Autism at the beginning: Microstructural and growth ...

Autism at the beginning: Microstructuraland growth abnormalities underlyingthe cognitive and behavioralphenotype of autism

ERIC COURCHESNE,a,b ELIZABETH REDCAY,a JOHN T. MORGAN,a

and DANIEL P. KENNEDYa

aUniversity of California, San Diego; and bChildren’s Hospital Research Center, San Diego

AbstractAutistic symptoms begin in the first years of life, and recent magnetic resonance imaging studies have discoveredbrain growth abnormalities that precede and overlap with the onset of these symptoms. Recent postmortem studiesof the autistic brain provide evidence of cellular abnormalities and processes that may underlie the recentlydiscovered early brain overgrowth and arrest of growth that marks the first years of life in autism. Alternativeorigins and time tables for these cellular defects and processes are discussed. These cellular and growthabnormalities are most pronounced in frontal, cerebellar, and temporal structures that normally mediate thedevelopment of those same higher order social, emotional, speech, language, speech, attention, and cognitivefunctions that characterize autism. Cellular and growth pathologies are milder and perhaps nonexistent in otherstructures ~e.g., occipital cortex!, which are known to mediate functions that are often either mildly affected orentirely unaffected in autistic patients. It is argued that in autism, higher order functions largely fail to developnormally in the first place because frontal, cerebellar, and temporal cellular and growth pathologies occur prior toand during the critical period when these higher order neural systems first begin to form their circuitry. It ishypothesized that microstructural maldevelopment results in local and short distance overconnectivity in frontalcortex that is largely ineffective and in a failure of long-distance cortical–cortical coupling, and thus a reduction infrontal–posterior reciprocal connectivity. This altered circuitry impairs the essential role of frontal cortex inintegrating information from diverse functional systems ~emotional, sensory, autonomic, memory, etc.! andproviding context-based and goal-directed feedback to lower level systems.

Autism begins in many ways. In the first weeksof life, a mother notices something is not rightabout her newborn baby boy: he has markeddifficulty coordinating his sucking and swal-lowing and sometimes he seems floppy andthen at other times strangely rigid. As the first

months go by, he appears to have good visualattention, perhaps too good, because some-times his attention seems stuck. He is alsounexpectedly sensitive to touch or sounds atone moment, but at others he seems almostcompletely oblivious to them. However, he iscuddly and he smiles and coos and seems tobe socially connected. Then, during the sec-ond half of his first year of life, his vocaliza-tions do not continue to develop and motherbecomes truly concerned because now it isnot just his motor, attention, and sensory re-sponses that are awry; he takes a reduced in-terest in her and rarely smiles or looks at her

The authors were supported by funds from the NationalInstitute of Mental Health ~2-ROI-MH36840! and Na-tional Institute of Neurological Disorders and Stroke ~2-ROI-NS19855! awarded to Eric Courchesne.

Address correspondence and reprint requests to: EricCourchesne, Center for Autism Research, 8110 La JollaShores Dr., Suite 201, San Diego, CA 92037; E-mail:[email protected].

Development and Psychopathology 17 ~2005!, 577–597Copyright © 2005 Cambridge University PressPrinted in the United States of AmericaDOI: 10.10170S0954579405050285

577

or anyone else. Instead, he seems to regardobjects with as much interest as people, whomhe regards with disinterest or even avoidance.By his first birthday, a frightening and sadthought occurs to the mother: perhaps my babyhas autism.

Another mother and father swell with joybecause their newborn baby girl is so perfect,and as months come and go, she grows moredelightful. Her interest in others grows andher skills at interacting and engaging othersbrings a sense of wonder to mother and father,and their dreams of what a special person sheis to become expand. Each day, the little tod-dler’s face is filled with different emotions asshe expands her social, emotional, speech, andlanguage skills. She seems as filled with lovefor her parents as they are for her. In the monthsbefore her second birthday, though, mothernotices that she has not been herself lately.What is it? Perhaps a little quieter, perhaps alittle bit less emotional. Worse, as her motherwatches, each day she seems to fade a bitmore until there is no doubt: her little girl isnot herself anymore. She has gradually be-come more and more remote; her face no lon-ger shows much emotion and she no longerdelights in others. In fact, she no longer seeksout mom and dad. She has slowly faded awayfrom them.

Thus, autism begins. In one case describedby Dawson, Osterling, Meltzoff, and Kuhl~2000!, the beginning was early, rapid, andunmistakable, but in another case describedby Lord ~C. Lord, personal communication,NAAR Conference, November 12, 2004!, signsof autism did not appear until later in the sec-ond year of life. New research has identifiedkey behavioral red flags for autism during thefirst 2 years of life ~see Table 1; Wetherby,Woods, Allen, Cleary, Dickinson, & Lord,2004!, and many of them involve a lack ofnormal socioemotional behavior and an appar-ent lack of the normal desire to make socio-emotional contact.

How could the desire for social connectionnot be there in an infant? Or, even more mys-teriously, how could that desire appear stronglyfor a time, only to slowly dwindle away, leav-ing a strange void? What could do this to aquality so essentially human and so essential

for human togetherness? What could snuff outsocial drive, a drive that ought to be as strongas the drive to breath, eat, or survive? Whatgoes wrong in how the brain is organized andoperates, and how might the impact of theseprocesses be mitigated or prevented?

Where (and When) to Start Looking forthe Neurobiological Bases of SocialDysfunction in Autism?

Over the past 40 years, cognitive psycholo-gists have attempted to identify cognitive def-icits underlying the myriad of behavioralsymptoms seen in autism. Contemporary re-search has identified deficits in complex pro-cessing ~Minshew, Goldstein, & Siegel, 1997!,weak central coherence ~Frith & Happe, 1994!,impairment in the dynamic control of orient-ing, disengaging, and switching attention~Courchesne, Townsend, Akshoomoff, Sai-toh, Yeung–Courchesne, Lincoln, James,Haas, Schreibman, & Lau, 1994; Townsend,Courchesne, Covington, Westerfield, Harris,Lyden, Lowry, & Press, 1999!, deficits in ex-ecutive functions ~Rogers & Pennington,1991!, deficits in theory of mind ~Baron–Cohen, Leslie, & Frith, 1985!, deficits inimitation ~Rogers & Pennington, 1991!, im-pairments in social and affective relations~Hobson, 1993!, and impairments in joint so-cial attention ~Mundy, 1995!. However, themajority of this research was not aimed atidentifying the specific neural bases of thisdisorder. With the use of cognitive neurosci-ence techniques ~i.e., functional magnetic res-

Table 1. Nine red flags for autism

1. Lack of appropriate eye gaze2. Lack of warm, joyful expressions with gaze3. Lack of sharing enjoyment or interest4. Lack of response to name5. Lack of coordination of gaze, facial expression,

gesture, and sound6. Lack of showing7. Unusual prosody8. Repetitive movements or posturing of the body9. Repetitive movements with objects

Note: Data adapted from Wetherby et al. ~2004!.

578 E. Courchesne et al.

onance imaging @fMRI# , positron emissiontomography @PET# , magnetic encephalogra-phy @MEG# , event-related potential @ERP# , andEEG! the brain bases of these behavioral def-icits can be elucidated. By utilizing the grow-ing basic and clinical literature of cognitiveneuroscience, evidence of cognitive and be-havioral dysfunctions in autism can point tocandidate brain regions of abnormality ~e.g.,see Mundy, 2003; Williams, Whiten, Sudden-dorf, & Perrett, 2001!.

Cognitive neuroscience studies of autismhave identified altered patterns of neurofunc-tional activity, specifically reduced activity inhigher order frontal, temporal, and cerebellarregions but normal to increased activity inlower order posterior visual regions ~for re-views, see Belmonte et al., 2004; Courchesne,Redcay, & Kennedy, 2004; Frith, 2003!. Thesehigher order regions are critical to the initia-tion, perception, and interpretation of socio-emotional and communicative functions, whichare strikingly impaired in autism, as well asother higher order cognitive, attention, andmemory functions that are also abnormal inthis disorder. Reduced frontal activation hasbeen reported in dorsal or medial frontal cor-tices in a theory of mind task ~Castelli, Frith,Happe, & Frith, 2002!, in response to sociallyfamiliar faces ~Pierce, Haist, Sedaghat, &Courchesne, 2004!, during a gender or emo-tion decision task ~Hubl, Bolte, Feineis–Matthews, Lanfermann, Federspiel, Strik,Poustka, & Dierks, 2003!, in working mem-ory tasks ~Luna, Minshew, Garver, Lazar, Thul-born, Eddy, & Sweeney, 2002!, in an embeddedfigures task ~Ring, Baron–Cohen, Wheel-wright, Williams, Brammer, Andrew, & Bull-more, 1999!, in an emotion Stroop task~Kennedy, Redcay, & Courchesne, 2004!, invisual spatial attention tasks ~Belmonte &Yurgelun–Todd, 2003!, and during sentencecomprehension ~Just, Cherkassky, Keller, &Minshew, 2004; Muller, Behen, Rothermel,Chugani, Muzik, Mangner, & Chugani, 1999;Muller, Chugani, Behen, Rothermel, Muzik,Chakraborty, & Chugani, 1998!. ERP studieshave consistently found reduced or absentphysiological responses from frontal cortexduring a variety of auditory and visual at-tention and orienting tasks ~Ciesielski,

Courchesne, & Elmasian, 1990; Courchesne,Kilman, Galambos, & Lincoln, 1984; Town-send, Courchesne, Covington, Westerfield,Harris, Lyden, Lowry, & Press, 1999!. Re-duced temporal lobe activity has been re-ported in higher order lateral temporal regionsduring processing of vocal sounds ~Gervais,Belin, Boddaert, Leboyer, Coez, Sfaello, Bar-thelemy, Brunelle, Samson, & Zilbovicius,2004!, speech sounds ~Boddaert, Belin, Cha-bane, Poline, Barthelemy, Mouren–Simeoni,Brunelle, Samson, & Zilbovicius, 2003; Bod-daert et al., 2004!, and faces ~Pierce, Müller,Ambrose, Allen, & Courchesne, 2001!, andduring a mentalizing task ~Castelli et al., 2002!.Further, ERP and MEG studies have identi-fied reduced or abnormal temporal lobe re-sponses to a variety of speech and nonspeechsounds ~Bruneau, Bonnet–Brilhault, Gomot,Adrien, & Barthelemy, 2003; Bruneau, Roux,Adrien, & Barthelemy, 1999; Ceponiene, Lep-isto, Shestakova, Vanhala, Alku, Naatanen, &Yaguchi, 2003; Dawson, Finley, Phillips, Galp-ert, & Lewy, 1988; Gage, Siegel, Callen, &Roberts, 2003! and during spatial tuning ofattention to selective auditory sound sources~Teder–Salejarvi, Pierce, Courchesne, & Hill-yard, 2005!. In contrast to these reductions ofactivity, recent fMRI studies report normalactivation in visual cortex in response to basicvisual stimuli ~Hadjikhani, Chabris, Joseph,Clark, McGrath, Aharon, Feczko, Tager–Flusberg, & Harris, 2004!. Furthermore, sev-eral of the above-mentioned fMRI studies thatshow reductions in frontal and0or temporalregions show increased activity in occipitalregions, including lateral extrastriate regions~Ring et al., 1999!, medial occipital cortex~Hubl et al., 2003!, and ventral occipital cor-tex ~Belmonte & Yurgelun–Todd, 2003!. Thesefindings suggest that posterior, lower orderregions may be relatively spared while ante-rior, higher order regions may be moreseverely affected. Furthermore, functional con-nectivity studies have revealed reduced func-tional connectivity between lower order andhigher order brain regions ~e.g., occipital tofrontal, Castelli et al., 2002; superior temporalto inferior frontal, Just et al., 2004; and pari-etal to frontal, Horwitz, Rumsey, Grady, &Rapoport, 1988!.

