Superior temporal sulcus anatomical abnormalities in childhood autism: a voxel-based morphometry MRI study

www.elsevier.com/locate/ynimg NeuroImage 23 (2004) 364 – 369

Superior temporal sulcus anatomical abnormalities in childhood autism: a voxel-based morphometry MRI study N. Boddaert, a,b,* N. Chabane, c H. Gervais, a C.D. Good, d M. Bourgeois, e M.-H. Plumet, c C. Barthe´le´my, f M.-C. Mouren, c E. Artiges, a Y. Samson, g F. Brunelle, a,b R.S.J. Frackowiak, d and M. Zilbovicius a a

ERM 0205 INSERM-CEA, Service Hospitalier Fre´de´ric Joliot, CEA, 91406, Orsay, France Service de Radiologie Pe´diatrique, Hoˆpital Necker Enfants Malades, AP-HP, Paris V 75015, Paris, France c Service de Pe´dopsychiatrie, Hoˆpital Robert Debre´, AP-HP Paris, France d Wellcome Department of Imaging Neuroscience, Institute of Neurology, UCL, Queen Square, London, UK e Service de Neurochirurgie, Hoˆpital Necker Enfants Malades, AP-HP, Paris V 75015, Paris, France f INSERM Unite´ 316, CHU Bretonneau, bd Tonnelle´, Tours, France g Service des Urgences Cerebro-Vasculaires, Groupe Hospitalier Pitie´-Salpeˆtrie`re, AP-HP, 75013 Paris, France b

Received 29 March 2004; revised 3 June 2004; accepted 14 June 2004

The underlying neurobiology of autism, a severe pervasive developmental disorder, remains unknown. Few neocortical brain MRI abnormalities have been reported. Using rest functional brain imaging, two independent studies have described localized bilateral temporal hypoperfusion in children with primary autism. In order to search for convergent evidence of anatomical abnormalities in autistic children, we performed an anatomical MRI study using optimized whole-brain voxelbased morphometry (VBM). High-resolution 3-D T1-weighted MRI data sets were acquired in 21 children with primary autism (mean age 9.3 F 2.2 years) and 12 healthy control children (mean age 10.8 F 2.7 years). By comparing autistic children to normal children, we found bilaterally significant decreases of grey matter concentration located in superior temporal sulcus (STS) (P < 0.05 corrected, after small volume correction; SVC). Children with autism were also found to have a decrease of white matter concentration located in the right temporal pole and in cerebellum P < 0.05 corrected, after small volume correction; SV(P < 0.05, corrected) compared to normal children. These results suggest that autism is associated with bilateral anatomical abnormalities localized in the STS and are remarkably consistent with functional hypoperfusion previously reported in children with autism. The multimodal STS areas are involved in highest level of cortical integration of both sensory and limbic information. Moreover, the STS is now recognized as a key cortical area of the ‘‘social brain’’ and is implicated in social perceptual skills that are characteristically impaired in autism. Therefore, the convergent anatomical and functional temporal abnormalities observed in autism may be important in the understanding of brain behavior relationships in this severe developmental disorder. D 2004 Elsevier Inc. All rights reserved. Keywords: Autism; Superior temporal sulcus; Morphometry

* Corresponding author. Pediatric Radiology Department, Necker Enfants-Malades Hospital, 149 rue de Se`vres 75015 Paris, France. Fax: +33-1-44-49-51-70. E-mail address: [email protected] (N. Boddaert). Available online on ScienceDirect (www.sciencedirect.com.) 1053-8119/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2004.06.016

