the perception and interpretation of the social world (Ostrom 1984). It has been suggested that the social impairments found in autism are linked to an inability to mentalise, a lack of understanding of mental states: of a 'theory of mind' (Baron-Cohen 1989, Frith 1989).

Brothers (1990,1992) proposed a specific brain system for social cognition, defining the end result of social cognition as the accurate perception of the dispositions and intentions of other individuals. From accumulated work with primates, Brothers identified three brain areas which are important for social cognition: the amygdala, the superior temporal sulcus, and the orbital-frontal cortex.

Hetzler and Griffin (1981) favoured bilateral temporal dysfunction as the underlying cause of autism, and Damasio and Maurer (1978) proposed bilateral dysfunction in structures including the ring of mesolimbic cortex. Kluver-Bucy syndrome (resulting from bilateral induced lesions of the amygdala and inferior temporal cortex) has been cited as a neurological model of autism (Damasio and Maurer 1978). Bachevalier (1991) found impairments of social interaction in monkeys with bilateral lesions of the amygdala and hippocampus. Bishop (1993), in a review paper, suggests that the limbic system and associated frontal lobe structures may well contribute to the neurobiological basis of autism.


Subject characteristics
Group        Age          Handedness     Sex       Socio-Economic Status1
          Mean  SD        Right Left  Male Female      Range   Mode
S        105.9 19.78      9     1     10   0           l-4     2
P        104.4 19.44      9     1     10   0           l-4     2
A        108.9 18.54      8     2      9   1           l-4     2
C        106.3 19.84      8     2      5   5           l-4     2
1Socio-economic status (as defined by the Office of Population Censuses and Surveys): 1 = professional etc occupations. 2 = intermediate occupations. 3(N) = skilled occupations. non-manual. 3(M) = skilled occupations, manual. 4 = partly skilled occupations. 5 = unskilled occupations. S = Semantic-pragmatic, P = Phonologic-syntactic. A = high-level autism, C = controls (normal).


In a related study (Shields et al. 1996) we have found evidence of neuropsychological similarities between semantic-pragmatic language disorder and high-level autism. Groups of children with these two disorders had similar patterns of results in neurological tests indicative of right hemisphere functional deficiency.

The aim of this study was to further clarify this link by examining some aspects of social cognition in groups of children with 'phonologic-syntactic disorder' (P); and 'semantic-pragmatic disorder' (S); in a group of children with high-level autism (A): and in a control group of normal children (C).

Our hypothesis was that the groups with semantic-pragmatic language disorder and with high-level autism would perform poorly on the tests of social cognition.

Four groups of 10 children in the age range 7 to 10 years participated in the study. The groups are described in more detail in the companion paper (Shields et al. 1996). Of the two groups with language impairment, one (P) had a disorder of language form (phonology and syntax) and one (S) had semantic and pragmatic difficulties. These two groups were recruited and constituted as described in the companion paper (Shields et al. 1996), and their communication symptoms were rated as described there, using Rapin and Allen's (1987) criteria for 'phonologic-syntactic syndrome' and 'semantic-pragmatic syndrome'. They all had special educational needs and were being educated in facilities for children of normal intelligence. None was known to have signs of brain pathology. None had been diagnosed as having autism.

Another group (A) comprised children with high-level autism according to DSM-III-R criteria.

A group of control subjects (C) comprised normal children, from the same area.

As described in the companion paper, age was controlled by matching across the four groups, each set of four subjects being within 9 months of age of each other. The groups were also matched for socio-economic status. None of our subjects with language disorder were girls, reflecting the greater prevalence of developmental language disorder in boys (Eme 1979). Since those tests which had norms had been standardised on groups of boys and girls, we selected a control group of boys and girls. Subjects' details are given in Table I.

Each subject was tested individually by the first author in their own school or home. Tests of social cognition included social comprehension questions, theory-of-mind procedures and a test to detect eye direction. Details of the contents of the test battery are given in the Appendix.

We predicted that groups S and A would perform less well than groups P and C. This prediction was tested by profile analysis (Morrison 1990, Johnson and Wichern 1992) to discover whether there were general differences between the groups, followed by more specific between-group comparisons. All data were rendered commensurable by conversion to percentages if necessary.