Autism at the beginning 579

Findings from the cerebellar cortex in au-tism appear to parallel those of the cerebralcortex in that a similar dissociation betweenhigher order and lower order processes andregions is seen. Although only a handful offunctional neuroimaging studies of the cerebel-lum in autism exist, these studies reveal re-duced cerebellar activity during higher ordertasks, including attention ~Allen & Courchesne,2003!, listening to and generating sentences~Muller et al., 1998, 1999!, and judgement offacial expression ~Critchley, Daly, Phillips,Brammer, Bullmore, Williams, Van Amels-voort, Robertson, David, & Murphy, 2000!with normal to increased activity during lowerorder motor tasks ~Allen & Courchesne, 2003;Allen, Muller, & Courchesne, 2004; Muller,Pierce, Ambrose, Allen, & Courchesne, 2001!.The cerebellum has reciprocal projections toboth prefrontal ~Middleton & Strick, 2001!and temporal ~Schmahmann & Pandya, 1991!lobes, and thus, not surprisingly, reductions ofactivity in the cerebellum during higher ordertasks are often seen with concurrent reduc-tions in prefrontal or temporal lobe activity~e.g., Muller et al., 1999!. These functionalfindings suggest a system of neural dysfunc-tion including higher order frontal, temporal,and cerebellar regions may underlie many ofthe social and speech abnormalities seen inolder children and adults with autism.

An important question is what structuraldevelopmental abnormalities lead to these out-come frontal, temporal, and cerebellar neuro-functional abnormalities. It is argued here thatuseful models and knowledge of the neuraldevelopmental biology that precedes and pro-duces the initial autistic behavioral dysfunc-tions are necessary on three fronts: they canprovide explanations for outcome neuro-behavioral deficits that have been carefullydocumented by behavioral and neuroimagingresearch. They can provide information thatconstrains and directs candidate causes thatcan plausibly lead to these more specific earlyneural defects. Last, they can provide specificneurobiological processes and defects for tar-geting in animal models; autism is a disorderof higher order social, emotional, speech, lan-guage, and cognitive functions that are notconvincingly modelable in rodents, which are

the animals of choice for genetic and environ-mental factors research on autism.

Although much data exist to document thebehavioral and cognitive impairments in au-tism, comparatively little is known about theunderlying microstructural neural defects. Ex-istent postmortem neuropathology studies thathave relied on visual qualitative inspection inautism have shown abnormalities in micro-anatomy within frontal, temporal ~Bailey, Luth-ert, Dean, Harding, Janota, Montgomery,Rutter, & Lantos, 1998!, limbic ~Bauman &Kemper, 2005; Kemper & Bauman, 1998!, andcerebellar regions ~Bailey et al., 1998; Kemper& Bauman, 1998; Ritvo, Freeman, Scheibel,Duong, Robinson, Guthrie, & Ritvo, 1986!.However, these qualitative studies have beenfew in number, and none of these abnormali-ties have been rigorously quantified. Further,the nature of observed abnormalities revealedis diverse, from increased cell packing densityto laminar abnormalities to ectopic neurons inwhite matter ~Bailey et al., 1998!. Indeed, nu-merous microstructural abnormalities likely un-derlie the diverse behavioral phenotype ofautism. Systematic and quantitative neuropath-ological investigations require such an invest-ment of time and effort that an uninformedsearch for potentially subtle, but important,neuropathological differences would be inef-ficient and impractical. Thus, knowledge ofcandidate brain regions of abnormality in au-tism, as gained through cognitive neurosci-ence, allows for an efficient hypothesis-drivenanalysis of neurobiological abnormalities inautism. Further, knowledge of regional neuro-functional abnormalities provides an essentialframework for interpreting the potential func-tional significance of postmortem neuronal andmolecular findings. For example, the neuro-imaging literature would predict neuronal de-fects that are more apparent in prefrontal andhigher order temporal regions but less appar-ent in occipital cortex. In sum, to understandthe emergence of autism, a multilevel analysisapproach is useful wherein behavior informscognitive neuroscience, which informs basicneurobiology, and vice versa.

In the following paper, we will focus onfindings from macroscopic ~e.g., MRI! andmicroscopic structural investigations of au-


tism, highlighting findings from very youngchildren with autism. Further, we will attemptto relate this new microstructural evidence withfindings of macroscropic structural and func-tional abnormalities in autism.

White and Gray Matter GrowthPathology in the First Years of Lifein Autism: MRI Evidence

The first years of life are uniquelyimportant for brain development

Figure 1 is an example of the sparse neuralconnections in the frontal cortex of a new-born. The circuitry necessary for complex in-formation processing and complex behaviordoes not yet exist. Instead, it will be created inthe first postnatal years by neuronal differ-entiation and growth, dendritic and axonalgrowth, axonal myelination, and a tremen-dous increase in synapse numbers. Higher or-der social, emotional, cognitive, attention,speech, and language functions are all medi-

ated by slowly maturing regions such as fron-tal cortex, while basic sensory and perceptualfunctions are mediated by relatively more rap-idly and earlier maturing systems.

In the frontal cortex at birth, the length ofthe pyramidal cell dendritic arbors in layer 3is only 3% of full mature size, and by 2 yearsof age it is only 48% ~see example in Fig-ure 1!; many additional years will pass beforefull size is reached. By comparison, in pri-mary visual cortex pyramidal dendritic arborsare already 33% of full size at birth, and by 2years they have reached mature size ~table 1.3in Huttenlocher, 2002!. Myelination followsthe same functional hierarchical pattern, withbasic-level systems developing earlier thanhigher order association systems ~Kinney,Brody, Kloman, & Gilles, 1988!. Comparedto more posterior cortices, frontal cortex un-dergoes synapse formation later, and for a lon-ger period of time, and develops far largerpyramidal neurons with far more synapses~about 100,000 vs. about 20,000! and far largerdendritic ~e.g., in layer 3, 6836 vs. 2900 mm!

Figure 1. Golgi-stained sections showing the growth of pyramidal neuron soma and dendrites in themiddle frontal gyrus. The normal newborn has sparse neural circuitry; then, with increasing age, there isa tremendous increase in the complexity of dendritic arborizations. In this frontal cortical area, thedendrite arbors for layer 3 pyramidal neurons, which are only 3% of mature size in the newborn, are stillonly about 50% by 2 years of age and do not reach 100% until the end of childhood ~see text!. From TheHuman Brain ~3rd ed!, by J. Nolte, 1993, St. Louis, MO: Mosby Year Book0Elsevier. Copyright 1993by Mosby Year Book0Elsevier. Reprinted with permission from Elsevier.


and axonal arbors and axonal projections ~Hut-tenlocher, 2002!.

It has long been recognized that this uniqueperiod of neural differentiation and circuit for-mation is also a time when the brain is partic-ularly vulnerable to abnormal events thatdisrupt the formation of cortical connectivity,producing aberrant circuits and functions andbehavioral deficits ~Dobbing, 1981; Hutten-locher, 2002; Kinney et al., 1988!. More slowlymaturing brain regions such as frontal cortexhave a longer window of vulnerability thanmore rapidly maturing ones, such as the oc-cipital cortex.

Findings in young autistic children

Recent evidence indicates that autism may in-volve brain growth pathology during this verywindow of vulnerability; it is early, brief, andage delimited ~for reviews, see Courchesne,2004; Courchesne et al., 2004; Courchesne &Pierce, 2005a; Dementieva, Vance, Donnelly,Elston, Wolpert, Ravan, DeLong, Abramson,Wright, & Cuccaro, 2005!. At birth, head cir-cumference in autism, and therefore brain size~Bartholomeusz, Courchesne, & Karns, 2002!,were found to be either equivalent to the nor-mal average or slightly smaller than normal~Courchesne, Carper, & Akshoomoff, 2003;Courchesne et al., 2001; Dementieva et al.,2005; Gillberg & de Souza, 2002; Lainhart,Piven, Wzorek, Landa, Santangelo, Coon, &Folstein, 1997!.1 During the first year of lifethe autistic brain grows at an abnormally ac-celerated rate ~Dementieva et al., 2005; Mann& Walker, 2003!, such that by the end of thiscritical period at 2–3 years of age, quantita-tive MRI studies show it to be about 10%larger than normal ~Courchesne et al., 2001;Piven, 2004; Sparks et al., 2002!.

Both cerebral and cerebellar white mattervolumes in autistic toddlers have been found

to be abnormally enlarged ~18 and 39%, re-spectively!; cerebral gray matter is enlargedby 11% ~Courchesne et al., 2001!. Most im-portantly, there were striking regional differ-ences in this overgrowth pathology: in 2- to4-year-old autistic toddlers, frontal cortex andfrontal white matter volumes were the mostabnormally enlarged, but the occipital lobesdid not differ significantly from normal ~Car-per, Moses, Tigue, & Courchesne, 2002!.Within the frontal cortex, dorsolateral and me-sial prefrontal cortices were most abnormal,but the precentral gyrus, like the occipital lobe,was not significantly different from normal~Carper & Courchesne, 2005!. Temporal graymatter and parietal white matter were also en-larged but not to the magnitude of frontal grayand white volumes ~Carper et al., 2002!; inthe limbic system, the amygdala was also en-larged by 4 years of age ~Sparks et al., 2002!.A PET study of 3- to 4-year-old autistic chil-dren reported hypoperfusion in frontal cortexin autism which was interpreted as evidenceof delayed frontal maturation ~Zilbovicius,Garreau, Samson, Remy, Barthelemy, Syrota,& Lelord, 1995!.