Introduction Autism is a complex, severe, and lifelong developmental disorder. Its main symptoms are social interaction and communication deficits (Kanner, 1943; Rapin, 1997). Autistic children have difficulties in processing emotional expressions and have narrow interests and poor imagination (Gillberg and Coleman, 1992). Since the first description by Leo Kanner in 1943, autism has intrigued the medical and scientific world because autism associates severe cognitive – behavioral problems in the absence of marked consistent cerebral dysmorphology. Therefore, a fundamental goal of any neurobiological study of autism is a description of brain regions that are of abnormal structure or dysfunctional. Once identified and the abnormalities characterized, better strategies for early diagnosis and treatment of autism may follow. The first MRI studies of autism were published at the end of the 1980s (Courchesne et al., 1988; Gaffney et al., 1987). Since these pioneering studies, about 200 studies have appeared in the literature. Most of them have focused on specific structures such as the cerebellum, the amygdala, the hippocampus, and the corpus callosum, but subtle neocortical changes have not been systematically searched. The cerebellum is one of the most studied structures in autism since a pioneer quantitative MRI study has showed evidence of hypoplasia of the vermian lobules VI and VII in a group of autistic adults (Courchesne et al., 1988), but other groups failed to replicate these findings (Piven et al., 1992, 1997). Some studies show increased volume of the amygdala in adults with autism (Howard et al., 2000) or in mentally retarded (MR) autistic children (Sparks et al., 2002), some have described a decreased volume in autistic adults (Aylward et al., 1999), and yet others reveal no significant abnormalities in MR or non-MR adults with autism (Haznedar et al., 2000). Likewise, to date no consistent hippocampal findings have been reported in autism. Concerning the hippocampus, some studies show no abnormalities in high-functioning and MR adults and children with autism

N. Boddaert et al. / NeuroImage 23 (2004) 364–369

(Haznedar et al., 2000; Howard et al., 2000; Piven et al., 1998; Saitoh et al., 1995), others report decreased volume in autistic adolescent and adults (Aylward et al., 1999) or increased volumes in MR autistic children (Sparks et al., 2002). Morphometric MRI studies have reported a small corpus callosum in MR adolescent patients with autism (Egaas et al., 1995; Manes et al., 1999) and in adults (Hardan et al., 2000; Piven et al., 1997). Finally several groups reported an increased total brain volume in children and adults with autism (Courchesne et al., 2001, 2003; Herbert et al., 2003; Piven et al., 1992, 1995, 1996; Sparks et al., 2002). Recently, quantitative structural imaging studies have benefited greatly from both new technologies for data acquisition and new approaches to image analysis. In addition, these newer methods are more adequate for the study of complex and subtle neocortical abnormalities, and some very recent results are very promising. Using a parametric mesh-based analytic technique to create a threedimensional model of the cerebral cortex and using detailed maps of 22 major sulci in stereotaxic space, Levitt et al. (2003) showed significant differences in cortical sulcal patterns in children with autism localized mainly in the frontal and temporal sulci. In the functional domain, most brain imaging studies in autism were performed with activation paradigms (for a review, see Cody et al., 2002) but few studies were performed at rest. Recently, two independent high-resolution PET and SPECT studies have described localized bilateral temporal hypoperfusion in children with primary autism. These rest functional abnormalities were centered in the superior temporal sulcus (STS) and superior temporal gyrus (STG) (Ohnishi et al., 2000; Zilbovicius et al., 2000). In order to search for convergent evidence of temporal anatomical abnormalities in autistic children, we performed an anatomical MRI study using whole-brain voxel-based morphometry. In contrast to techniques relying on inspection and manual demarcation of structures, VBM is unbiased toward particular regions. All of the stages of image processing are automated and the software is widely available (SPM99, Wellcome Department of Imaging Neuroscience, London, UK, http://www.fil.ion.ucl.ac.uk/ spm/). It enables detection of subtle changes between two groups. A pioneering study in high-functioning adults with autism (Asperger’s syndrome) using VBM was published in 1999 by Abell et al. (1999) showing frontotemporal grey matter abnormalities. Since this publication, VBM has benefited from substantial improvements. Therefore, we used the optimized VBM method (Good et al., 2001) to examine possible localized abnormalities in grey and white matter concentration in children with primary autism. On the basis of previous functional data (Ohnishi et al., 2000; Zilbovicius et al., 2000), we predicted that MRI abnormalities may be localized in the temporal lobes.