Tests on assumptions
With sample sizes as small as 10, tests on assumptions were unlikely to give very clear conclusions. Furthermore, in 14 of the 16 samples, an appreciable number (between three and eight) of the observations were tied. This was particularly true of the Eye Direction Detection test, where at least half the observations were tied in each of the four samples. If one attempts to test for univariate normality by computing Pearson correlation co-efficients, r, from quantile-quantile plots of the ordered scores versus standard normal quantiles, as suggested by Johnson and Wichern (1992), misleadingly high values of r are obtained as a result of the ties. Therefore, in view of the small sample sizes and numerous ties, tests on univariate normality could not yield useful information. Furthermore, the ties would remain even if the data were transformed. e.g. to logits, in an attempt to seek data likely to conform better to normal distributions.

The equality of the covariance matrices was evaluated using the test devised by Box (1950). The limitations of using this procedure in our study are explained elsewhere (Shields et al. 1996). Box's statistic distributed as F was computed in addition to the more usual X2 statistic. F was significant at a level of p=0.0001 and X2 at p<0.025, suggesting that equality of covariance matrices could probably not be assumed.

Since parametric multivariate analyses were probably inappropriate in this context (Shields et al. 1996), we decided to use randomisation tests (Edginton 1980, Manly 1991) as the basis of the analyses. Such an approach would still enable multivariate analyses, such as profile analysis, to be computed, but would overcome the problem of violation of parametric assumptions by establishing significance by randomisation.

Profile analysis
Test statistics were computed for a form of profile analysis (Morrison 1990) based on the two-way mixed-model analysis of variance (ANOVA). In our randomisation version of this test, the significance of the computed F ratios was determined by relating them to their randomisation distributions using Manly's (1991) 'two-stage' randomisation method for a design involving both an independent-samples factor and a repeated-measures factor. In this and in all the other randomisation tests described below, the randomisation distribution was established from 9999 random permutations of the data plus the actual data set.

Running the randomisation test on the ANOVA version of profile analysis, we found significance levels of p<0.000l for the Groups effect, p<0.0001 for the repeated-measures (Tests) effect and p=0.0172 for the interaction (Parallelism) effect. We could not conclude that the profiles were parallel (as shown by the third significance level) and there was therefore some doubt about the scope of the first result that the groups differed significantly. The significance of the repeated-measures effect will be ignored, because differences between the mean scores across the tests are not of interest. It was therefore necessary to examine the individual tests in the battery separately.

The graph of the profiles (Fig. I) showed clearly that the S and A groups were approximately coincident, and that their mean results were below those of the P and C groups, as predicted. The size of this difference varied across the tests in the battery, being particularly strong for the Theory of Mind and Social Comprehension B tests.

Fig. I. Mean scores of children in four experimental groups on tests of social cognition. P = phonologic-syntactic disorder, S = semantic-pragmatic disorder, A = high-level autism, C = control group of normal children. Error bars show ħISEM.


Right hemisphere battery: one-way ANOVAs
Test                      F value of        Significance level by
                         actual data         randomisation test
Theory of Mind                26.13                <0.0001
Eye Direction Detection       5.21                  0.0039
Social Comprehension W1       16.66                <0.0001  
Social Comprehension B2       14.06                <0.0001
1From the Wechsler Preschool and Primary Scale of Intelligence Verbal Comprehension subtest (Wechsler 1967)
2From the British Ability Scales Social Comprehension subtest (Elliott et al. 1977)


To examine differences between the groups within each of the four tests separately, F statistics from individual one-factor independent-samples analyses of variance were computed in randomisation tests. They gave the results presented in Table II.

Modified Fisher PLSD tests on individual comparisons gave the results presented in Table III. Again a randomisation process based on F statistics was employed for each comparison between means. Asterisks to indicate statistical significance were attached only if the p value was less than a critical value derived from application of the Bonferroni inequality, excluding comparisons for which the means were not expected to differ. This gave p = 0.05/4 = 0.0125 as the critical p value. The comparisons between P and C, and between S and A, are included in the table merely to show that, if one tests these differences against the same criterion, they are not significant. However, there would be some evidence, if one adopted the 0.05 'per comparison' error rate, that groups P and C differed significantly on two tests.


