Untitled

How somatic cortical maps differ in autistic and typical brainsMehmet Akif Coskuna, Larry Varghesea, Stacy Reddochc, Eduardo M. Castillod,Deborah A. Pearsonc, Katherine A. Lovelandc, Andrew C. Papanicolaoudand Bhavin R. Shetha,b The comorbidity of ‘core characteristics’ and sensorimotor abnormalities in autism implies abnormalities in brain Health | Lippincott Williams & Wilkins.
development of a general and pervasive nature and atypical organization of sensory cortex. By using magnetoencephalography, we examined the cortical Keywords: autism, cortical organization, magnetoencephalography, response to passive tactile stimulation of the thumb and index finger of the dominant hand and lip of the individuals with autism spectrum disorder and typically developing aDepartment of Electrical and Computer Engineering, bCenter forNeuroEngineering and Cognitive Systems, University of Houston, Houston, persons. The distance between the cortical representations Departments of cPsychiatry and Behavioral Sciences and dPediatrics, The of thumb and the lip was significantly larger in the autism University of Texas Health Science Center at Houston, Houston, Texas, USA group than in typicals. Moreover, in cortex, the thumb is Correspondence to Bhavin R. Sheth, PhD, University of Houston, Houston, typically closer to the lip than the index finger. This was not observed in persons with autism. Our findings are arguably Tel: + 1 713 743 4935; fax: + 1 713 743 4444; e-mail: [email protected] the first demonstration of abnormality in sensory organization in the brains of persons with Received 25 September 2008 accepted 17 October 2008 individuals in the autism group met our research criteria Children with autistic spectrum disorders often have for an autism spectrum disorder, as determined by using unusual reactions to certain sensory stimuli. Across the Autism Diagnostic Observation Schedule [10] and reports, 44–88% of individuals with autism have abnormal Autism Diagnostic Interview, Revised [11] administered sensitivity to tactile stimuli [1]. It is possible that by trained clinicians. Five individuals in the autism group abnormal responsivity to touch is a sensory perceptual were clinically classified as pervasive developmental problem (see Ref. [2]) or stems from some deviation in disorder-not otherwise specified, one as Asperger syn- sensory cortical organization. Of importance, these drome, and the remaining eight as autistic disorder. Two unusual behaviors appear early in development during persons with autism and three typicals were female.
the time the somatic map is being formed [3]. The Potential participants were excluded if there was association of sensorimotor differences and social- evidence of brain injury, seizure disorder, or neurotropic emotional impairments in autism suggests that both infection or disease, or had a history of identified severe areas of function could be affected by abnormalities in psychopathology, such as bipolar disorder, schizophrenia, brain development of a more general nature that have or behavior problems severe enough to make accurate the potential to adversely affect development in and reliable testing difficult. All participants were right multiple systems. As a first step to elucidate this relation- handed as determined by the Edinburgh Handedness ship, we investigated somatotopy in the brains of young Inventory [12]. All in the autism group were high functioning; full-scale IQs (FSIQs) and verbal IQs (VIQs) (MEG). Somatotopic maps of various body parts derived from the Wechsler Abbreviated Scale of Intelli- (e.g. lip, trunk, and shoulder) and even of the individual gence [13] were greater than 85 (mean ± 1 SEM: FSIQ, fingers of a single hand have been obtained in past studies 105 ± 5; VIQ, 103 ± 5; and performance IQ, 105 ± 4).
by using MEG [4–6]. In addition, MEG has also Prior informed consent was obtained from all participants, been used to investigate the biological basis of autism or participants and their parents, under a protocol approved by the University of Texas Health ScienceCenter-Houston and the University of Houston.
Pneumatically driven mechanical taps (1.8 kg/cm2) of (mean ± SEM = 18.6 ± 1.0 years) and 16 typically devel- 40 ms duration (20 ms rise time) were individually oping persons (typicals, 19.5 ± 1.0 years) participated. All applied to the right thumb, right index finger, and the c 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
right lip (4D Neuroimaging Inc., San Diego, California, Data from two of 14 persons with autism and three of 16 USA). Participants were informed that a pressure pulse typicals were removed on the basis of the following two would be delivered and that all they had to do was close criteria: (i) the hot spots for thumb, index finger, and lip their eyes, stay still, and relax. A training block containing were not in line with known somatotopic order. According five stimuli before the actual recording familiarized the to known topography of somatosensory cortex, the hot spot corresponding to the index finger is superior to thehot spots of the thumb and lip, and the hot spotcorresponding to the thumb is superior to that of the lip Data acquisition and magnetoencephalography analysis [6,15]. Among excluded participants, one or both were All MEG recordings used a whole-head neuromagnet- not observed; (ii) the automated dipole estimation ometer containing an array of 248 gradiometers (Magnes method failed to obtain a reasonable hot spot location.