Method Subjects Twenty-one children with a primary autistic disorder, of which 16 were boys, were selected among patients attending pediatric psychiatry outpatient units. They were aged from 7 to 15 years (mean age 9.3 F 2.2 years). All children met the DSM-IV diagnosis criteria for autistic disorder. The Autism Diagnostic Interview—Revised (ADI-R) was performed in 18 autistic children and confirmed the diagnosis (Table 1) (Lord et al., 1994). The 21 autistic children were also evaluated with the Behavior Summa-

365

Table 1 ADI-R indicates Autism Diagnostic Interview—Revised Scores for the ADI-R

Mean score

F

Range

ADI-R ADI-R ADI-R ADI-R ADI-R

24.4 11.9 17.3 6.7 4.7

5.6 2.6 2.5 2.2 1.0

13 – 30 7 – 14 15 – 20 3 – 11 1–5

social domain nonverbal communication communication domain stereotypy domain onset

rized Evaluation Scale (Barthelemy et al., 1997). The three autistic children who were not tested with the ADI-R were not different from the others using this scale. Mental retardation was assessed by intellectual quotient (IQ) or developmental quotient (DQ) determined with the Wechsler Intelligence Scale for Children (WISC-R) or the Brunet – Le´zine developmental test (Brunet and Le´zine, 1976). The mean IQ and DQ global values were 41.9 F 21.3. For 11 autistic children, IQ was assessed with WISC-R, the mean IQ was 55.8 F 15.8; range from 40 to 91. For 10 autistic children, who had more severe mental retardation, we used the Brunet – Lezine developmental test; the mean DQ was 24 F 9; range from 12 to 39. Meticulous clinical evaluation allowed selection of a homogeneous group of children with severe primary autism without neurological disorder or seizure. We excluded from study patients with (1) known infectious, metabolic, or genetic diseases; (2) chromosomal abnormality; (3) seizures; (4) identifiable neurological syndromes or focal signs; and (5) motion artefacts on MRI. All children were tested to look at chromosomal abnormalities and none of them had fragile X syndrome. All subjects were free of medication for at least a month before MRI. An ethics committee approved the study and all examinations were performed with the written informed consent of parents. Children with autism were compared to a control group composed by 12 healthy children, 7 boys, aged from 7 to 15 years (mean age 10.8 F 2.7 years). None of them had a history of neurological or psychiatric disorders and they all had normal schooling. The autistic and control children groups were matched for age (P = 0.1) and sex (P = 0.3). Subjects with prominent normal anatomical variants (e.g., mega cisterna magna, cavum septum pellucidum) or technical artifacts on MRI were excluded (12 autistic children). In addition, three autistic children were excluded because of MRI abnormalities (one child with periventricular leukomalacia, two children with major ventricular dilatation). Brain imaging Magnetic resonance imaging was performed on 1.5 T Signa General Electric scanner using a 3-D T1-weighted FSPGR sequence (TR/TE/TI/NEX: 10.5/2.2/600/1, flip angle 10j, matrix size 256  192, yielding 124 axial slices, and a thickness of 1.2 mm, field of view 22 cm). In addition, a conventional MRI examination was performed for each child including T2 and FLAIR sequences. The MRI examination lasted 15 min. In all autistic children, MRI studies were performed during sleep induced by premedication (7 mg/kg of sodium pentobarbital). All images were visually inspected by two pediatric neuroradiologists (NB and FB). Images were analyzed on a Sun Ultra 10 workstation (Sun Microsystems, Mountain view, CA) using MATLAB 5.3 (Math-

366

N. Boddaert et al. / NeuroImage 23 (2004) 364–369

Works, Natick, MA) and SPM 99. Optimized voxel-based morphometric (VBM) analysis includes five steps (Good et al., 2001). (1) Customized templates: The creation of separate grey and white matter MRI templates of normal children in our control group in order to reduce any potential bias in spatial normalization. (2) Segmentation and extraction of a brain image: This procedure involves segmentation of the original structural MR images into grey and white matter images in native space. (3) Normalization of grey or white matter images using our customized template: The extracted segmented grey or white matter images are normalized to our grey or white matter templates thus preventing any contribution of nonbrain voxels and affording optimal spatial normalization of grey or white matter. This procedure corrects for global changes in brain size. In order to facilitate optimal segmentation, the optimized normalization parameters are reapplied to the original whole-brain structural images (in native space). (4) Segmentation of normalized wholebrain images: The optimally normalized whole-brain images are then segmented into grey and white matter, CSF and non-CSF partitions. (5) Smoothing: Optimally normalized and segmented images are smoothed with a 12-mm full-width half-maximum (FWHM) kernel.

