In Press,

Cerebral Cortex

Underconnectivity

in autism

Functional and anatomical cortical underconnectivity in autism:Evidence from an fMRI study of an executive function task andcorpus callosum morphometry

Marcel Adam Just

, Vladimir L. Cherkassky

Timothy A. Keller

, Rajesh K. Kana

, Nancy J. Minshew

Center for Cognitive Brain ImagingCarnegie Mellon University, Pittsburgh, PA

Departments of Psychiatry and Neurology,University of Pittsburgh School of Medicine, Pittsburgh, PA

Abstract

: The brain activation of a group of high-functioning autistic participants was measuredusing fMRI during the performance of a Tower of London task, in comparison to a control groupmatched with respect to IQ, age, and gender. The two groups generally activated the same corticalareas to similar degrees. However, there were three indications of underconnectivity in the groupwith autism. First, the degree of synchronization (i.e. the functional connectivity, or the correlationof the time series of the activation) between the frontal and parietal areas of activation was lowerfor the autistic than the control participants. Second, relevant parts of the corpus callosum, throughwhich many of the bilaterally activated cortical areas communicate, were smaller in cross-sectionalarea in the autistic participants. Third, within the autism group but not within the control group, thesize of the genu of the corpus callosum was correlated with frontal-parietal functional connectivity.These findings suggest that the neural basis of altered cognition in autism entails a lower degree of integration of information across certain cortical areas resulting from reduced intra-corticalconnectivity. The results add support to a new theory of cortical underconnectivity in autism,which posits a deficit in integration of information at the neural and cognitive levels.

Keywords

: autism, functional connectivity, fMRI, corpus callosum, executive function

Abbreviations

: fMRI = functional Magnetic Resonance Imaging; L DLPFC = Left DorsolateralPrefrontal Cortex; ROI = Region of Interest; TOL = Tower of London.Newly emerging theories of neurologicalfunctioning in autism are highlighting interregionalfunctional and anatomical connectivity as a likely keyfeature of the pathophysiology. Several recentfunctional neuroimaging studies provide evidence of a lower degree of coordination among activated brainareas in autism. A recent study of sentencecomprehension (Just et al., 2004) found that the brainactivity was less synchronized across activated brainareas (i.e. there was reduced functional connectivity)in autism. Studies of social cognition (Castelli et al.,2002) and working memory (Luna et al., 2002) alsosuggest aberrant functional connectivity in the brainsof individuals with autism. The corticalunderconnectivity theory of autism (Just et al., 2004)provides an integrating framework for the newfindings, and also provides useful extensions to previoustheories of autism. Very briefly, underconnectivitytheory proposes that autism is a cognitive andneurobiological disorder associated withunderfunctioning of integrative circuitry, resulting in adeficit in integration of information at the neural andcognitive levels.In addition to functional imaging studies,anatomical studies also present evidence for abnormalconnectivity in autism. Courchesne et al. (2001) foundan abnormal developmental trajectory of white matter inautism, such that 2-3 year old boys with autism hadincreased cerebral and cerebellar white matter volume.Herbert et al. (2004) found localized white matterenlargement in the outer radiate compartment of thewhite matter in children with autism. Herbert et al.