An equally striking and important secondphase of growth pathology in autism followson the heels of the early overgrowth: abnor-mally slow or arrested growth. A recent meta-analysis of 12 MRI brain volume studiesshowed an autism brain size difference fromnormal of about 11% between 1 and 3 yearsof age, which declined through childhood~when the normal, but not autistic, brain con-tinues to grow!; by adolescence the autismbrain size difference from normal is only about1–2% ~Redcay & Courchesne, 2005!. Be-tween 2– 4 and 6–8 years of age, frontal andtemporal cortical gray matter increase by 20and 17% in normal children but change byonly 1 and �1%, respectively, in autism~Carper et al., 2002!. The dorsolateral sub-region of frontal cortex increases by 27% from2–5 years to 5–9 years of age in normals, butby only 7% in autistic children ~Carper &Courchesne, 2005!. White matter growth islikewise retarded; for example, between 2 and4 years and 7 and 11 years of age, frontalwhite matter volume increases by 45% in nor-mal children, but by only 13% in autistic chil-

1. There is excellent agreement among reports of birthhead circumference in autism: it was 34.7 cm in Gill-berg and de Souza ~2002!, 34.17 cm in Lainhart et al.~1997!, and 34.65 cm in Courchesne et al. ~2003!.Normal average birth head circumference is between34.5 and 35.9 cm, depending on the clinical normsused for reference ~Bartholomeusz et al., 2002!.


dren ~Carper et al., 2002!, and between 2 and3 years of age and adolescence, cerebellar whitematter volume increases by 50% in normalchildren, but by only 7% in autistic children~Courchesne et al., 2001!. Thus, regions thatshow the greatest early overgrowth in autism,frontal lobe and cerebellar white matter, alsoshow sharply reduced or arrested growththereafter.

Last, in contrast to the cerebellar whitematter changes just described, during earlydevelopment in autism there appears to be areduction in the size of one or another regionof the cerebellar vermis, which is largely graymatter ~review in Courchesne, 2004!. In oneMRI study with over 200 autistic and controlsubjects ~making this the largest autism MRIstudy ever done!, Hashimoto, Tayama, Mura-kawa, Yoshimoto, Miyazaki, Harada, andKuroda ~1995, p. 229! reported underdevelop-ment of the cerebellar vermis in autistic indi-viduals ranging from infants to adolescents.In another MRI study of 3- to 9-year-old chil-dren, reduction in the size of cerebellar ver-mis lobules VI–VII was found to be specificto autistic children compared with normal, frag-ile X, fragile X with autism, Down syndrome,and Down syndrome with autism children~Kaufmann, Cooper, Mostofsky, Capone,Kates, Newschaffer, Bukelis, Stump, Jann, &Lanham, 2003!. In a third recent MRI study ofautistic children, the posterior portion of thecerebellar vermis was found to be signifi-cantly reduced in size ~Levitt, Blanton,Capetillo–Cunliffe, Guthrie, Toga, & Mc-Cracken, 1999!. These findings are consistentwith some previous developmental MRI stud-ies reporting hypoplasia of the one or anotherregion of the cerebellar vermis ~e.g., Ciesiel-ski, Harris, Hart, & Pabst, 1997; Courchesne,Yeung–Courchesne, Press, Hesselink, & Jerni-gan, 1988!.

Findings in older autistic childrenand adults

Recent structural imaging results on older au-tistic children are compatible with this evi-dence of gray and white matter abnormality inautistic toddlers. For example, Herbert et al.

~2003! reported greater white matter volumein 7- to 11-year-old autistic children and in asecond study found that this deviation fromnormal was greatest in frontal white matterunderlying cortex ~so-called “radiate whitematter”! and least in occipital lobe ~Herbert,Ziegler, Makris, Filipek, Kemper, Normandin,Sanders, Kennedy, & Caviness, 2004!. Har-dan, Jou, Keshavan, Varma, and Minshew~2004! found abnormally increased gyrifica-tion in frontal cortex in an autistic sampleranging in age from 8 to 50 years. Levitt,Blanton, Smalley, Thompson, Guthrie, Mc-Cracken, Sadoun, Heinichen, and Toga ~2003!reported that some frontal and temporal sulciare abnormally shifted superiorly and posteri-orly in autistic children. Altered serotoninsynthesis in the cerebello–thalamo–frontalpathway has been described in autistic chil-dren ~Chugani, Muzik, Rothermel, Behen,Chakraborty, Mangner, da Silva, & Chugani,1997! and a theory has been proposed linkingserotonin abnormality during prenatal devel-opment to minicolumn and other neural abnor-malities in autism ~Chugani, 2000!.

Microstructural Defects in Autism:Prominent Cerebral andCerebellar Abnormalities

According to in vivo MRIs of autistic toddlersand young children, the frontal lobe, whichwould be predicted to be the most vulnerableduring postnatal neuronal growth, differentia-tion, and circuit formation, is indeed the struc-ture with the greatest early growth pathologyand later functional pathology. Conversely, oc-cipital lobes would be predicted to be the leastvulnerable, and in fact, they show a nonsignif-icant difference from normal both structurallyand functionally. Temporal and parietal corti-ces fall between these two extremes. This raisesthe hypothesis proposed by Courchesne andPierce ~2005a, 2005b! that fine quantitativemapping of neural microstructure and whitematter in autism will show that the greatestpathology occurs in cortical association re-gions that have the latest and most protractedneuronal and functional developmental timetables.


Similarly, the cerebellum is vulnerable dur-ing the first 2 years of postnatal life becausethe genesis of cerebellar cells continues wellinto the first and perhaps second year of life.It is unique in this regard because neurogen-esis during brain development is prenatal forall other major brain structures; whether thisfinal phase of cytogenesis in the cerebellumgenerates neurons, glia or both is uncertain.This vulnerable structure also develops abnor-mally in autism in the first years of life accord-ing to the in vivo MRI data reviewed above,and of all MRI abnormalities, the most strik-ing and pronounced is overgrowth of cerebel-lar white matter ~Courchesne et al., 2001!.

Neuron numbers: Excess in cerebrum,reduction in cerebellum

In collaborative and as yet unpublished re-search projects, neuron numbers in the cere-brum, cerebellum, and subcortical structuresin a set of autistic and control postmortemcases aged 4–67 years were counted ~Schmitz,2004; Wegiel, 2004!. These were among thefirst studies to utilize modern stereological pro-cedures in a study of the autistic brain. Thereappeared to be a tendency toward an excessnumber of cerebral cortical neurons was foundin autistic cases as compared to controls~Schmitz, 2004!. Interestingly, the magnitudeof this excess showed an age-related declineacross child to adult autism cases. It is not yetknown whether there are regional differencesin the excess neuron numbers or in the rateand magnitude of the age-related decline innumbers.

In contrast, decreased neuron numbers werereported for the cerebellum and basal ganglia~Wegiel, 2004!. Cerebellar Purkinje neuronnumbers were decreased by about 30% andthe volume of the cerebellum was reduced byabout 19%. This loss of Purkinje neurons wasseen at all ages in the autism sample from 4-to 67-year-olds. Large decreases in neuronnumbers were also reported for the nucleusaccumbens and basal ganglia, but neuronnumbers in the hippocampus were remarkablysimilar in autistic and control cases. This quan-titative and stereological study adds to a largenumber of observational reports that cerebel-

lar Purkinje neuron numbers are abnormallyreduced in autism ~see reviews in Courchesne,1997, 2004!. Cerebellar Purkinje neuron lossmay be the single most commonly reportedneuronal abnormality in the autism literatureto date.

Implications of increased cortical neuronsand decreased Purkinje neurons

Many fundamental questions remain beforethe implications of an increase in cortical neu-rons and axons can be fully interpreted. Arethere regional increases in neurons and axonsthat parallel regional increases in MRI grayand white matter volume? The only existentstereological count of cerebral neurons wasglobal, not regional ~Schmitz, 2004!. In ourpilot study, we reported what appeared to bean increased number of neurons in layer 3throughout frontal cortical regions but not inprimary visual cortex. Although this fits thereports of regional MRI volumetric differ-ences, it was not a stereologically conductedcount. It will be valuable for future studies toperform regional stereological neuron counts.Because cerebral neurons in different layersare generated at different but somewhat over-lapping developmental times, it will also beimportant to determine whether or not the in-creases are layer specific. There have beenhypotheses that autism involves deficient cor-tical inhibitory control, and so it will be inter-esting to know if there is an imbalance in thenumbers of excitatory pyramidal neurons andinhibitory interneurons such as chandelier cells.Because roughly 75–80% of cortical neuronsare pyramidal cells ~Jones, 1984!, an aberrantincrease in their numbers could have a signif-icant overall effect on both total neuron countsand gray matter volume even if the number ofinhibitory neurons remained constant or evendecreased slightly.

An excess number of cerebral neurons couldbe because of any one of several possibilities:a failure to correctly regulate the number ofneurons produced during the neurogenesisstage of prenatal development in autism, adeficit or delay in apoptosis so that too manysurvive into postnatal life, or a compensatoryneural genesis during perinatal or postnatal


life that is triggered by adverse events such asthose that ignite the neuroinflammatory reac-tion reported by Vargas, Nascimbene, Krish-nan, Zimmerman, and Pardo ~2005!.

There is no experimental data as yet thatcould speak to the abnormal neurogenesis pos-sibility. However, if the excess resulted froman early prenatal failure to regulate genesis,then it might be expected that this early ex-cess would cause the brain to appear to beenlarged at birth as well as after birth, but infact, in the majority of cases, head circumfer-ence at birth in autism is normal to slightlysmaller than normal. On the other hand, if theexcess in neuron numbers is regionally lim-ited to only late maturing frontal and otherhigher order association cortices, then per-haps at birth its presence might not be appar-ent and only becomes so across the first monthsand years of postnatal life.

The second possibility involving abnormaldeficits or delays in apoptosis is compatible withobservations of abnormally increased levels ofmolecules that should promote apoptosis mol-ecules in frontal and cerebellar cortices in adultautistic postmortem cases ~Araghi–Niknam &Fatemi, 2003! and of an age-related decline inthe magnitude of the excess numbers from child-hood to adult years in autism ~Schmitz, 2004!.

The third possibility involving an aberrantlate compensatory genesis could help explainwhy the brain at birth in autism is not abnor-mally enlarged in the great majority of cases.It is now appreciated that such a “compensa-tory” late genesis of neural and glial cells canoccur in response to some adverse conditions~Vaccarino & Ment, 2004!.

Abnormally slowed or arrested growth thatfollows the brief period of overgrowth in theautistic toddler could be because of a late on-set of apoptosis as just discussed above, be-cause of the conclusion of the period of latecompensatory neural and glial genesis alsojust discussed, or because the excess of neu-rons and connections produces dysfunctionalneural activity that causes elimination or re-tarded growth of ineffective neurons, axons,and dendritic and axonal arbors. Certainly thereare a number of clear microstructural signs ofarrest of growth, loss, and insufficient devel-opment in the older child or adult with autism,

each of which could underlie the arrested orslowed growth of gray and white matter thathas been observed in MRI studies of older au-tistic children and adults. Already mentionedwere abnormal age-related declines in cerebralneuron numbers, the presence of abnormal lev-els of molecules that might promote apoptosis,and abnormally narrow minicolumns. Small cellsize has been noted in the cingulate cortex~Kemper & Bauman, 1998! and in other frontalregions ~Buxhoeveden, Semendeferi, Schen-kar, Switzer, & Courchesne, 2005! in older au-tistic postmortem cases.