VBM analysis Grey matter abnormalities By comparing the autistic group to normals, the exploratory analysis (P < 0.001, uncorrected for multiple comparisons, Z score > 3.15) revealed significant decrease in grey matter concentration only in the temporal lobes bilaterally. No abnormality was found outside the temporal lobes. The grey matter decrease was precisely located on the fundus of the superior temporal sulci (STS) corresponding to Brodmann area 21 (Fig. 1 and Table 2). By performing the small volume correction for the superior temporal regions, we have observed a significant bilateral grey matter decrease (P < 0.05 corrected for multiple comparisons) located exactly in the same STS regions. White matter abnormalities Significant decreases in white matter concentration were found in the autistic group compared to normal children in the right pole of the temporal lobe close to Brodmann area 38 and in the left cerebellar hemisphere (P < 0.05, corrected for multiple comparisons) (Table 3).

Statistical analysis

Discussion

Regionally specific differences in grey and in white matter between groups were assessed statistically using a two-tailed contrast testing for an increased or decreased probability of a voxel being grey or white matter; concentration changes were assessed using segmented images. Normalization for global differences in voxel intensity across scans was done by inclusion of the global mean voxel value as a confounding covariate in an analysis of covariance (ANCOVA) while preserving regional differences in grey or white matter. For the grey matter analysis, we have first performed an exploratory analysis using height threshold at P < 0.001 uncorrected for multiple comparisons, extension threshold for 40 voxels. Secondly, in order to search for grey matter abnormality precisely in the temporal regions that were previously found to be functionally abnormal in autistic children, we used a small volume correction (SVC) for this region. This small volume was determined using a mask derived from the previous rest PET study (Zilbovicius et al., 2000). This mask was obtained with SPM99 from the contrast of mentally retarded children versus autistic children (P < 0.05 corrected for cluster level, previously published by Zilbovicius et al., 2000), defining the region of rest hypoperfusion in autistic children (search in 5793 voxels; see Fig. 2b). So, for the grey matter analysis, we have secondly performed an analysis using P < 0.05 corrected for multiple comparisons with SVC. For the white matter analysis, because there was not a priori hypothesis, only voxels surviving the correction for multiple comparisons at P < 0.05 were reported.

Using anatomical MRI and optimized voxel-based morphometry, we found bilateral grey matter decrease localized in the superior temporal lobes in children with primary autism. Moreover, the temporal superior sulcal grey matter abnormalities observed in children with autism are entirely consistent with previously rest functional brain imaging data obtained in autism (Zilbovicius et al., 2000). Fig. 2 shows that the grey matter decrease (Fig. 2a) we report in children with autism is almost identically located to those of previously described PET functional hypoperfusion (Fig. 2b) (Zilbovicius et al., 2000). The control groups were different in the two studies (MRI and PET). Only five autistic children were

Results

Fig. 1. MRI grey abnormalities in autistic children. Brain areas with significant grey matter decreases in autistic children compared to normal control children are superimposed on left and right lateral surfaces of a rendering of the T1-weighted anatomical template image in Talairach space. A statistical threshold of Z > 3.15 (P < 0.001) was used for display purposes; peaks reaching statistical significance are listed in Table 2. Plots show relative normalized grey matter concentration for each autistic child (green dot) and for each normal child (red diamond).

Visual inspection of the MRI Clinical review of anatomical MRI was performed by two pediatric neuroradiologists (NB and FB), and no abnormality was found in both autistic and healthy children.

N. Boddaert et al. / NeuroImage 23 (2004) 364–369

367

Table 2 Decrease in grey matter concentration in autistic children compared to normal control children Anatomical region

Voxels numbers

Z score

Right superior temporal sulcus (BA 21)*

440

3.75

38

24

4

196

3.49 3.15 3.59

40 51 38

34 36 27

6 9 3

Left superior temporal sulcus (BA 21)*

Talairach coordinates x

y

z

Approximate Brodmann area (BA) numbers associated with anatomical regions are given in parentheses. Z score as well as coordinates in Talairach stereotaxic space correspond to local maxima; x, distance (mm) to right (+) or left ( ) of the midsagittal line; y, distance anterior (+) or posterior ( ) to vertical plane through the anterior commissure; and z, distance above (+) or below ( ) the intercommissural (AC-PC) line. * P < 0.05 corrected for multiple comparisons with SVC.