Two-tailed p values
Test                      P vs S  P vs A  P vs C  S vs A  S vs C  A vs C
Theory of Mind            0.0005* 0.0003* 0.0241  0.7588  0.0001* 0.0001*
Eye Direction Detection   0.02082 0.0118* 0.6298  0.6303  0.0255  0.0158
Social Comprehension W    0.0012* 0.0020* 0.0394  0.2612  0.0001* 0.0002*
Social Comprehension B    0.0037* 0.0045* 0.3095  1.000   0.0002* 0.0008
p<0.0125; see text for discussion.


One-tailed p values
Test                      P vs S  P vs A  P vs C  S vs A  S vs C  A vs C
Theory of Mind            0.0005* 0.0003                  0.0001* 0.0001*
Eye Direction Detection   0.0146  0.0056*                 0.0118* 0.0088*
Social Comprehension W    0.0009* 0.0007*                 0.0001* 0.0001*
Social Comprehension B    0.0017* 0.0025*                 0.0002* 0.0002*
*p<0.0125; see text for discussion.


Comparison of P+C vs S+A

Test                     One-tailed              Two-tailed
Theory of Mind             0.0001                  0.0001
Eye Direction Detection    0.0001                  0.0002
Social Comprehension W     0.0001                  0.0001
Social Comprehension B     0.0001                  0.0001


Where a difference had been predicted, its direction had also been predicted. We could therefore employ one-sided significance tests for the four sets of predicted differences. Table IV assesses the one-sided significance levels, against the Bonferroni p value of 0.0125. Since no predictions about differences were made for the P-to-C and S-to-A comparisons, one-sided tests were not computed in these cases. All but one of the one-tailed values were deemed 'significant' on this criterion. In a few cases the one-tailed values did not differ from two-tailed for technical reasons associated with the randomisation tests.

The most direct test of the prediction that groups S and A should do less well than groups P and C (which overcomes the disadvantage that comparisons between tests are not independent) was to compute specific contrasts between the S and A groups combined, on the one hand, and the P and C groups combined, on the other. The results of randomisation tests, using F statistics, of these contrasts are presented in Table V. (Although one- and two-sided p values are both given, the p values are so low the distinction hardly matters.)

The results of these specific comparisons were very clear. The combined P and C sample consistently, and very significantly, did better than the combined S and A sample.

Clear support was found for the hypothesis that the groups with semantic-pragmatic language disorder and with high-level autism would both do poorly on tests of social cognition.

Phonologic-syntactic language disorder may be associated with immature theory-of-mind skills; or the lower score of group P vs group C may result from the demands made by this task on language-comprehension skills. However, it is clear that semantic-pragmatic language disorder and high-level autism involve a severe deficit in this area.

The significant differences between groups S+A and groups P+C for social cognition tasks support the opinion that 'semantic-pragmatic disorder' is a disorder of the autistic spectrum and that the weaknesses in communicative competence in children with the disorder may result from, or be associated with, an underlying cognitive deficit which is not primarily linguistic in nature.

The results of this and the parallel study (Shields et al. 1996) link semantic-pragmatic language disorder and high-level autism in two ways: first, the two groups share a pattern of results indicative of right hemisphere functional deficiency (e.g. poor performance of visual and visuo-constructional tasks); second, they share a pattern of results indicative of social-cognitive dysfunction. It is worth considering briefly how these findings might be accommodated in neuropsychological terms.

There have previously been suggestions that autism can be characterised as a disorder of the right hemisphere (Sarvis 1960, Fein et al. 1984, Goodman 1989). With regard to the 'right hemisphere battery' of tests, the results of the first study (Shields et al. 1996) appear to fit with this notion. However, there has also been speculation over many years that abnormalities of frontal lobe and/or limbic structures may lie at the root of autistic disorder, a view that has gained favour in recent years. Bishop (1993) argues persuasively that the notion of frontal lobe and limbic system dysfunction provides the most promising working hypothesis not only in terms of the search for neurobiological correlates of this disorder but also in providing insight into the range of psychological impairments characteristic of the condition. As an example of the convergence of psychological and neurobiological thinking, Baron-Cohen and Ring (1994) hypothesise that the core cognitive deficits of autism may be traced to neural circuitry encompassing the superior temporal sulcus, the orbitofrontal regions and the amygdala. Such speculation is in line with the view of Brothers (1990), who has proposed that there is a specific brain system responsible for mediating social cognition, which incorporates the amygdala (concerned with the regulation of emotional response), the superior temporal sulcus (concerned with various aspects of face perception), and the orbitofrontal cortex (concerned with the regulation of behaviour within social contexts).