WH3600, 4D Neuroimaging Inc.). The instruments were This usually occurred because a spurious extremum placed in a magnetically shielded and sound-attenuated occurred in the contour map, which is the automated room (Vacuumschmelze Gmbh & Co., KG, Hanau, method used for dipole calculation. Excluded partici- Germany). There were 2000 epochs of stimulation of pants, thus, had abnormally large or small interhot spot the index finger and lip each and 700 epochs of distance. Manual estimation yielded reasonable hot spot stimulation of the thumb in separate blocks. A single locations but we did not use those estimates to maintain epoch lasted 575 ms, and included a 120-ms prestimulus objectivity. Of interest, the present findings are robust to baseline. The signal was high-pass analog filtered (1.0 Hz either method: manual estimation yielded statistically cutoff) and the data acquired at a rate of 32 kHz significant findings similar to the findings obtained using were decimated to a final sampling rate of 290 Hz using a two-stage decimating digital FIR low-pass filter (100 Hzcutoff). Any epochs that contained exaggerated moments,such as eye blinks ( > 2pT) were discarded. Portions of the signal that were correlated to sensors placed far away Statistical analysis used SPSS 15 for Windows (SPSS Inc., from the head were likely to be noise, so were subtracted Chicago, Illinois, USA). Student’s t-tests (two-tailed) were out. Remaining epochs were ensemble averaged.
used to examine group differences in age, IQ, head volume,and distance between the hot spots of thumb, index finger,and lip. Linear correlations between interhot spot distance The representation of a body part in cortex, or cortical and a host of variables related to IQ and brain volume1/3 ‘hot spot,’ was the location of the earliest equivalent were computed and analyzed for significance.
current source, or dipole, of the evoked contralateralcortical response to its tactile stimulation. To avoidexperiment bias and thereby enhance replicability and objectivity of the process, hot spots corresponding to Stimulation evoked a characteristic cortical response with stimulation of the thumb, index finger, and lip were several components at varying latencies that began with localized in the brain by using a fully automated method a strong response in the contralateral cortex 60–90 ms validated earlier [14]. Only current sources that had following the touch. The evoked cortical MEG response a field variance greater than 94% and a localized volume was modeled as coming from a single source or column(s) less than 20 cm3 were accepted. The pairwise distances in cortex. The calculations yielded the location of the between the centers of the hotspots of the thumb, index predominant site of activation in cortex in response to finger, and lower lip thus obtained were compared stimulation of the given body part, that is the cortical hot spot. We examined the distance in cortex between thehot spots of the lip and of each of the fingers.
To estimate brain volume, we modeled the head asa rectangular prism; that is to say, each axis along the head First, we measured the reliability of the cortical maps.
constituted a dimension of the prism. We used the The raw data obtained from each participant were split fiducial points and head shape digitization points to into two equal halves, or blocks, of trials corresponding to calculate the height, width, and length of the prism.
early and later recordings. For each block, we computed Prism width is defined as the anteroposterior distance the distance between the cortical representations of between the location of the nasion on the forehead thumb and index finger on the one hand and the lip on (bridge of the nose) and the most posterior point at the the other. The discrepancy between the two blocks was back of the head; prism length is the distance between less than 0.2 cm on average and was not significantly the preauricular (ear) coil locations corresponding to the different from 0 for each pairwise comparison (P > 0.8 on external meati along the mediolateral axis; prism height is each). Thus, our analysis indicates that reliable and stable the distance between the most superior point on the cortical maps were obtained, in line with past studies conducted by us showing that somatic maps obtained Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Somatic cortical maps in autism Coskun et al.
with MEG are reliable and highly concordant with more was not uniform, however; the distance between the cortical hot spots corresponding to the index fingerand lip was numerically (Fig. 1b, inset), though not Next, we compared somatic distances between the two significantly, smaller on average in the autism group groups of participants by using all available data.
The distance between the representations in cortex of P = 0.18, two-tailed t-test). The inhomogeneity in thumb and lip was about 38% greater on average in the somatic map extent was revealed in a clearer way, autism group (2.26 ± 0.13 cms) compared with typical when the ratios of the index finger–lip distance to the (1.63 ± 0.08 cm; Fig. 1a). The difference was significant thumb–lip distance for the two groups were compared.
[t(23) = 3.97, P < 0.001, two-tailed t-test]. This expan- Typically in cortex, the representation of the thumb as sion in cortical distance in the brains of the autism group compared with that of the index finger is significantlycloser to the representation of the lip (thumb – lip/indexfinger – lip = 0.74 in our typical sample). In contrast, the cortical representations of the thumb and index fingerwere nearly equidistant from that of the lip in our autism group (thumb – lip/index finger – lip = 0.99). The ratio ofdistance between thumb and lip hot spots on the one handand index finger and lip on the other, that is (thumb – lip)/ (index finger – lip), in the group of typicals was significantly different from one (0.74; P < 0.01; Fig. 1b), but not for the autism group (0.99; P > 0.8; Fig. 1b) and the ratio between the groups was significantly different as well (P = 0.003).