In Press,

Cerebral Cortex

Underconnectivity

in autism

2suggested an ongoing postnatal process involvingwhite matter in autism that primarily affectsintrahemispheric and corticocortical connections. Adiffusion tensor imaging study found reducedfractional anisotropy (indicating a lower degree of coherence of directionality) in white matter adjacentto the ventromedial prefrontal cortices, anteriorcingulate gyri, temporoparietal junctions, and in thecorpus callosum (Barnea-Goraly et al., 2004). At amuch more fine-grained level, Casanova et al. (2002)found more numerous and abnormally narrowminicolumns in the frontal and temporal cortex inautism, creating an abundance of short connectivefibers relative to long ones, which may indicate adeficiency in long distance (inter-regional)connectivity. The converging findings of functionalconnectivity abnormalities and white matterabnormalities in autism in several studies suggest thatalterations in cortical connectivity and thecommunication among cortical regions may be partof the pervasive core processing deficits in autism(Herbert et al., 2004).One of the anatomical regions that has emergedas a target of autism research is the corpus callosum,which mediates the inter-hemispheric communicationamong cortical areas underpinning higher levelcognitive function. Several morphometric studiesreport abnormalities, especially reduction in size, invarious subregions of the corpus callosum in autism(e.g. Piven et al., 1997; Manes et al., 1999; Hardan etal., 2000; Vidal et al., 2003). Chung et al. (2004)found lower white matter density (an index for neuralconnectivity) in the genu, rostrum and splenium of the corpus callosum in individuals with autism, andsuggested that this reduction might result in impairedinter-hemispheric connectivity in frontal, temporaland occipital regions. The interhemispheric fibersfrom the inferotemporal and occipital lobes (posteriorareas) traverse the splenium (the posterior part of thecorpus callosum), whereas fibers from the frontallobes traverse the genu and rostrum (Pandya &Seltzer, 1986). Therefore it is possible thatabnormalities in the subregions of the corpuscallosum could disrupt the functional connectivityamong cortical regions in the two hemispheres.The particular task used for examining brainactivation in the current fMRI study, the Tower of London puzzle, is considered a test of executivefunction. Some of the strongest experimentalevidence for executive dysfunction in autism so farinvolves the Tower of London (TOL), a task requiring planning and goal-management ability.Several investigations using Tower tasks have foundsignificant impairments in autism relative to matchedcontrols (Minshew et al., 2002; Ozonoff et al., 1991;Hughes et al., 1994; Bennetto et al., 1996; Ozonoff &Jensen, 1999; Ozonoff & McEvoy, 1994). The executiveprocessing in the Tower of London task has been shownin normal individuals to evoke prominent activationbilaterally in prefrontal and parietal areas (Newman etal., 2003).Brain activation during tests of executive functionhas not been widely investigated in autism (Hill & Frith,2003). It is not known whether executive functiondeficits in autism are due to impairment withinprefrontal cortex itself or to some other underlyingsystem-wide deficit such as its connectivity to otherregions. If there is general reduction in the functionalconnectivity among the brain regions in autism, asshown in a sentence comprehension task (Just et al.,2004), then one would expect reduced communicationand integration among the brain regions to alsoundermine executive function in TOL.Our study focused on the interdependence of functionally related brain regions during theperformance of a TOL task. The theoretical rationale forthis focus is that it is becoming clear that thinking is anemerging property of a large-scale network of collaborating cortical areas. Therefore, to characterizeneural functioning in autism, it may be necessary toexamine the cortical activation at a systems level ratherthan at the level of local brain regions (Just et al., 2004).One way to measure the synchronization among brainregions is to compute the correlation or covariancebetween the activation levels in two activated areas oversome time period. This measure generally showssystematic synchronization between areas, modulated bya number of variables. The synchronization is taken asevidence of

functional

connectivity

(

Friston, 1994;Horwitz, Rumsey, & Donohue, 1998).

The termfunctional connectivity has been used to describe theinterdependence of functionally related brain regions.The synchrony of the blood flow fluctuations in thefunctionally related brain regions implies the existenceof neuronal connections that facilitate coordinatedactivity. Functional connectivity between two brainregions is assessed as the correlation between pairs of measurements of cerebral blood flow (PET) or bloodoxygenation level (fMRI). Castelli et al. (2002) usedPET based correlation of activation levels between tworegions of interest across the participants in a theory of mind study and predicted that the visual areas may notbe properly connected with the cognitive areas in autism.Two older functional imaging studies using coarser-grain measures (e.g. Horwitz et al., 1988; Zilbovicius et