It is interesting to speculate on whether the pattern of results observed for the semantic-pragmatic and autistic groups in the two studies might be accommodated in terms of abnormalities within this proposed neural circuitry. We suggest that the present results could be explained, at least in part, by abnormalities restricted to the amygdala, arguably the key component of the hypothesised brain system in view of its rich interconnectivity with frontal and temporal regions as well as with subcortical structures concerned with autonomic and endocrine functions underlying emotional response (Amaral et al. 1992). In recent years a picture of the cognitive consequences of damage to the amygdala has begun to emerge and the reported effects of restricted bilateral amygdala lesions in humans bear some similarity to the profile of results reported for the semantic-pragmatic and autistic groups in the present study. For example, specific impairments of visual memory and face processing have been observed in the context of preserved canonical language skills (Jacobson 1986, Tranel and Hyman 1990, Adolphs et al. 1994, Allman and Brothers 1994, Young et al. 1995). Allman and Brothers noted the possible relevance of such studies to the understanding of autism, and there had already been some suggestion that the Kluver-Bucy syndrome (which depends critically upon bilateral amygdala damage) carries parallels with autism (Damasio and Maurer 1978). In the light of the present observations, the comparison of autistic cognitive functions and deficits associated with acquired amygdala damage promises to be a fruitful line of inquiry.

Our studies illustrate both strengths and limitations of the neuropsychological approach. Comparison of test profiles across different types of developmental and acquired cognitive disorder serve to highlight points of convergence and suggest common underlying neurological abnormalities, but the tests themselves lack the power to indicate definitively which brain systems may be involved (the amygdala system, or right neocortical areas). However, by bringing candidate brain systems into view, neuropsychological studies give direction to neuropathological studies. The way forward in understanding the neurological basis of developmental disorders is likely to involve a combination of neuropsychological and neuropathological methods, including structural and functional brain imaging.

In conclusion, our data are consistent with a notion of right hemisphere dysfunction as a contributor to both semantic-pragmatic language disorder and autism, but it is possible that this reflects abnormalities within a more complex functional-anatomical system, perhaps involving the amygdala.

SC1 - Theory of Mind (Wimmer and Perner 1983, Baron-Cohen et al. 1985, Perner and Wimmer 1985, Baron-Cohen 1989, Perner et al. 1989, Happe 1991):
tests of first- and second-order false belief and deception, which assess the ability to mentalise - to appreciate the content of other people's thoughts and beliefs. The skill of mentalizing develops with age in normal children, but is weak in children with autism.
SC2 - Eye Direction Detection (Baron-Cohen 1992): a technique for assessing the perception of gaze in the comprehension of mental states. Children with autism find this hard and tend to give egocentric responses.
SC3 and SC4 - Social comprehension (W and B): questions from the Wechsler Preschool and Primary Scale of Intelligence Verbal Comprehension sub-test (Wechsler 1967), and from the British Ability Scales Social Comprehension (Elliott et al. 1977) sub-test, which tap social cognition. Norms were not used, but children in the age range involved should find these questions easy, whereas children with autism have difficulty with such tests (Frith 1989).

Accepted for publication 1st August 1995.

Authors' Appointments
*Jane Shields, MPhil, PhD, DipCST, AIL, Speech and Language Therapist, Storm House School, National Autistic Society. UK.
Rosemary Varley. BSc. MA. PhD. Lecturer. Speech Sciences, University of Sheffield.
Adrian Simpson BA. PhD. Lecturer, Department of Psychology, University of Sheffield.
Paul Broks, BA. MSc. DPhil. CPsychol. Senior Lecturer, School of Psychology, University of Birmingham.

The authors acknowledge the help of Simon Baron-Cohen and Francesca Happe in devising the test battery used. They also thank the subjects who participated, and their parents, therapists and teachers.

*Correspondence to first author at Priory Annexe: Storm House School, St Wilfred's Road, Cantley, Doncaster. DN4 6AH. UK.