The difference in somatic map extent in autism versus typical is not attributable to differences in brain size at the time of the recordings. First, although our autism grouphad slightly larger brain volumes (4295 ± 112 cm3) than the typicals (4004 ± 97 cm3), but it was not significantlyso. Second, interhot spot distance (thumb – lip) and brainsize (brain volume1/3) were not significantly correlated in either of the two groups (R2 < 0.01, P > 0.4). Finally, even after normalization by head size, thereby yielding a dimensionless ratio – (thumb-lip)/(brain volume1/3) – thumb and lip hot spots remained significantly more distant in the autism group compared with typicals (P = 0.03).
Thus, differences in brain volume at the time of the recordings between the two groups could not entirely explain the larger distance between the thumb and lip The difference in somatic map extent in autism versus typical is also not attributable to differences in IQ. The autism group was all high-functioning individuals with FSIQs and VIQs greater than 85, and mean FSIQ, VIQ, andperformance IQ above 100 (average IQ). Nonetheless, our typicals had significantly higher IQs. Of importance,however, interhot spot distance (thumb – lip) was not significantly correlated with VIQ, performance IQ, or FSIQ in either group (R2 < 0.1, P > 0.4 in all six cases).
Somatic map distances in autism. (a) The distance (mean ± SEM)in centimeters between the representations in somatic cortex corresponding to the right thumb and lip in persons with autismdisorder (AD) (light gray bar) and typicals (black bar). (b) The Cortical representations of the thumb and of the lip in (thumb–lip)/(index finger–lip) cortical distance ratio (mean ± SEM) in our sample of individuals with autism spectrum disorders autism group and typicals is shown. The inset shows the distance in were significantly more distant compared with those of cortex between the representations, or ‘hot spots’, of index finger andlip. *P < 0.05; **P < 0.01; ***P < 0.001.
typically developing individuals. The representation ofthe lip was equidistant from the representations of the Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
thumb and index finger in the cortex of individuals with circuitry, are thus likely to be involved in giving rise to the autism, in contrast to the cortex of typical individuals.
increased distance. Complicating matters further, the This atypical nature of somatic map extent in autism is distance between the lip and the index finger was not among the first demonstrations of an abnormality in larger in the Autism group compared with typical, but sensory organization in the brains of persons with autism.
slightly smaller. In summary, the evidence in totalitysuggests that any relationship between brain growth and Dynamic modulation of functional sensory organization in somatic map development is likely to be complex.
somatosensory cortex has been observed but only whenthe individual is performing a motor task with cognitive demands [16]. Participants in this study did not perform The present findings show subtle but significant any cognitive or motor task; therefore, differences in deviations from normal in the organization of somatosen- cognitive ability or motor activity between the two sory cortex in individuals with autism, and reveal groups, if they exist at all, are unlikely to play any role abnormalities in brain development and cortical circuitry in the present findings. The experimental condition in in autism that go beyond brain areas and circuitry this study is equivalent to a rest condition in terms of task corresponding to autism’s ‘core characteristics’.
demand – a common baseline in neuroimaging studies.
Moreover, amplitude and latency, are remarkably robust to change Financial disclosures: the authors report no competing in level of attention [17]. In total, differences in interests. The research was supported by a grant from the attention, motor task, or cognitive demand – if they exist National Alliance for Autism Research – AutismSpeaks at all – are unlikely to account for abnormalities in (BRS). M.A.C. was supported in part by a Presidential functional sensory organization in autism we observed.
fellowship from the University of Houston.
People with autism have abnormally large heads early in childhood, between 1 and 4 years of age; as they near Baranek GT, Parham LD, Bodfish JW. Sensory and motor features in autism: adulthood, the difference in head size compared with assessment and intervention. In: F Volkmar RP, Klin A, Cohen D, editors.
Handbook of autism and pervasive developmental disorders. Hoboken, typical individuals is largely gone [18]. The development New Jersey: Wiley; 2005. pp. 831–857.
of the somatic map occurs very early in life, which is Guclu B, Tanidir C, Mukaddes NM, Unal F. Tactile sensitivity of normal and around the same time that the overgrowth is taking place autistic children. Somatosens Mot Res 2007; 24:21–33.