Increased numbers of cerebral neuronscould underlie the volume increases in cere-bral gray matter that have been reported inMRI studies. In addition, an excess of neuronslikely means an excess of axons, and this couldunderlie volume increases in white matter thathave also been reported in MRI studies ofautism. These excesses mean cortical connec-tivity must be abnormal. Some have specu-lated that autism involves cerebral anatomicaloverconnectivity ~Casanova, 2004!, while oth-ers have argued in favor of cerebral functionalunderconnectivity ~Belmonte & Yurgelun–Todd, 2003; Horwitz et al., 1988; Just et al.,2004!. It has also been suggested that the keyis an abnormally increased ratio of excitationto inhibition ~Rubenstein & Merzenich, 2003!.Here it is proposed that each of these defectsunderlies autism: there is an abnormal ratio ofexcitation to inhibition, there is anatomicaloverconnectivity but primarily within frontalcortical regions, and there is functional under-connectivity but principally in long-distancefrontoposterior reciprocal pathways. More-over, it is proposed that connectivity patternsand defects change with development.

Minicolumns are abnormally narrowin frontal and temporal cortexbut not occipital cortex

The minicolumn is a fundamental unit of in-formation processing. Figure 2 shows sche-matics of a minicolumn. The minicolumn is aroughly columnar vertical assembly of pyra-midal neurons and interneurons, their inter-connections, and input and output axons thatextend from layer 6 up to the cortical surface.


586

Each is 30– 60 mm in diameter with perhaps80–100 pyramidal neurons per minicolumn. Inhumans, minicolumns in frontal association cor-tex are nearly twice the diameter and severaltimes the volume of those in primary sensorycortices such as the primary visual cortex ~Bux-hoeveden & Casanova, 2002!. These 80–100pyramidal neurons are thought to have origi-nated from a common precursor cell duringneurogenesis in the second and third trimes-ters. Thus, the number of minicolumns in ce-rebral cortex reflects the number of precursorcells that gave rise to each minicolumn. Ex-pansion of cerebral cortex and increase in itsprocessing power across evolution up to hu-mans is due to an increase in the number ofprenatally generated minicolumns. The verti-cal array of pyramidal cells is surrounded byneuropil space; in humans, this neuropil spacethat contains synapses and dendrites has ex-panded greatly, allowing for more complex mi-crocircuitry and refined and powerful neuralcomputation within each minicolumn ~Buxho-eveden & Casanova, 2002!. These vertical as-semblies of interconnected pyramidal neuronshave large inhibitory interneurons ~e.g., chan-delier cells! that modulate the synchronous out-put of clusters of minicolumns. Synchronousprocessing and signaling locally in minicol-umns provide more powerful coherence of out-put to other cortical regions and cortico–corticalcoupling.The function of minicolumns is to pro-vide fine tuning of information processing andlearning within a cortical column.

The first quantitative study of minicolumnsize in autism examined layer 3 and foundminicolumns to be abnormally narrow in onefrontal and two temporal cortical areas ~BA 9,

21, and 22; Casanova, Buxhoeveden, Switala,& Roy, 2002!. In a pilot study, we aimed todetermine which frontal regions have this mini-column abnormality, and whether this abnor-mality is present at the time of clinical onsetof autism. We measured minicolumn size inlayer 3 throughout dorsal, mesial, and orbitalfrontal cortex in a postmortem 3-year-oldautistic case; minicolumn size in primary vi-sual cortex was measured as a contrast site~Buxhoeveden et al., 2005!. To test whetherthere is arrest of minicolumn growth in frontalcortex in autism, we compared results fromthis 3-year-old to those from a postmortem41-year-old adult with autism. We measuredbetween 1,600 and 2,000 individual minicol-umns per case, and we also measured con-trols. Minicolumn and neuropil sizes weresignificantly reduced throughout dorsal, me-sial, and orbital frontal cortices in the 41-year-old adult with autism, being almost half thevolume of normal adults. In addition, minicol-umns and their surrounding neuropil space inthe 3-year-old autistic case were nearly thesame size as this 41-year-old autistic adult,which suggests the hypothesis that in autismminicolumn growth is arrested sometime dur-ing early childhood. In contrast to findings infrontal cortices, in both the 3-year-old and41-year-old autistic cases, minicolumn size wasnormal in primary visual cortex. In addition,in the frontal cortex in these autistic cases,there appeared to be an excess number of neu-rons in each minicolumn, but this was notstereologically confirmed. The reduction infrontal minicolumn size in our small pilot studywas greater than that reported for minicol-umns in temporal cortex in the Casanova et al.

Figure 2. The minicolumn is a fundamental unit of information processing. ~a! It is a roughly columnarvertical assembly of pyramidal neurons and interneurons, their interconnections, and input and outputaxons that extend from layer 6 up to the cortical surface. Each is 30– 60 mm in diameter with perhaps80–100 pyramidal neurons per minicolumn. ~b! A schematic closeup of cortical layer 3 within aminicolumn. The shaded area represents the neuropil space that surrounds the vertical assembly ofneurons. This area contains mostly synapses, dendrites, and unmyelinated axons. The unshaded sectionis where the majority of the cell soma reside and it contains myelinated axons. From “MorphologicalDifferences Between Minicolumns in Human and Nonhuman Primate Cortex,” by D. P. Buxhoeveden,A. E. Switala, E. Roy, M. Litaker, and M. F. Casanova, 2001, American Journal of Physical Anthropol-ogy, 115. Copyright 2001 by American Journal of Physical Anthropology. Adapted with permissionfrom Wiley–Liss, a subsidiary of John Wiley & Sons, Inc.


~2002! study. Together, the Casanova studyand our pilot study reveal regional differencesin minicolumn size in autism, with maximummaldevelopment in the dorsal and orbital fron-tal cortex, somewhat less in the temporal cor-tex, and none detectable in the primary visualcortex.

Implications of underdevelopedminicolumns

An interesting question is whether the in-crease in cerebral neuron numbers indicatestoo many minicolumns or too many neuronsper minicolumn. The number of minicolumnsin cortex in autism has not been quantified.Certainly, a modest increase in the number ofotherwise normal minicolumns cannot ex-plain the emergence of autistic behavior in thefirst 2 years of life. On the contrary, one canimagine the opposite view: a modest increasein the number of otherwise normal minicol-umns could produce a modest improvement ininformation processing power.

In contrast, an excess number of neuronsper minicolumn would likely disrupt the nor-mal intrinsic microcircuitry and physiologicalfunctioning within each minicolumn, espe-cially if there is an imbalance between thenumber of excitatory pyramidal cells per col-umn and the number of inhibitory cells. Be-cause normally more neurons are generatedthan are later retained, an excess of neuronsper minicolumn would be indicative of a fail-ure of the normal neuron selection and prun-ing process of apoptosis. The retention of anabnormal number of neurons and their connec-tions would alter minicolumn circuit designand function. Because minicolumn cell assem-blies are thought to be the most basic unit ofrefined information processing in cortex, theirdefect could be part of the explanation of in-formation processing deficiencies in autism.

A more profound disturbance of minicol-umn circuit design and function would resultfrom migration abnormality. Migration de-fects have been observed in several postmor-tem autism cases ~Bailey et al., 1998!, andKennedy, Semendeferi, and Courchesne ~un-published! have also seen rafts of cells in ab-errant locations and disoriented pyramidal

neurons in the frontal cortex. Migration abnor-mality could fracture or interrupt the funda-mental vertical integrative feature of theminicolumn organizational design, cause anexcess of neurons in some minicolumns butreduction in others, and0or produce an excessof neurons settling in certain layers withinminicolumns and a deficit in other layers. Suchmigration-based defects would lead to frac-tionated and incompletely or aberrantly formedminicolumn vertical circuitry, as well as animbalance between excitation and inhibitionwithin and between minicolumns. This couldexplain why minicolumns are “narrow” or “un-derdeveloped” and an excess of neuron num-bers could explain why cortical gray mattervolumes are larger than normal.

The presence of normal minicolumns inprimary sensory cortex may signal that theautistic brain retains the capacity for detailedand refined processing of lower level visualinformation, and in fact, normal visual activa-tion in autism has been found by recent fMRIstudies ~Hadjikhani et al., 2004!. Thus, corti-cal areas of impaired and spared minicolumnsparallel areas of impaired and spared functionin autism. To our knowledge, this is the firstevidence of such a regionally specific cere-bral cytoarchitectonic–cerebral function par-allel in autism. Major strides in understandingthe fundamental neural bases of autistic be-havior may ensue from combined investiga-tions of regional differences in minicolumns,MRI volumetric growth patterns, and neuro-functional activity.

The double defect of aberrant migrationand an excess of neurons due to failure ofnormal apoptosis would be devastating tominicolumn circuit formation and function.Although a simple increase in minicolumnnumbers does not by itself explain abnormalfunction and behavior in the first 2 years oflife in autism, dysfunctional minicolumn as-sembly could explain numerous findings ofdysfunctional frontal activity. The underdevel-opment of minicolumns in frontal cortex likelysignals the failure of the normal emergence ofa diversity of highly specialized vertical func-tional units that are necessary for the refinedprocessing of and learning about informationcritical to higher order functions. Indeed, ab-


normalities in a variety of higher order social~Castelli et al., 2002; Pierce et al., 2004!, emo-tional ~Kennedy et al., 2004!, and cognitivefunctions ~Belmonte & Yurgelun–Todd, 2003;Just et al., 2004; Luna, Minshew, Garver, Lazar,Thulborn, Eddy, & Sweeney, 2002! have beendemonstrated by neuroimaging in many stud-ies of older children and adults with autism.

Abnormal glia activation andneuroinflammation are presentin frontal lobes and cerebellum

Vargas et al. ~2005! found evidence of astro-glial and microglial activation and neuroin-flammation in both white and gray mattersamples taken from posterior cerebellar hemi-spheres, the middle frontal gyrus and anteriorcinguate gyrus in 5- to 44-year-old autisticpostmortem cases. Macrophage chemoattrac-tant protein ~MCP-1! and tumor growth factor-beta-1 ~TGF-b-1! derived from glia were themost prevalent cytokines. CSF taken from liv-ing autistic children also showed a markedincrease in MCP-1. In all three regions, therewas enlargement of astroglial cell bodies andtheir processes. Microglial activation waspresent in the cerebellum and cerebral cortexand its underlying white matter; the cerebel-lum had the most pronounced microglial acti-vation. In the cerebellum, glial activation wasassociated with degenerating Purkinje neu-rons, granule cells and axons, and in nearly allautistic brain examined, there was Purkinjeand granule cell loss. Degenerating Purkinjecells were strongly immunoreactive for TGF-b-1. In the anterior cingulate and middle fron-tal gyrus, microglial activation was prominentat the junction of cortex and underlying whitematter. There was no evidence of an adaptiveimmune reaction ~e.g., T-cell infiltration ordeposition of immunoglobulin! in the autisticbrains, but evidence of deposition of comple-ment membrane attack complexes was ob-served in the cerebellum, apparently associatedwith Purkinje cells. Interestingly, compared tothe cerebellum and middle frontal gyrus, theanterior cingulate gyrus had a more complexarray of increased proinflammatory and mod-ulatory cytokines.