In Press,

Cerebral Cortex

Underconnectivity

in autism

3al. 1995) implicated lower inter-regional brainconnectivity in autism.In fMRI studies, functional connectivitymeasurements are based on the correlation of theactivation time-series in pairs of brain areas. The timeseries in this study included an observation every 3sec (i.e. a TR of 3 sec) while participants wereperforming the TOL task. The general assumption isthat the functioning of voxels whose activation levelsrise and fall together is coordinated. The functionalconnectivity was measured between some of the keyareas involved in executive processing, and then wascompared between the autism and controlparticipants. The main hypothesis was that therewould be a lower level of functional connectivityamong the autism participants in the frontal-parietalnetwork.The functional connectivity in the TOL task,which is known to engage prefrontal and parietalareas bilaterally (Newman et al., 2003), might welldepend on the corpus callosum as part of thebiological infrastructure that permits communicationamong brain areas. This study measured the size of the various segments of the corpus callosum of eachparticipant in the functional imaging study,hypothesizing that the sizes of key areas would besmaller in the autistic participants, following similarprevious findings in purely morphometric studies(Egaas et al., 1995; Hardan et al., 2000; Piven et al.,1997). Moreover, for the first time, this study testsfor a correlation between the size of various corpuscallosum segments and frontal-parietal functionalconnectivity. The secondary hypothesis was that inthe participants with autism, there would be apositive correlation, because the size of the corpuscallosum is constraining the functional connectivity.In the control group, there should be no correlationbecause there is no constraint on informationprocessing imposed by the size of their corpuscallosum and their neural connectivity. That is, theirneural resources and neural connectivity are assumedto always be adequate to meet these task demands.

Methods

Participants.

Eighteen high-functioningindividuals with autism (mean age 27.1 years, SD =11.9) and eighteen healthy participants (mean age24.5 years, SD = 9.9) were included in the study (FullScale and Verbal IQ scores of 80 or above, as shownin Table 1. The diagnosis of autism was establishedusing the ADI-R (Autism Diagnostic Interview-Revised, Lord et al., 1994), the ADOS-G (AutismDiagnostic Observation Schedule-General, Lord et al.,2000), and confirmed by expert clinical diagnosis. Nineof the participants with autism were taking psychotropicmedications. Of these, six were taking only onemedication, a serotonin reuptake inhibitor, but not on theday of the scan. All participants were required to be ingood medical health. Potential autistic participants wereexcluded if they had evidence of an associatedinfectious, genetic, or metabolic disorder, such asfragile-X syndrome or tuberous sclerosis. Potentialcontrol and autistic participants were also excluded if found to have evidence of birth asphyxia, head injury, ora seizure disorder. Exclusions were based on neurologichistory and examination, physical examination, andchromosomal analysis or metabolic testing if indicated.Written informed consent was obtained from participantsand/or their guardians, using procedures approved by theUniversity of Pittsburgh Medical Center InstitutionalReview Board.The control participants were communityvolunteers recruited to match the autistic participants onage, Full Scale IQ, gender, race, and family of originsocioeconomic status, as measured by the Hollingsheadmethod (Hollingshead, 1957). Potential controlparticipants were screened by questionnaire, telephone,face-to-face interview, and observation during screeningpsychometric tests such as the Wechsler AbbreviatedScales of Intelligence (WASI) (Wechsler, 1999) and TheWide Range Achievement Test 3 (WRAT3). Familyhistory of developmental and neuropsychiatric disorderswas obtained using a questionnaire specificallydeveloped for the CPEA research program. Exclusionarycriteria, evaluated through these procedures, includedcurrent or past history of psychiatric and neurologicdisorders, birth injury, developmental delay, schoolproblems, acquired brain injury, learning disabilities,and medical disorders with implications for the centralnervous system or those requiring regular medicationusage. Potential control participants were also screenedto exclude those with a family history of autism,developmental cognitive disorder, learning disability,affective disorder, anxiety disorder, schizophrenia,obsessive compulsive disorder, or other neurologic orpsychiatric disorder thought to have a geneticcomponent. There were no statistically reliabledifferences between the autistic and control participantsin age or IQ. All participants were Caucasian. Twelve of the participants with autism and 9 control participantswere previously included in a study of functionalconnectivity in a sentence comprehension task (Just etal., 2004), and one participant with autism waspreviously included in a study of functional connectivityin a verbal working memory task (Koshino et al., 2005).