Baranek GT. Efficacy of sensory and motor interventions for children with in the brains of children with autism. Thus, it appears autism. J Autism Dev Disord 2002; 32:397–422.
likely that the overgrowth in autism has an influence on Baumgartner C, Doppelbauer A, Deecke L, Barth DS, Zeitlhofer J, the development of the somatic map. Rather than Lindinger G, et al. Neuromagnetic investigation of somatotopy of humanhand somatosensory cortex. Exp Brain Res 1991; 87:641–648.
studying a possible outcome of the brain overgrowth as Castillo EM, Simos PG, Wheless JW, Baumgartner JE, Breier JI, Billingsley RL, we are doing here with our young adult participants, it et al. Integrating sensory and motor mapping in a comprehensive MEG would be more suitable to study the relationship between protocol: clinical validity and replicability. Neuroimage 2004; 21:973–983.
Suk J, Ribary U, Cappell J, Yamamoto T, Llinas R. Anatomical localization overgrowth and somatic map abnormality [19]. For this, revealed by MEG recordings of the human somatosensory system.
one would have to replicate this study with young Electroencephalogr Clin Neurophysiol 1991; 78:185–196.
children, and observe the development in sensory Wilson TW, Rojas DC, Reite ML, Teale PD, Rogers SJ. Children andadolescents with autism exhibit reduced MEG steady-state gamma organization longitudinally both during and after the responses. Biol Psychiatry 2007; 62:192–197.
overgrowth has occurred. MEG recordings of young Gage NM, Siegel B, Roberts TP. Cortical auditory system maturational children 5–6 years old and younger have been successfully abnormalities in children with autism disorder: an MEG investigation. BrainRes Dev Brain Res 2003; 144:201–209.
carried out [20–22]. Moreover, there are no cognitive Gage NM, Siegel B, Callen M, Roberts TP. Cortical sound processing in demands placed on our participants. Our paradigm thus children with autism disorder: an MEG investigation. Neuroreport 2003; appears transferable to young children.
Lord C, Rutter M, DiLavore PC, Risi S. Autism Diagnostic ObservationSchedule (ADOS) Manual. Los Angeles: Western Psychological Services; The relationship between brain overgrowth and somatic map development, if one exists, may not be as Rutter M, Le Couteur A, Lord C. Autism Diagnostic Interview – RevisedManual. Los Angeles: Western Psychological Services; 2003.
straightforward as one might imagine. For example, one Oldfield RC. The assessment and analysis of handedness: the Edinburgh is tempted to speculate that the larger brains in people Inventory. Neuropsychologia 1971; 9:97–113.
with autism imply a stretched out somatic map. This Wechsler D. Wechsler Abbreviated Scale of Intelligence Manual.
San Antonio, Texas: The Psychological Corporation; 1999.
would explain the larger distance between the cortical Papanicolaou AC, Pazo-Alvarez P, Castillo EM, Billingsley-Marshall RL, representations of the lip and the thumb in our autism Breier JI, Swank PR, et al. Functional neuroimaging with MEG: normativelanguage profiles. Neuroimage 2006; 33:326–342.
group. However, even when the overgrowth was taken Hlustik P, Solodkin A, Gullapalli RP, Noll DC, Small SL. Somatotopy in into account, the effect, that is, the larger lip–thumb human primary motor and somatosensory hand representations revisited.
distance in the cortical map in the autism group, Schaefer M, Heinze HJ, Rotte M. Task-relevant modulation of primary persisted. Other factors, besides the forces responsible somatosensory cortex suggests a prefrontal-cortical sensory gating system.
for increasing brain volume, such as those affecting Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Somatic cortical maps in autism Coskun et al.
Gomez Gonzalez CM, Clark VP, Fan S, Luck SJ, Hillyard SA. Sources of Fujioka T, Ross B, Kakigi R, Pantev C, Trainor LJ. One year of musical training attention-sensitive visual event-related potentials. Brain Topogr 1994; affects development of auditory cortical-evoked fields in young children.
Brain 2006; 129 (Pt 10):2593–2608.
Redcay E, Courchesne E. When is the brain enlarged in autism? A meta- Paetau R, Hamalainen M, Hari R, Kajola M, Karhu J, Larsen TA, et al.
analysis of all brain size reports. Biol Psychiatry 2005; 58:1–9.
Magnetoencephalographic evaluation of children and adolescents with Courchesne E, Redcay E, Morgan JT, Kennedy DP. Autism at the intractable epilepsy. Epilepsia 1994; 35:275–284.
beginning: microstructural and growth abnormalities underlying the Rojas DC, Maharajh K, Teale PD, Kleman MR, Benkers TL, Carlson JP, et al.
cognitive and behavioral phenotype of autism. Dev Psychopathol 2005; Development of the 40Hz steady state auditory evoked magnetic field from ages 5 to 52. Clin Neurophysiol 2006; 117:110–117.
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Source: http://ece.ee.uh.edu/sites/ece.ee.uh.edu/files/44/files/Somatic%20Maps%20in%20Autism--NeuroReport%202009.pdf