Implications of neuroinflammationand glial activation

Glial activation in cerebellar and frontal whitematter reported by Vargas et al. ~2005! mightcontribute to the significant increases in cer-ebellar and frontal lobe white matter volumesdocumented by in vivo MRI in 2- to 4-year-old autistic children ~Carper et al., 2002;Courchesne et al., 2001!, as well as to theretarded growth in white matter after this earlyovergrowth period. The reported glial activa-tion involved enlargement of glial cell bodiesin the autism postmortem cases ~Vargas et al.,2005!; although in that study the number ofglial cells was not counted, it is possible thatin autism there might also be an increase inglial cell numbers as part of the neuroinflam-mation reaction. Glial activation in white mat-ter that increases glial cell size and numbercould be part of the explanation for the vol-ume increases seen in structural MRI of youngautistic children. It might also be speculatedthat when glial activation reaches a plateau orsteady state, it would no longer drive volumeincreases; if so, this might explain the secondphase of growth pathology in autism: abnor-mally slow or arrested growth.

In the anterior cingulate gyrus and middlefrontal gyrus in the Vargas et al. ~2005! study,microglial activation was prominent at the junc-tion of cortex and underlying white matter. Invivo diffusion tensor imaging also reports whitematter abnormality in these same two frontalzones in older autistic children ~Barnea–Goraly, Kwon, Menon, Eliez, Lotspeich, &Reiss, 2004! and an in vivo MRI study reportsgreater volume in autism in the white matterunderlying cortex, the zone of so-called radi-ate white matter tracts ~Herbert et al., 2004!.Because brain size is normal to smaller thannormal at birth in autism and the abnormalacceleration of brain size begins in the firstpostnatal months, if glial activation does playa role in the overgrowth then it must begin todo so sometime between late in the third tri-mester and the first postnatal months of life.

It is uncertain how the neuroinflammationdescribed by Vargas et al. ~2005! might berelated to the recent report of an increase inthe number of cerebral neurons ~Schmitz,


2004!, but there are several possibilities. Oneis that the neuroinflammatory activity signalscompensatory reactions that limit the processof apoptosis in cerebral cortex thus sparingtoo many neurons. Another is that the neuroin-flammatory activity signals a late onset, com-pensatory proliferation of glia and cererbralcortical neurons ~see review in Vaccarino &Ment, 2004!. Still another is that genetic fac-tors lead to overproduction of both neural andglial cells as in the example of the p27 geneknock-out model mentioned below.

In either event, the overproduction of neu-rons, and probably glial cells, may itself trig-ger a still later compensatory response; in theSchmitz ~2004! study there was an age-relateddecline in the amount of excess cerebral neu-ron numbers; another study ~Araghi–Niknam& Fatemi, 2003! found evidence in postmor-tem autism tissue of reduced anti-apoptoticand increased pro-apoptotic molecules in thecerebral cortex. The age-related decline in ex-cess cerebral neuron numbers is compatiblewith the in vivo MRI studies of the youngautistic brain reviewed earlier. Namely, by 2–4years of age, frontal and temporal gray mattervolumes are abnormally enlarged, but there-after they fail to increase in volume, whereasnormal cortex grows by about 20% in sizeacross childhood ~Carper et al., 2002;Courchesne et al., 2001!.

Glial cells play key roles in brain organiza-tion during development as well as in neuroin-flammatory reactions. They are involved inneural migration, minicolumn structural for-mation, and minicolumn function, as well asin apoptosis, including that of Purkinje cells~Marin–Teva, Dusart, Colin, Gervais, van Roo-ijen, & Mallat, 2004!. For example, they nor-mally provide crucial signaling to migratingneurons and growing axons, and disruption ofglial development and activity would disruptneural development and axon connectivity pat-terns. They are hypothesized to play an essen-tial role in minicolumn structural and functionaldevelopment ~Colombo & Reisin, 2004!. Dis-ruption of this role may prevent the emer-gence of functionally discrete minicolumns.Glial developmental abnormalities, which ap-parently occur in autism, may play a part inthe genesis of microstructural abnormalities

in autism including migration defects, mini-column abnormality, connectivity defects andabnormal neuron numbers. Further research isneeded to determine whether the glial and mo-lecular abnormalities described by Vargas et al.~2005! are fundamentally neuroinflammatoryreactions that begin prenatal or early postnatalor reflect aberrations in genetic mechanismsthat regulate the normal role of glia duringneural development and organization.

Excess glial production and0or activationhave the potential to produce any or all of thepreviously discussed microstructural find-ings, including frontal minicolumn abnormal-ities and increased neuron counts. Abnormalglial activation might also help explain somemacrostructural findings in autism such as cer-ebellar and frontal white matter overgrowthand gray matter reduction in the cerebellum.In addition to its potential explanatory capac-ity for micro- and macrostructural findings,the increase in glial activity also could under-lie theories of autism based on functional im-aging studies, such as local overconnectivityand long-distance underconnectivity, based onthe capacity of glia to alter synaptic strength~Ullian, Christopherson, & Barres, 2004!. Ex-cess glial activation in the cerebellum is alsoin accord with previously discussed findingsof abnormal cerebellar function ~Allen &Courchesne, 2003; Allen et al., 2004!.

Summary and Conclusions

Evidence now supports the hypothesis thatgrowth pathology throughout the first years ofpostnatal life prevents the developmental for-mation of neural circuitry in frontal, temporal,and cerebellar cortices that is essential forhigher order social, emotional, language,speech, and cognitive functions ~Courchesne& Pierce, 2005a, 2005b!. In the autistic infantor toddler, the reason these higher order func-tions do not appear when they should or ap-pear only in a nascent form and then regress,is that neural maldevelopment in these partic-ular cortices precedes and prevents these es-sential circuits from forming in the first place.The reason that it is not until the second andmore commonly the third year of life before itis realized that a toddler has autism, is be-


cause these frontal, temporal, and cerebellarcircuits normally have a late and protracteddevelopment and do not normally “come on-line” until the second and third years of life.Thus, in the first year of life, the infant withautism and the normal infant are not easilydistinguished from each other: that is, boththe autistic and the normal infant lack manyof these higher order functions ~e.g., seeTable 1; most of the red flags of autism arebehaviors that are also undeveloped in the nor-mal infant during the first month of life!.

At the macroscopic structural level, thisprocess of neural maldevelopment is signaledby an abnormally accelerated rate of braingrowth after birth that is not sustained beyondearly childhood. By 2 to 4 years of age, thestructures most abnormally enlarged are fron-tal and temporal gray and white matter, theamygdala, and cerebellar white matter. Struc-tural data from older children and adults withautism shows an abnormal shift in the relativelocation of several frontal and temporal sulciand increased folding of frontal cortex; whitematter underlying frontal and temporal corti-ces may be especially abnormally increased involume and may have abnormal diffusion pat-terns that are indicative of either abnormallyoriented axons or other white matter pathol-ogy. Neurofunctional abnormalities in frontal,temporal, and cerebellar cortices have beendemonstrated in children and adults with au-tism via a large number of fMRI, PET, andERP experiments ~Belmonte et al., 2004; Bel-monte & Yurgelun–Todd, 2003; Chugani et al.,1997; Dawson, Osterling, Rinaldi, Carver, &McPartland, 2001; Friedman, Shaw, Artru,Richards, Gardner, Dawson, Posse, & Dager,2003; Hughes, Russell, & Robbins, 1994; Lunaet al., 2002; McEvoy, Rogers, & Pennington,1993; Minshew et al., 1997; Pennington &Ozonoff, 1996; Pierce et al., 2004; Rumsey &Hamburger, 1990; Townsend et al., 2001; Zil-bovicius et al., 1995!.

At the microstructural level, both neuronaland glial abnormalities are present in cerebraland cerebellar structures. In the cerebrum, neu-ron numbers may be increased, but in the cer-ebellum they are significantly decreased.Although cerebral neuron numbers may be in-creased, in some frontal regions neurons may

be too small and too densely packed, signsindicating underdevelopment. Minicolumns infrontal and temporal association cortices ~butnot in primary visual cortex! are too narrow, afinding suggestive of deviant development ofthis fundamental unit of neural microcircuitry.In frontal and cerebellar gray and white mat-ter, astroglia and microglia are activated andmolecular signals of a neuroinflammatory re-action are present, both findings suggestive ofeither an on-going but delayed developmentalstage of apoptosis or an on-going innate in-flammatory reaction to some yet to be identi-fied trigger ~e.g., a chemical or pathogenexposure, a genetic defect, etc.!. Finally, ac-cording to visual inspection ~and not yet ver-ified by quantitative experiments!, in someautistic postmortem cases neural migration de-fects and abnormally oriented pyramidal cellsare both present in frontal and temporal corti-ces and migration defects are present in thecerebellum.

These microstructural defects are presentin autistic cases as young as 3 and 4 years,ages during which in vivo MRI studies findabnormal enlargement of these same cerebraland cerebellar structures. Thus, these defectsmight play a role in the genesis of enlarge-ment of those structures. Certainly an excessof cerebral neurons with its attendant excessof axons and activated glial cells ~with per-haps an increase in the number of glial cells!might each contribute to the gray and whitematter volume increases seen in the first yearsof life in autism. There is no evidence of anincrease in the number of minicolumns, al-though it has been speculated that there mustbe an excess of them because cerebral graymatter volumes are increased. However, thisis not the only explanation. Gray matter vol-umes might be increased because cortex isthicker due to an excess of neurons and pan-laminar glial activation. Minicolumns mightbe thin but taller than normal. Another possi-bility is that minicolumns are initially gener-ated in normal numbers but become structurallyand functionally fractionated due to migrationabnormalities ~perhaps because glial guid-ance signals are abnormal!, the failure of apop-tosis to establish a normal number of neuronswithin a column, or the pathological glial ac-


tivation, which might disturb the ability ofastroglia to perform its developmental role inminicolumn functional organization. Together,an excess of neurons and fractionated minicol-umns might give the misimpression of an in-crease in the number of columns that wereoriginally generated in early prenatal life.