included in both the present MRI and previous PET studies. The exclusion of these five children does not change the results, indicating that the convergence is not due to subject overlap. The two groups were still matched for age and gender even after the five overlapping children were removed. These bilateral temporal abnormalities are also very close to previously described abnormalities demonstrated with SPECT (Ohnishi et al., 2000). However, the present study has methodological limitation concerning the choice of the control group. The group of autistic children we have studied had an associated mental retardation, as the majority of the autistic children and the control group were not matched for the IQ. Therefore, these preliminary results need to be replicated in a larger IQ-matched population of autistic and control children. Nevertheless, previous functional PET and SPECT results indicating temporal abnormality in autism were obtained by comparing IQ-matched autistic and mentally retarded control children. In this study, the grey matter anomalies are centered on the superior temporal sulcus (STS), which is increasingly recognized as a key component of the ‘‘social brain’’ (Allison et al., 2000). Neuroimaging studies in normal subjects and single-cell recordings in monkeys have emphasized the role of this structure in the processing of biological movements, including movements of the eyes, mouth, hands, and body, and in social perception (Allison et Table 3 Decrease in white matter concentration in autistic children compared to normal control children Anatomical region

Voxels numbers

Z score

Talairach coordinates x

y

Right temporal pole (BA 38)* Left cerebellum*

814

4.90

50

8

1209

4.99

34

z 28 60

39

Approximate Brodmann area (BA) numbers associated with anatomical regions are given in parentheses. Z score as well as coordinates in Talairach stereotaxic space correspond to local maxima; x, distance (mm) to right (+) or left ( ) of the midsagittal line; y, distance anterior (+) or posterior ( ) to vertical plane through the anterior commissure; and z, distance above (+) or below ( ) the intercommissural (AC-PC) line. * P < 0.05 corrected for multiple comparisons.

Fig. 2. Convergent anatomical and functional temporal lobe anomalies in children with primary autism. (a) SPM glass brain represents the superior temporal regions where autistic children had a significant decrease of grey matter concentration. (b) The same region had a significant decrease of regional cerebral blood flow measured with PET (Zilbovicius et al., 2000). The control groups were different in the two studies, and only five autistic children were included in both MRI and PET studies. The exclusion of these five children did not modify the results, indicating that the convergent results are not due to subject overlap.

al., 2000; Blakemore and Decety, 2001; Hietanen and Perrett, 1996). Children with autism have deficits in the perception of eye gaze, poor eye contact during communication, and difficulties accessing information to infer the mental state of others (Howard et al., 2000). ‘‘I had no idea that other people communicated through subtle eye movements,’’ says an adult with autism, ‘‘until I read it in a magazine five years ago.’’ Such a capacity may be a prerequisite for higher level appreciation of the minds of others and is part of a larger cognitive domain called ‘‘theory of mind’’ or social cognition, which is severely impaired in autism (BaronCohen et al., 1999; Frith, 2001; Happe et al., 1996). There is also evidence that the STS is implicated in successful imitation (Rizzolatti et al., 2001) and in human voice perception (Belin et al., 2000), essential skills for interpersonal communication. Deficits of imitation and voice perception are thought to be associated with autism (Rapin, 1997; Williams et al., 2001). Most of the input to the STS is derived from third-order sensory association areas and others polymodal areas (parietal, prefrontal cortex, limbic, and paralimbic regions) (Barnes and Pandya, 1992; Seltzer and Pandya, 1994), suggesting that multimodal STS areas are involved in the highest level of cortical integration of both sensory and limbic information. Superior temporal areas would therefore be primordial in the construction of coherent internal representations since these areas are central to the integration of complex perceptual multimodal information (Gloor, 1997). STS damage could yield a set of complex cognitive and affective deficits that are characteristic of autism. Therefore, we believe that the bilateral anatomical grey matter abnormalities of the STS regions and the associated functional alterations may play a role in the emergence of autistic symptoms. However, it is important to acknowledge that the functional significance of reduced grey matter is unknown. The type of cells involved in the grey matter reduction is unknown and the reduced grey matter could be due to increased white matter within the structure. Our results agree with earlier data implicating the temporal lobes in acquired clinical models of autism (Bolton and Griffiths, 1997; Chugani et al., 1996; Gillberg, 1991). Autistic behavior has been reported in temporal lobe pathology, such as epilepsy and herpes simplex encephalitis (Gillberg, 1991). Recent neuroimaging