Two groups of children with contrasting types of developmental language disorder (phonologic-syntactic and semantic-pragmatic) were compared with a group of children with high-level autism and with a control group of normal children, on tests of social cognition (theory of mind: social comprehension: and detection of eye direction). The similarly poor performances of the semantic-pragmatic group and the autistic group suggest that semantic-pragmatic language disorder lies on the autistic spectrum.

Connaissance dans les troubles développementaux du langage et l'autisme de haut niveau
Deux groupes d'enfants avec des types contrastants de trouble développemental du langage (phonologique-syntaxique et sémantique-pragmatique) ont été comparés à un groupe d'enfants avec autisme de haut niveau et avec un groupe contrôle d'enfants normaux sur des tests de connaissance sociale (théorie de l'esprit: compréhension sociale: et détection de la direction oculaire). Des performances également pauvres dans le groupe sémantique-pragmatique et le groupe autistique suggerent que le trouble du langage de type sémantique-pragmatique appartient au spectre de l'autisme.

Soziale Fähigkeiten bei entwicklungsbedingten Sprachsörungen und hochgradigem Autismus
Zwei Gruppen von Kindern mit verschiedlichen entwicklungsbedingten Sprachstörungen (phonologisch-syntaktisch und semantisch-pragmatisch) wurden mit einer Gruppe von Kindern mit hochgradigem Autismus und mit einer Kontrollgruppe gesunder Kinder anhand von Test zur Erfassung sozialer Fähigkeiten (Theory of Mind, soziales Verständn und Fertlegung der Blickrichtung) verglichen Die ähnlich schlecten Ergebnisse der semantisch-pragmatischen Gruppe und der autistischen Gruppe lassen vermuten, daß die semantisch-pragmatisch Sprachstörungen im autistischen Spektrum liegt.

Cognición social en las alteraciones del desarrollo del lenguaje y autismo de alto nivel
Dos grupos de niños con tipos contrastados de alteración del desarrollo del lenguaje (fonolódgico-sintáctico y semántico-pragmático) fueron comparados con un grupo de niños con autismo de alto nivel y con un grupo control de niños normales utilizando pruebas de cognición social (teorfs de la mente, comprensión social y detección de la dirección ocular). La similtud de los pobres resultados en el grupo cottnicit'tn social (ieor{s de Ia inenie. cttmprensi6n social y detccci6n de Ia dirección ocular). La slinilitud de los pobres resultados en ci grupo semántico-pragmático y en el autistico sugieren que la alteración semantico- pragmática del lenguaje está situada en el expectro autistico.