Brain enlargement is absent at birth in atleast 94% or more of autistic infants ~reviewin Courchesne & Pierce, 2005a!. This meanseither that the environmental and0or geneticdefects and processes that generate acceler-ated growth during the first years of life beginin perinatal or early postnatal periods, or thatthey remain occult in some way until then.Several scenarios can be conceptualized thatwould produce the observed microstructuraland macrostructural outcomes. Adverse events~e.g., Glasson, Bower, Petterson, de Klerk,Chaney, & Hallmayer, 2004! during late pre-natal, perinatal, or early postnatal life mighttrigger a neuroinflammatory reaction in fron-tal cortex including glial activation and gen-esis of new neural cells. Alternately, the numberof cortical neurons generated might be nor-mal, but for some reason there is a delay orfailure in apoptosis in frontal and possiblyother association cortices. In that event, glialactivation reflects a delayed, deviant, or in-effective apoptotic process. Another possibil-ity is that perhaps the number of corticalneurons generated far exceeds normal, andthis triggers a prolonged, on-going “correc-tive” apoptotic process that abnormally ex-tends into postnatal life. In addition to scenariosin which an inflammation-inducing insult orproliferative error causes multiple further dis-ruptions, neuronal overproliferation and glialactivation could share a common genetic root.One example of a potential genetic base formany of the observed micro and macrostruc-tural changes is the p27 gene. Its loss causesdysregulation of cell proliferation cycles, re-sulting in a 250% increase in glia cell num-bers in the cerebellum and a 30% increase inhippocampal neurons ~Casaccia–Bonnefil,Tikoo, Kiyokawa, Friedrich, Chao, & Koff,1997!.

A common theme across these and otherplausible scenarios is that in autism at birth,there is an imbalance of neuron numbers with

an excess in the frontal cortex and perhapsother association cortical regions but not inbasic-level systems, and there is concurrentglial activation. These abnormalities couldset in motion a cascade of maldevelopmentin the frontal cortex via several pathways~Courchesne & Pierce, 2005b!. First, early de-velopment normally involves not only progres-sive growth processes, but also regressive,selection, and elimination processes ~Hutten-locher, 2002; Quartz & Sejnowski, 1997!.Thus, abnormal cellular processes that createthe overgrowth and then arrested growth dur-ing this vulnerable developmental time couldin turn accelerate, retard, or disrupt selectionand regressive processes that would normallybe occurring. Second, the activation of astro-glia might disrupt the ability of astroglia toplay their normal postnatal role in frontal mini-column functional development. Third, the ex-cess of frontal cortical neurons after the normalstage of apoptosis ~which is normally largelycompleted prenatal! might impede the refine-ment of within-minicolumn circuits, tip theexcitatory–inhibitory balance in minicolumnstowards excess excitation, and abnormally in-crease the target size for long-distance axonsfrom posterior lower level systems whichwould effectively dilute their impact on fron-tal neural functioning. Further, following thesimple principle that neurons that fire to-gether wire together, the abnormal excess offrontal neurons, in the absence of normal lo-cal inhibitory modulation, might be predictedto create local and very short distance eddiesof excitation that develop into excessively over-connected but dysfunctional local and short-distance circuits. Conversely, long-distancecortical–cortical connectivity would be de-creased because its development depends onspatiotemporally coherent bursts of activity.The net functional result is diminished impactof low-level information on frontal activityand diminished impact of frontal activity onposterior systems. In effect, then, frontal cor-tex is, relative to normal, “disconnected” fromother cortical and subcortical structures, andinstead frontal cortex mainly “talks with it-self” ~Courchesne & Pierce, 2005b!. The cen-tral function of frontal cortex, to integratediverse information from multiple systems and


provide directive and adaptive feedback, doesnot develop in autism.

In the beginning in autism, we hypothesizethat frontal functions fail to develop normallyin the first years of life and so the first signs ofautism are absent or deviant frontal-mediatedbehaviors. Underlying frontal dysfunctionmay be defective minicolumn microcir-cuitry, excessive but disorganized local andshort-distance connectivity, and deficient long-distance reciprocal cortical–cortical connec-tivity. Low-level, basic information processingmay be relatively spared but it may fail to beutilized in the serve of high-order context-based, goal-directed behavior. Thus, aberrantcortical long-distance and local circuitry wouldimpair the essential role of frontal cortex inintegrating information from diverse and dis-tant functional systems ~emotional, sensory,autonomic, memory, etc.! and in providingcontext-based and goal-directed feedback tolower level systems.

This hypothesis of neural information pro-cessing is corroborated by neurofunctionalfindings of reduced activity in higher orderfrontal and temporal regions but normal toincreased activity in lower order posterior re-gions, as well as reduced functional connec-tivity between regions as described above. Alack of input from higher order associationregions would lead to a cognitive processingstyle that is fragmented and incomplete. Rather,sensory cortices may receive input normallybut fail to send and receive inputs from higherorder association regions, due to deficient longdistance connectivity. This may account forthe behavioral findings of decreased contex-tual and social processing but enhanced pitchand visual–perceptual processing.

It is important that these findings of re-duced neural activity in higher order regions

and reduced functional connectivity are mostlyfrom older children and adults. Cognitiveneuroscience studies of the younger autisticchild are needed to determine if similar neuro-functional findings are present during the earlyperiod of deviant brain development. As seenfrom the anatomical evidence outlined in thispaper, there are macro- and microstructuraldifferences between young children with au-tism and adults with autism, including mea-sures of total brain size, gray and white mattervolumes, and neuron numbers. Thus, it is un-likely that neurofunctional findings from theolder autistic child or adult will be generaliz-able to the infant or toddler in whom autism isjust emerging. Unfortunately, the new field ofdevelopmental cognitive neuroscience has beenseverely slowed by the difficulty in adaptingcognitive neuroscience technologies used inadults to infants and children. The first docu-mentation of cognitive ERP components ininfants ~Courchesne, Ganz, & Norcia, 1981!came nearly 2 decades after they were firstdocumented in adults. The first fMRI study ofhealthy, typically developing infants was car-ried out in 2001 ~Anderson, Marois, Colson,Peterson, Duncan, Ehrenkranz, Schneider,Gore, & Ment, 2001!, following almost a de-cade with scores of fMRI studies of healthyadults. Clearly, there is an enormous gap inour understanding of the neural developmentof cognitive processes in humans. Such workis of vital importance to link early structuralfindings ~i.e., microscopic postmortem andmacroscopic MRI! with the emergence of theautistic cognitive and behavioral phenotype.

Significant gains in our understanding ofthe neuroscience of autism will likely be madeby embracing a combined neurobiological andcognitive neuroscience approach directed to-ward the early development of autism.

References

Allen, G., & Courchesne, E. ~2003!. Differential effectsof developmental cerebellar abnormality on cognitiveand motor functions in the cerebellum: An fMRI studyof autism. American Journal of Psychiatry, 160,262–273.

Allen, G., Muller, R. A., & Courchesne, E. ~2004!. Cer-ebellar function in autism: Functional magnetic reso-nance image activation during a simple motor task.Biological Psychiatry, 56, 269–278.

Anderson,A. W., Marois, R., Colson, E. R., Peterson, B. S.,Duncan, C. C., Ehrenkranz, R. A., Schneider, K. C.,Gore, J. C., & Ment, L. R. ~2001!. Neonatal auditoryactivation detected by functional magnetic resonanceimaging. Magnetic Resonance Imaging, 19, 1–5.

Araghi–Niknam, M., & Fatemi, S. H. ~2003!. Levels ofBcl-2 and P53 are altered in superior frontal and cer-ebellar cortices of autistic subjects. Celullar and Mo-lecular Neurobiology, 23, 945–952.


Bailey, A., Luthert, P., Dean, A., Harding, B., Janota, I.,Montgomery, M., Rutter, M., & Lantos, P. ~1998!. Aclinicopathological study of autism. Brain, 121,889–905.

Barnea–Goraly, N., Kwon, H., Menon, V., Eliez, S.,Lotspeich, L., & Reiss, A. L. ~2004!. White matterstructure in autism: Preliminary evidence from diffu-siontensorimaging.BiologicalPsychiatry,55,323–326.

Baron–Cohen, S., Leslie, A. M., & Frith, U. ~1985!. Doesthe autistic child have a “theory of mind”? Cognition,21, 37– 46.

Bartholomeusz, H. H., Courchesne, E., & Karns, C. ~2002!.Relationship between head circumference and brainvolume in healthy normal toddlers, children, and adults.Neuropediatrics, 33, 239–241.

Bauman, M. L., & Kemper, T. L. ~2005!. Neuroanatomicobservations of the brain in autism: A review andfuture directions. International Journal of Develop-mental Neuroscience, 23, 183–187.

Belmonte, M. K., Cook, E. H., Jr., Anderson, G. M.,Rubenstein, J. L., Greenough, W. T., Beckel–Mitchener,A., Courchesne, E., Boulanger, L. M., Powell, S. B.,Levitt, P. R., Perry, E. K., Jiang, Y. H., DeLorey,T. M., & Tierney, E. ~2004!. Autism as a disorder ofneural information processing: Directions for re-search and targets for therapy ~1!. Molecular Psychi-atry, 9, 646– 663.

Belmonte, M. K., & Yurgelun–Todd, D. A. ~2003!. Func-tional anatomy of impaired selective attention andcompensatory processing in autism. Cognitive BrainResearch, 17, 651– 664.

Boddaert, N., Belin, P., Chabane, N., Poline, J. B., Bar-thelemy, C., Mouren–Simeoni, M. C., Brunelle, F.,Samson, Y., & Zilbovicius, M. ~2003!. Perception ofcomplex sounds: Abnormal pattern of cortical activa-tion in autism. American Journal of Psychiatry, 160,2057–2060.

Boddaert, N., Chabane, N., Belin, P., Bourgeois, M., Royer,V., Barthelemy, C., Mouren–Simeoni, M. C., Phil-ippe, A., Brunelle, F., Samson, Y., & Zilbovicius, M.~2004!. Perception of complex sounds in autism: Ab-normal auditory cortical processing in children. Amer-ican Journal of Psychiatry, 161, 2117–2120.

Bruneau, N., Roux, S., Adrien, J. L., Barthelemy, C. ~1999!.Auditory associative cortex dysfunction in childrenwith autism: Evidence from late autitory evoked po-tentials ~N1 wave-T complex!. Clinical Neurophysi-ology, 110, 1927–1924.

Bruneau, N., Bonnet–Brilhault, F., Gomot, M., Adrien,J. L., & Barthelemy, C. ~2003!. Cortical auditory pro-cessing and communication in children with autism:Electrophysiological0behavioral relations. Inter-national Journal Psychophysiology, 51, 17–25.

Buxhoeveden, D. P., Semendeferi, K., Schenkar, N., Swit-zer, R., & Courchesne, E. ~2005!. Reduced mini-columns in frontal cortex in patients with autism.Unpublished manuscript.

Buxhoeveden, D. P., & Casanova, M. F. ~2002!. The mini-column hypothesis in neuroscience. Brain, 125~Pt 5!,935–951.

Buxhoeveden, D. P., Switala, A. E., Roy, E., Litaker, M.,& Casanova, M. F. ~2001!. Morphological differencesbetween minicolumns in human and nonhuman pri-mate cortex. American Journal of Physical Anthropol-ogy, 115, 361–371.