368

N. Boddaert et al. / NeuroImage 23 (2004) 364–369

studies have shown an association between temporal lobe abnormalities and the occurrence of secondary autism (Bolton and Griffiths, 1997; Chugani et al., 1996). In addition, neuropathological studies showed temporal lobes abnormalities in patients with autism (Bauman and Kemper, 1985; Casanova et al., 2002). All of these behavioral, lesional, structural, and functional brain imaging studies at rest suggest a link between temporal lobe dysfunction and autistic behavior. The finding of white matter abnormalities was more unexpected and remains therefore to be confirmed. However, in the temporal poles, these abnormalities may reflect a reduction of fiber bundles connecting the STS with limbic and paralimbic regions (Barnes and Pandya, 1992; Seltzer and Pandya, 1994). These regions are also involved in social and emotional skills (Adolphs, 1999) and may be impaired in autism since recent functional MRI studies in autism have shown an abnormal pattern of amygdala activation during emotional analysis of faces in autistic patients (BaronCohen et al., 1999). Furthermore, Castelli et al. (2002) have shown abnormal patterns of temporal pole activation in autistic subjects while watching scenarios that normally evoke mental state attributions. In addition, the lateral cerebellum is also concerned with the specific recognition of ‘‘biological motion’’ and imitation of action (Blakemore and Decety, 2001). In conclusion, we found bilateral temporal grey matter decrease in children with primary autism, which are consistent with recently described bilateral temporal lobe hypoperfusion in autistic children. The localization of both functional and anatomical abnormalities is highly consistent with the typical symptoms of autism since they are centered on the STS, a key cortical area for social interaction, emotional reactions, and for interpersonal communication.

Acknowledgments Supported by Programme Hospitalier de Recherche CliniqueMiniste`re de la Sante´ (France), France-Te´lecom Foundation, and France Foundation.

References Abell, F., Krams, M., Ashburner, J., Passingham, R., Friston, K., Frackowiak, R.S.J., Happe, F., Frith, C., Frith, U., 1999. The neuroanatomy of autism: a voxel-based whole brain analysis of structural scans. NeuroReport 10, 1647 – 1651. Adolphs, R., 1999. Social cognition and the human brain. Trends. Cogn. Sci. 3, 469 – 479. Allison, T., Puce, A., McCarthy, G., 2000. Social perception from visual cues: role of the STS region. Trends. Cogn. Sci. 4, 267 – 278. Aylward, E., Minshew, N., Goldstein, G., Honeycutt, N., Augusting, A., Yates, K., Barta, P.E., Pearlson, G.D., 1999. MRI volumes of amygdala and hippocampus in non-mentally retarded autistic adolescents and adults. Neurology 53, 2145 – 2150. Barnes, C.L., Pandya, D.N., 1992. Efferent cortical connections of multimodal cortex of the superior temporal sulcus in the rhesus monkey. J. Comp. Neurol. 318, 222 – 244. Baron-Cohen, S., Ring, H.A., Wheelwright, S., Bullmore, E.T., Brammer, M.J., Simmons, A., Williams, S.C., 1999. Social intelligence in the normal and autistic brain: an fMRI study. Eur. J. Neurosci. 11, 1891 – 1898. Barthelemy, C., Roux, S., Adrien, J.L., Hameury, L., Guerin, P., Garreau, B., Fermanian, J., Lelord, G., 1997. Validation of the revised Behavior Summarized Evaluation Scale. J. Autism Dev. Disord. 27, 139 – 153.