Adolphs R, Tranel D Domasio H, Domasio A, (1994) Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala. Nature 372: 669-72.
Allman J, Brothers L, (1994) Faces, fear and the amygdala Nature 372: 613-4.
Amarai DG, Price JL, Pitkanen A, Carmichael ST. (1992) Anatomical organisation of the primate
amygdaloid complex. In: The Amygdala: Neurobiological Aspects of Emotion, Memory and Mental Dysfunction. New York: Wiley-Liss p 1-66.
Bachevalier J. (1991) An animal model for childhood autism: memory loss and socioemotional disturbances following neonatal damage to the limbic system in monkeys. In: Tamminga CA, Schulz SC, editors. Advances in Neuropsychiatry and Psychopharmacology 1: 129-40.
Baron-Cohen S (1989) The autistic child's theory of mind: the case of specific developmental delay. Journal of Child Psychology and Psychiatry 30: 285-98.
- (1992) Are children with autism blind to the mentalistic significance of the eyes? Paper given at a meeting of the British Neuropsychological Society Queen Square London 2nd November, 1992.
Leslie AM, Frith U. (1985) Does the autistic child have a 'theory of mind'? Cognicion 21: 37-46.
Ring H. (1994) A model of the mind reading system: neuropsychological and neurobiological perspectives. In: Mitchell P, Lewis C. editors. Origins of an Understanding of Mind. Hillsdale, New Jersey: Lawrence Erlbaum Associates.
Bishop DVM. (1993) Autism, executive functions and theory of mind: a neurological perspective. Journal of Child Psychology and Psychiatry 34: 279-93. (Annotation)
Box GEP. (1950) Problems in the analysis of growth and wear curves. Biometrics 6: 362-89.
Brothers L. (1990) The social brain: a project for integrating primate behaviour and neuropsychology in a new domain. Concepts in Neuroscience 1: 27-51.
- (1992) Perception of social acts in primates: cognition and neurobiology. Seminars in the Neurosciences 4: 409-14.
Damasio AR, Maurer RG (1978) A neurological model for childhood autism. Archives of Neurology 35: 777-86.
Edginton ES (1980) Randomization Tests. New York: Dekker
Elliott CD, Murray DJ, Pearson LS. (1977) British Ability Scales. Windsor: NFER-Nelson.
Eme RF. (1979) Sex differences in childhood psychopathology: a review. Psychological Review 86: 574-95
Fein D, Humes M, Kaplan E, Lucci D, Waterhouse L (1984) The question of left hemisphere dysfunction in infantile autism. Psychological Bulletin 95: 258-81
Frith. U. (1989) Autism - explaining the enigma. Oxford: Basil Blackwell.
Goodman R. (1989) Infantile autism: a syndrome of multiple primary deficits? Journal of Autism and Developmental Disorders 19: 409-24.
Happe FGE. (1991) Theory of Mind and Communication in Autism (Thesis). University of London.
Hetzler BE, Griffin IL (1981) Infantile autism: and the temporal lobe of the brain. Journal of Autism and Developmental Disorders 11: 317-30.
Jacobson R. (1986) Disorders of facial recognition, social behaviour and affect after combined bilateral amygdalotomy and subcordate tractotomy: a clinical and experimental study. Psychological Medicine 16: 439-50.
Johnson RA, Wichern W. (1992) Applied Multivariate Statistical Analysis. 3rd edition. Englewood Cliffs. NJ: Prentice Hall.
Manly BFJ. (1991) Randomization and Monte Carlo Methods In Biology. London: Chapman and Hall.
Morrison DF. (1990) Multivariate Statistical Methods. 3rd edition. New York: McGraw-Hill.
Ostrom TM. (1984) The sovereignty of social cognition In: Wyer RF, Skrull TK, editors. Handbook of Social Cognition. Vol 1. Hillsdale, New Jersey: Lawrence Erlbaum Associates.
Perner J, Frith U, Leslie AM, Leekam SR, (1989) Exploratian of the autistic child's theory of mind: knowledge, belief and communication. Child Development 60: 689-700.
Perner J, Wimmer H. (1985) John thinks that Mary thinks that...: attribution of second order beliefs by five to ten year old children. Journal of Experimental Child Psychology 39: 437-71.
Rapin I, Allen DA (1987) Developmental Dysphasia and Autism in Preschool Children: Characteristics and Subtypes. Proceedings of the First International Symposium on Specific Speech and Language Disorders in Children. University of Reading UK.
Sarvis MA (1960) Psychiatric implications of temporal lobe damage. Psychoanalytic Study of the Child 15: 454-81.
Shields J, Varley R, Broks P, Simpson A. (1996) Hemispheric function in developmental language disorders and autism. Developmental Medicine and Child Neurology 38: 473-86.
Tranel D, Hyman BT. (1990) Neuropsychological correlates of bilateral amygdala damage. Archives of Neurology 47: 349-55.
Wechsler D. (1967) Wechsler Pre-school and Primary Scale of Intelligence. New York:
Psychological Corporation.
Wimmer H, Perner J.(1983) Beliefs about belief: representation and the constraining function of wrong belief in young children's understanding of deception. Cognition 13:103-28.
Wing L, Gould J. (1979) Severe impairments of social interaction and associated abnormalities in children: epidemiology and classification. Journal of Autism and Developmental Disorders 9: 11-30.
Young AW, Aggleton JP, Hellawell DJ, Johnson M, Broks P, Hanley IR. (1995) Face processing impairments after amygdalotomy. Brain 118: 15-24.


Authors' Note:
The authors regret that they are not in a position to offer opinions, assessments or consultations on individual children.
Dr Jane Shields, The National Autistic Society, Earlybird Project, Hoylands House, Barnsley Road, Silkstone. Near Barnsley. S75 4NG

Copyright Holders Note:
Permission has been granted to put a copy of this article on this World Wide Web page by the publishers of Developmental Medicine and Child Neurology. It is not for reproduction without permission.

Back to NAS Surrey Branch Welcome Page