Carper, R. A., & Courchesne, E. ~2005!. Localized en-largement of the frontal cortex in early autism. Bio-logical Psychiatry, 57, 126–133.

Carper, R. A., Moses, P., Tigue, Z. D., & Courchesne, E.~2002!. Cerebral lobes in autism: Early hyperplasiaand abnormal age effects. NeuroImage, 16, 1038–1051.

Casaccia–Bonnefil, P., Tikoo, R., Kiyokawa, H., Friedrich,J., Chao, M., & Koff, A. ~1997!. Oligodendrocite pre-curser differentiation is perturbed in the absence ofthe cylin-dependent kinase inhibitor p27kip1. Genesand Development, 11, 2335–2346.

Casanova, M. F. ~2004!. White matter volume increaseand minicolumns in autism. Annals of Neurology, 56,453; author reply 454.

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!. Au-tism, Asperger syndrome and brain mechanisms forthe attribution of mental states to animated shapes.Brain, 125, 1839–1849.

Ceponiene, R., Lepisto, T., Shestakova, A., Vanhala, R.,Alku, P., Naatanen, R., & Yaguchi, K. ~2003!. Speech-sound-selective auditory impairment in children withautism: They can perceive but do not attend. Proceed-ings of the National Academy of Sciences USA, 100,5567–5572.

Chugani, D. ~2000!. Autism. In M. Ernst & J. Rumsey~Eds.!, The foundation and future of functional neuro-imaging in child psychiatry ~pp. 171–188!. New York:Cambridge University Press.

Chugani, D. C., Muzik, O., Rothermel, R. D., Behen,M. E., Chakraborty, P. K., Mangner, T. J., da Silva,E. A., & Chugani, H. T. ~1997!. Altered serotoninsynthesis in the dentato–thalamo–cortical pathway inautistic boys. Annals of Neurology, 14, 666– 669.

Ciesielski, K.T, Harris, R., Hart, B., & Pabst, H. ~1997!.Cerebellar hypoplasia and frontal lobe cognitive def-icits in disorders of early childhood. Neuropsycholo-gia, 35, 643– 655.

Ciesielski, K. T., Courchesne, E., & Elmasian, R. ~1990!.Effects of focused selective attention tasks on event-related potentials in autistic and normal individuals.Electroencephalography and Clinical Neurophysiol-ogy, 75, 207–220.

Colombo, J., & Reisin, H. ~2004!. Interlaminar astrogliaof the cerebral cortex: A marker of the primate brain.Brain Research, 1006, 126–131.

Courchesne, E. ~1997!. Brainstem, cerebellar and limbicneuroanatomical abnormalities in autism. CurrentOpinion in Neurobiology, 7, 269–278.

Courchesne, E. ~2004!. Brain development in autism: Earlyovergrowth followed by premature arrest of growth.Mental Retardation and Developmental DisabilitiesResearch Reviews, 10, 106–111.

Courchesne, E., Carper, R., & Akshoomoff, N. ~2003!.Evidence of brain overgrowth in the first year of lifein autism. Journal of the American Medical Associa-tion, 290, 337–344.

Courchesne, E., Ganz, L., & Norcia, A. M. ~1981!. Event-related brain potentials to human faces in infants. ChildDevelopment, 52~3!.

Courchesne, E., Karns, C., Davis, H. R., Ziccardi, R.,Carper, R., Tigue, Z., Pierce, K., Moses, P., Chisum,H. J., Lord, C., Lincoln, A. J., Pizzo, S., Schreibman,L., Haas, R. H., Akshoomoff, N., & Courchesne,R. Y. ~2001!. Unusual brain growth patterns in earlylife in patients with autistic disorder: An MRI study.Neurology, 57, 245–254.

Courchesne, E., Kilman, B. A., Galambos, R., & Lincoln,A. J. ~1984!. Autism: Processing of novel auditory


information assessed by event-related brain poten-tials. Electroencephalography and Clinical Neurophys-iology, 59, 238–248.

Courchesne, E., & Pierce, K. ~2005a!. Brain overgrowthin autism during a critical time in development: Im-plications for frontal pyramidal neuron and interneu-ron development and connectivity. InternationalJournal of Developmental Neuroscience, 23, 153–170.

Courchesne, E., & Pierce, K. ~2005b!. Why the frontalcortex in autism might be talking only to itself: Localoverconnectivity but long-distance disconnection. Cur-rent Opinion in Neurobiology, 15, 225–230.

Courchesne, E., Redcay, E., & Kennedy, D. P. ~2004!.The autistic brain: Birth through adulthood. CurrentOpinion in Neurology, 17, 489– 496.

Courchesne, E., Townsend, J., Akshoomoff, N. A., Saitoh,O., Yeung–Courchesne, R., Lincoln, A. J., James,H. E., Haas, R. H., Schreibman, L., & Lau, L. ~1994!.Impairment in shifting attention in autistic andcerebellar patients. Behavioral Neuroscience, 108,848–865.

Courchesne, E.,Yeung–Courchesne, R., Press, G.A., Hesse-link, J. R., & Jernigan, T. L. ~1988!. Hypoplasia ofcerebellar vermal lobules VI and VII in autism. NewEngland Journal of Medicine, 318, 1349–1354.

Critchley, H., Daly, E., Phillips, M., Brammer, M., Bull-more, E., Williams, S., Van Amelsvoort, T., Robert-son, D., David, A., & Murphy, D. ~2000!. Explicit andimplicit neural mechanisms for processing of socialinformation from facial expressions: A functional mag-netic resonance imaging study. Human Brain Map-ping, 9, 93–105.

Dawson, G., Finley, C., Phillips, S., Galpert, L., & Lewy,A. ~1988!. Reduced P3 amplitude of the event-relatedbrain potential: Its relationship to language ability inautism. Journal of Autism and Developmental Disor-ders, 18, 493–504.

Dawson, G., Osterling, J., Meltzoff, A. N., & Kuhl, P.~2000!. Case study of the development of an infantwith autism from birth to two years of age. Journal ofApplied Developmental Psychology, 21, 299–313.

Dawson, G., Osterling, J., Rinaldi, J., Carver, L., & Mc-Partland, J. ~2001!. Brief report: Recognition memoryand stimulus–reward associations: Indirect supportfor the role of ventromedial prefrontal dysfunction inautism. Journal of Autism and Developmental Dis-orders, 31, 337–341.

Dementieva, Y. A., Vance, D. D., Donnelly, S. L., Elston,L. A., Wolpert, C. M., Ravan, S. A., DeLong, G. R.,Abramson, R. K., Wright, H. H., & Cuccaro,M. L. ~2005!. Accelerated head growth in early devel-opment of individuals with autism. Pediatric Neurol-ogy, 32, 102–108.

Dobbing, J. ~1981!. The later development of the brainand its vulnerability. In J. A. Dobbing ~Ed.!, Scientificfoundations of pediatrics ~2nd ed., pp. 744–759!. Lon-don: William Heinemann Medical Books.

Friedman, S. D., Shaw, D. W., Artru, A. A., Richards,T. L., Gardner, J., Dawson, G., Posse, S., & Dager,S. R. ~2003!. Regional brain chemical alterations inyoung children with autism spectrum disorder. Neu-rology, 60, 100–107.

Frith, C. ~2003!. What do imaging studies tell us aboutthe neural basis of autism? Novartis Foundation Sym-posium, 251, 149–166; discussion 166–176, 281–197.

Frith, U., & Happe, F. ~1994!. Autism: Beyond “theory ofmind.” Cognition, 50, 115–132.

Gage, N. M., Siegel, B., Callen, M., & Roberts, T. P.

~2003!. Cortical sound processing in children withautism disorder: An MEG investigation. NeuroRe-port, 14, 2047–2051.

Gervais, H., Belin, P., Boddaert, N., Leboyer, M., Coez,A., Sfaello, I., Barthelemy, C., Brunelle, F., Samson,Y., & Zilbovicius, M. ~2004!. Abnormal cortical voiceprocessing in autism. Nature Neuroscience, 7, 801–802.

Gillberg, C., & de Souza, L. ~2002!. Head circumferencein autism, Asperger syndome, and ADHD: A compar-ative study. Developmental Medicine and Child Neu-rology, 44, 296–300.

Glasson, E. J., Bower, C., Petterson, B., de Klerk, N.,Chaney, G., & Hallmayer, J. F. ~2004!. Perinatal fac-tors and the development of autism: A populationstudy. Archives of General Psychiatry, 61, 618– 627.

Hadjikhani, N., Chabris, C., Joseph, R., Clark, J., McGrath,L., Aharon, I., Feczko, E., Tager–Flusberg, H., & Har-ris, G. ~2004!. Early visual cortex organization inautism: An fMRI study. NeuroReport, 15, 267–270.

Hardan, A. Y., Jou, R. J., Keshavan, M. S., Varma, R., &Minshew, N. J. ~2004!. Increased frontal cortical fold-ing in autism: A preliminary MRI study. PsychiatryResearch, 131, 263–268.

Hashimoto, T., Tayama, M., Murakawa, K., Yoshimoto,T., Miyazaki, M., Harada, M., & Kuroda, Y. ~1995!.Development of the brainstem and cerebellum in au-tistic patients. Journal of Autism and DevelopmentalDisorders, 25, 1–18.

Herbert, M. R., Ziegler, D. A., Deutsch, C. K., O’Brien,L. M., Lange, N., Bakardjiev, A., Hodgson, J., Adrien,K. T., Steele, S., Makris, N., Kennedy, D., Harris,G. J., & Caviness, V. S., Jr. ~2003!. Dissociations ofcerebral cortex, subcortical and cerebral white mattervolumes in autistic boys. Brain, 126, 1182–1192.

Herbert, M. R., Ziegler, D. A., Makris, N., Filipek, P. A.,Kemper, T. L., Normandin, J. J., Sanders, H. A., Ken-nedy, D. N., & Caviness, V. S., Jr. ~2004!. Localiza-tion of white matter volume increase in autism anddevelopmental language disorder. Annals of Neurol-ogy, 55, 530–540.

Hobson, R. P. ~1993!. The emotional origins of socialunderstanding. Philosophical Psychology, 6, 227–249.

Horwitz, B., Rumsey, J. M., Grady, C. L., & Rapoport,S. I. ~1988!. The cerebral metabolic landscape in au-tism. Intercorrelations of regional glucose utilization.Archives of Neurology, 45, 749–755.

Hubl, D., Bolte, S., Feineis–Matthews, S., Lanfermann,H., Federspiel, A., Strik, W., Poustka, F., & Dierks, T.~2003!. Functional imbalance of visual pathways in-dicates alternative face processing strategies in au-tism. Neurology, 61, 1232–1237.

Hughes, C., Russell, J., & Robbins, T. W. ~1994!. Evi-dence for executive dysfunction in autism. Neuropsy-chologia, 32, 477– 492.