Bauman, M., Kemper, T.L., 1985. Histoanatomic observations of the brain in early infantile autism. Neurology 35, 866 – 874. Belin, P., Zatorre, R.J., Lafaille, P., Ahad, P., Pike, B., 2000. Voice-selective areas in human auditory cortex. Nature 403, 309 – 312. Blakemore, S.J., Decety, J., 2001. From the perception of action to the understanding of intention. Nat. Rev., Neurosci. 2, 561 – 567. Bolton, P.F., Griffiths, P.D., 1997. Association of tuberous sclerosis of temporal lobes with autism and atypical autism. Lancet 349, 392 – 395. Brunet, O., Le´zine, I., 1976. Echelle de Developpement Psychomoteur de la Premiere Enfance, second ed. PUF, Paris. Casanova, M.F., Buxhoeveden, D.P., Switala, A.E., Roy, E., 2002. Minicolumnar pathology in autism. Neurology 58, 428 – 432. Castelli, F., Frith, C., Happe, F., Frith, U., 2002. Autism, Asperger syndrome and brain mechanisms for the attribution of mental states to animated shapes. Brain 125, 1839 – 1849. Chugani, H.T., Da Silva, E., Chugani, D.C., 1996. Infantile spasms: III. Prognostic implications of bitemporal hypometabolism on positron emission tomography. Ann. Neurol. 39, 643 – 649. Cody, H., Pelphrey, K., Piven, J., 2002. Structural and functional magnetic resonance imaging of autism. Int. J. Dev. Neurosci. 20, 421 – 438 (Review). Courchesne, E., Yeung-Courchesne, R., Press, G., Hesselink, J., Jernigan, T., 1988. Hypoplasia of cerebellar vermal lobules VI and VII in autism. N. Engl. J. Med. 318, 1349 – 1954. Courchesne, E., Karns, C., Davis, H., Ziccardi, R., Carper, R., Tigue, Z., Chisum, H.J., Moses, P., Pierce, K., Lord, C., Lincoln, A.J., Pizzo, S., Schreibman, L., Haas, R.H., Akshoomoff, N.A., Courchesne, R.Y., 2001. Unusual brain growth patterns in early life in patients with autistic disorder: an MRI study. Neurology 57, 245 – 254. Courchesne, E., Carper, R., Akshoomoff, N., 2003. Evidence of brain overgrowth in the first year of life in autism. JAMA 290, 337 – 344. Egaas, B., Courchesne, E., Saitoh, O., 1995. Reduced size of the corpus callosum in autism. Arch. Neurol. 52, 794 – 801. Frith, U., 2001. Mind blindness and the brain in autism. Neuron 32, 969 – 979. Gaffney, G.R., Tsai, L.Y., Kuperman, S., Minchin, S., 1987. Cerebellar structure in autism. Am. J. Dis. Child. 141, 1330 – 1332. Gillberg, I.C., 1991. Autistic syndrome with onset at age 31 years: herpes encephalitis as a possible model for childhood autism. Dev. Med. Child Neurol. 33, 920 – 924. Gillberg, C., Coleman, M., 1992. The Biology of Autistic Syndromes, second ed. Mac Keith Press, London. Gloor, P., 1997. The Temporal Lobe and Limbic System Oxford Univ. Press, New York. Good, C.D., Johnsrude, I.S., Ashburner, J., Henson, R.N., Friston, K.J., Frackowiak, R.S., 2001. A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 14, 21 – 36. Happe, F., Ehlers, S., Fletcher, P., Frith, U., Johansson, M., Gillberg, C., Dolan, R., Frackowiak, R., Frith, C., 1996. ‘Theory of mind’ in the brain. Evidence from a PET scan study of Asperger syndrome. NeuroReport 8, 197 – 201. Hardan, A., Minshew, N., Keshavan, M., 2000. Corpus callosum size in autism. Neurology 55, 1033 – 1036. Haznedar, M., Buchsbaum, M., Wei, T., Hof, P., Cartwright, C., Bienstock, C., 2000. Limbic circuitry in patients with autism spectrum disorders studied with positron emission tomography and magnetic resonance imaging. Am. J. Psychiatry 157, 1994 – 2001. Herbert, M., Ziegler, D., Deutsch, C., O’Brien, L., Lange, N., Bakardjiev, A., Hodgson, J., Adrien, K.T., Steele, S., Makris, N., Kennedy, D., Harris, G.J., Caviness Jr., V.S., 2003. Dissociation of cerebral cortex, subcortical and cerebral white matter volumes in autistic boys. Brain 126, 1182 – 1192. Hietanen, J.K., Perrett, D.I., 1996. A comparison of visual responses to object- and ego-motion in the macaque superior temporal polysensory area. Exp. Brain Res. 108, 341 – 345. Howard, M.A., Cowell, P.E., Boucher, J., Broks, P., Mayes, A., Farrant, A., Roberts, N., 2000. Convergent neuroanatomical and behavioural