Huttenlocher, P. ~2002!. Neural plasticity: The effects ofenvironment on the development of cerebral cortex.Cambridge, MA: Harvard University Press.

Jones, E. G. ~1984!. Laminar distribution of cortical ef-ferent cells. In A. Peters & E. G. Jones ~Eds.!, Cere-bral cortex: Cellular components of the cerebral cortex~pp. 521–553!. New York: Plenum Press.

Just, M. A., Cherkassky, V. L., Keller, T. A., & Minshew,N. J. ~2004!. Cortical activation and synchronizationduring sentence comprehension in high-functioningautism: Evidence of underconnectivity. Brain, 127,1811–1821.

Kaufmann, W. E., Cooper, K. L., Mostofsky, S. H., Ca-pone, G. T., Kates, W. R., Newschaffer, C. J., Bukelis,


I., Stump, M. H., Jann, A. E., & Lanham, D. C. ~2003!.Specificity of cerebellar vermian abnormalities in au-tism: A quantitative magnetic resonance imaging study.Journal of Child Neurology, 18, 463– 470.

Kemper, T., & Bauman, M. ~1998!. Neuropathology ofinfantile autism. Journal of Neuropathology and Ex-perimental Neurology, 57, 645– 652.

Kennedy, D. P., Redcay, E., & Courchesne, E. ~2004!.Cognition, emotion, and the resting state: An fMRIstudy of neurofunctional abnormalities in autism. Pa-per presented at the IMFAR, Sacramento, CA.

Kinney, H. C., Brody, B. A., Kloman, A. S., & Gilles,F. H. ~1988!. Sequence of central nervous systemmyelination in human infancy. II. Patterns of myeli-nation in autopsied infants. Journal of Neuropathol-ogy and Experimental Neurology, 47, 217–234.

Lainhart, J. E., Piven, J., Wzorek, M., Landa, R., Santan-gelo, S. L., Coon, H., & Folstein, S. E. ~1997!. Macro-cephaly in children and adults with autism. Journal ofthe American Academy of Child and Adolescent Psy-chiatry, 36, 282–290.

Levitt, J. G., Blanton, R., Capetillo–Cunliffe, L., Guthrie,D., Toga, A., & McCracken, J. T. ~1999!. Cerebellarvermis lobules VIII–X in autism. Progress in Neuro-psychopharmacology and Biological Psychiatry, 23,625– 633.

Levitt, J. G., Blanton, R. E., Smalley, S., Thompson, P. M.,Guthrie, D., McCracken, J. T., Sadoun, T., Heinichen,L., & Toga, A. W. ~2003!. Cortical sulcal maps inautism. Cerebral Cortex, 13, 728–735.

Luna, B., Minshew, N. J., Garver, K. E., Lazar, N. A.,Thulborn, K. R., Eddy, W. F., & Sweeney, J. A. ~2002!.Neocortical system abnormalities in autism: An fMRIstudy of spatial working memory. Neurology, 59,834–840.

Mann, T. A., & Walker, P. ~2003!. Autism and a deficit inbroadening the spread of visual attention. Journal ofChild Psychology and Psychiatry and Allied Disci-plines, 44, 274–284.

Marin–Teva, J. L., Dusart, I., Colin, C., Gervais, A., vanRooijen, N., & Mallat, M. ~2004!. Microglia promotethe death of developing Purkinje cells. Neuron, 41,535–547.

McEvoy, R. E., Rogers, S. J., & Pennington, B. F. ~1993!.Executive function and social communication deficitsin young autistic children. Journal of Child Psychol-ogy and Psychiatry andAllied Disciplines, 34, 563–578.

Middleton, F. A., & Strick, P. L. ~2001!. Cerebellar pro-jections to the prefrontal cortex of the primate. Jour-nal of Neuroscience, 21, 700–712.

Minshew, N. J., Goldstein, G., & Siegel, D. J. ~1997!.Neuropsychologic functioning in autism: Profile of acomplex information processing disorder. Journalof the International Neuropsychological Society, 3,303–316.

Muller, R. A., Behen, M. E., Rothermel, R. D., Chugani,D. C., Muzik, O., Mangner, T. J., & Chugani, H. T.~1999!. Brain mapping of language and auditory per-ception in high-functioning autistic adults: A PET study.Journal of Autism and Developmental Disorders, 29,19–31.

Muller, R. A., Chugani, D. C., Behen, M. E., Rothermel,R. D., Muzik, O., Chakraborty, P. K., & Chugani,H. T. ~1998!. Impairment of dentato–thalamo–corticalpathway in autistic men: Language activation datafrom positron emission tomography. Neuroscience Let-ters, 245, 1– 4.

Muller, R. A., Pierce, K., Ambrose, J. B., Allen, G., &

Courchesne, E. ~2001!. Atypical patterns of cerebralmotor activation in autism: A functional magnetic res-onance study. Biological Psychiatry, 49, 665– 676.

Mundy, P. ~1995!. Joint attention and social–emotionalapproach behavior in children with autism. Develop-ment and Psychopathology, 7, 63–82.

Mundy, P. ~2003!. Annotation: The neural basis of socialimpairments in autism: The role of the dorsal medial–frontal cortex and anterior cingulate system. Journalof Child Psychology and Psychiatry and Allied Dis-ciplines, 44, 793–809.

Nolte, J. ~1993!. The human brain ~3rd ed.!. St. Louis,MO: Mosby Year Book.

Pennington, B. F., & Ozonoff, S. ~1996!. Executive func-tions and developmental psychopathology. Journal ofChild Psychology and Psychiatry and Allied Dis-ciplines, 37, 51–87.

Pierce, K., Haist, F., Sedaghat, F., & Courchesne, E. ~2004!.The brain response to personally familiar faces inautism: Findings of fusiform activity and beyond.Brain, 127, 2703–2716.

Pierce, K., Müller, R.-A., Ambrose, J., Allen, G., &Courchesne, E. ~2001!. Face processing occurs out-side the fusiform “face area” in autism: Evidence fromfMRI. Brain, 124, 2059–2073.

Piven, J. ~2004, May 7!. Longitudinal MRI study of 18–35month olds with autism. Paper presented at the Inter-national Meeting for Autism Research, Sacramento,California.

Quartz, S. R., & Sejnowski, T. J. ~1997!. The neural basisof cognitive development: A constructivist manifesto.Behavioral Brain Science, 20, 537–556; discussion556–596.

Redcay, E., & Courchesne, E. ~2005!. When is the brainenlarged in autism? A meta analysis of all brain sizereports. Biological Psychiatry, 58, 1–9.

Ring, H. A., Baron–Cohen, S., Wheelwright, S., Wil-liams, S. C., Brammer, M., Andrew, C., & Bullmore,E. T. ~1999!. Cerebral correlates of preserved cog-nitive skills in autism: A functional MRI study ofembedded figures task performance. Brain, 122,1305–1315.

Ritvo, E. R., Freeman, B. J., Scheibel, A. B., Duong, T.,Robinson, H., Guthrie, D., & Ritvo, A. ~1986!. LowerPurkinje cell counts in the cerebella of four autisticsubjects: Initial findings of the UCLA–NSAC au-topsy research report. American Journal of Psychia-try, 143, 862–866.

Rogers, S. J., & Pennington, B. F. ~1991!. A theoreticalapproach to the deficits in infantile autism. Develop-ment and Psychopathology, 3, 137–162.

Rubenstein, J. L., & Merzenich, M. M. ~2003!. Modelof autism: Increased ratio of excitation0inhibition inkey neural systems. Genes Brain and Behavior, 2,255–267.

Rumsey, J. M., & Hamburger, S. D. ~1990!. Neuropsycho-logical divergence of high-level autism and severedyslexia. Journal of Autism and Developmental Dis-orders, 20, 155–168.

Schmahmann, J. D., & Pandya, D. N. ~1991!. Projectionsto the basis pontis from the superior temporal sulcusand superior temporal region in the rhesus monkey.Journal of Comparative Neurology, 308, 224–248.

Schmitz, C. ~2004, November 13!. New aspects in theneuropathology of the forebrain in autism. Paper pre-sented at the Integrating the Clinical and Basic Sci-ences of Autism: A Developmental Biology Workshop,Ft. Lauderdale, FL.


Sparks, B. F., Friedman, S. D., Shaw, D. W., Aylward, E.,Echelard, D., Artru, A. A., Maravilla, K. R., Giedd,J. N., Munson, J., Dawson, G., & Dager, S. R. ~2002!.Brain sructural abnormalities in young children withautism spectrum disorder. Neurology, 59, 184–192.

Teder–Salejarvi, W. A., Pierce, K., Courchesne, E., &Hillyard, S. A. ~2005!. Auditory spatial localizationand attention deficits in autistic adults. Cognitive BrainResearch, 23, 221–234.

Townsend, J., Courchesne, E., Covington, J., Westerfield,M., Harris, N. S., Lyden, P., Lowry, T. P., & Press,G. A. ~1999!. Spatial attention deficits in patients withacquired or developmental cerebellar abnormality.Journal of Neuroscience, 19, 5632–5643.

Townsend, J., Westerfield, M., Leaver, E., Makeig, S.,Jung, T.-P., Pierce, K., & Courchesne, E. ~2001!. Event-related brain response abnormalities in autism: Evi-dence for impaired cerebello–frontal spatial attentionnetworks. Cognitive Brain Research, 11, 127–145.

Ullian, E. M., Christopherson, K. S., & Barres, B. A.~2004!. Role for glia in synaptogenesis. Glia, 47,209–216.

Vaccarino, F. M., & Ment, L. R. ~2004!. Injury and repairin developing brain. Archives of Diseases in Child-hood. Fetal and Neonatal Edition, 89, F190–F192.

Vargas, D. L., Nascimbene, C., Krishnan, C., Zimmer-man, A. W., & Pardo, C. A. ~2005!. Neuroglial acti-vation and neuroinflammation in the brain of patientswith autism. Annals of Neurology, 57, 67–81.

Wegiel, J. ~2004, November 13, 2004!. Subcortical neuro-pathology. Paper presented at the Integrating the Clin-ical and Basic Sciences of Autism: A DevelopmentalBiology Workshop, Ft. Lauderdale, FL.

Wetherby, A. M., Woods, J., Allen, L., Cleary, J., Dickin-son, H., & Lord, C. ~2004!. Early indicators of autismspectrum disorders in the second year of life. Journalof Autism and Developmental Disorders, 34, 473–493.

Williams, J. H., Whiten, A., Suddendorf, T., & Perrett,D. I. ~2001!. Imitation, mirror neurons and autism.Neuroscience and Biobehavioral Reviews, 25, 287–295.

Zilbovicius, M., Garreau, B., Samson, Y., Remy, P., Bar-thelemy, C., Syrota, A., & Lelord, G. ~1995!. Delayedmaturation of the frontal cortex in childhood autism.American Journal of Psychiatry, 152, 248–252.


Date post:	18-Dec-2021
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Autism at the beginning: Microstructural and growth ...

Documents