N. Boddaert et al. / NeuroImage 23 (2004) 364–369 evidence of an amygdala hypothesis of autism. NeuroReport 11, 2931 – 2935. Kanner, L., 1943. Autistic disturbances of affective contact. Nerv. Child 2, 217 – 250. Levitt, J., Blanton, R., Smalley, S., Thompson, P., Guthrie, D., McCracken, J., Sadoun, T., Heinichen, L., Toga, A.W., 2003. Cortical sulcal maps in autism. Cereb. Cortex 13, 728 – 735. Lord, C., Rutter, M., Le Couteur, A., 1994. Autism Diagnostic Interview— Revised: a revised version of a diagnostic interview for caregivers of individuals with possible pervasive developmental disorders. J. Autism Dev. Disord. 24, 659 – 685. Manes, F., Piven, J., Vrancic, D., Nanclares, V., Plebst, C., Starkstein, S., 1999. An MRI study of the corpus callosum and cerebellum in mentally retarded autistic individuals. J. Neuropsychiatry Clin. Neurosci. 11, 470 – 474. Ohnishi, T., Matsuda, H., Hashimoto, T., Kunihiro, T., Nishikawa, M., Uema, T., Sasaki, M., 2000. Abnormal regional cerebral blood flow in childhood autism. Brain 123, 1838 – 1844. Piven, J., Nehme, E., Simon, J., Barta, P., Godfrey, P., Folstein, S., 1992. Magnetic resonance imaging in autism: measurement of the cerebellum, pons, and fourth ventricle. Biol. Psychiatry 31, 491 – 504. Piven, J., Ardnt, S., Bailey, J., Havercamp, S., Andreasen, N., Palmer, P., 1995. An MRI study of brain size in autism. Am. J. Psychiatry 152, 1145 – 1149. Piven, J., Ardnt, S., Bailey, J., Andreasen, N., 1996. Regional brain enlargement in autism: a magnetic resonance imaging study. J. Am. Acad. Child Adolesc. Psychiatry 35, 530 – 536.

369

Piven, J., Bailey, J., Ransom, B., Arndt, S., 1997. An MRI study of the corpus callosum in autism. Am. J. Psychiatry 154, 1051 – 1056. Piven, J., Bailey, J., Ransom, B., Arndt, S., 1998. No difference in hippocampus volume detected on magnetic resonance imaging in autistic individuals. J. Autism Dev. Disord. 28, 105 – 110. Rapin, I., 1997. Autism. N. Engl. J. Med. 337, 97 – 104. Rizzolatti, G., Fogassi, L., Gallese, V., 2001. Neurophysiological mechanisms underlying the understanding and imitation of action. Nat. Rev., Neurosci. 2, 661 – 670. Saitoh, O., Courchesne, E., Egaas, B., Lincoln, A., Schreibman, L., 1995. Cross-sectional area of the posterior hippocampus in autistic patients with cerebellar and corpus callosum abnormalities. Neurology 45, 317 – 324. Seltzer, B., Pandya, D.N., 1994. Parietal, temporal, and occipital projections to cortex of the superior temporal sulcus in the rhesus monkey: a retrograde tracer study. J. Comp. Neurol. 343, 445 – 463. Sparks, B., Friedman, S., Shaw, D., Aylward, E., Echelard, D., Artru, A., Maravilla, K.R., Giedd, J.N., Munson, J., Dawson, G., Dager, S.R., 2002. Brain structural abnormalities in young children with autism spectrum disorder. Neurology 59, 184 – 192. Williams, J.H., Whiten, A., Suddendorf, T., Perrett, D.I., 2001. Imitation, mirror neurons and autism. Neurosci. Biobehav. Rev. 25, 287 – 295. Zilbovicius, M., Boddaert, N., Belin, P., Poline, J.B., Remy, P., Mangin, J.F., Thivard, L., Barthelemy, C., Samson, Y., 2000. Temporal lobe dysfunction in childhood autism: a PET study. Am. J. Psychiatry 157, 1988 – 1993.