Research report
Neuroanatomic contributions to slowed orienting of attention in children with autism

https://doi.org/10.1016/S0926-6410(99)00006-3Get rights and content

Abstract

Previous research has demonstrated that adult autistic patients are abnormally slow to orient attention, with degree of slowed orienting associated with severity of cerebellar hypoplasia. This research was extended to children who, at ages two through six, met diagnostic criteria for autism and underwent magnetic resonance imaging (MRI). An average of 3 years later, when old enough to participate in behavioral experiments, the children returned to the laboratory and completed a spatial attention paradigm. Degree of slowed attentional orienting to visual cues was significantly correlated with degree of cerebellar hypoplasia, but not with size of other neuroanatomic regions. Additionally, there was a trend for orienting speed to differ between diagnostic outcome subgroups; children with confirmed diagnoses of autism at time of behavioral testing had larger orienting deficits than those who no longer met diagnostic criteria for autism. This research is among the first to establish a specific brain-behavior link in autistic children.

Introduction

To date, relatively few studies of direct quantitative brain-behavior associations in autistic individuals have been attempted. For the most part, brain-behavior relationships in autism have been inferred either from behavioral findings, in the absence of concomitant neuroanatomic analyses, or from postmortem and imaging studies without concomitant behavioral analyses. PET and other metabolic studies conducted with autistic individuals have, for the most part, examined passive brain metabolism rather than specific hypothesis-driven behaviors, and have yielded somewhat inconsistent results [e.g., Refs. 11, 40, 47, 58]. Specific cognitive operations have been addressed in neurophysiological studies of autistic individuals; however, such methodology is limited in its ability to localize neural abnormalities, and therefore does not readily provide associations between specific cognitive and neural abnormalities. Studies utilizing both carefully measured neuroanatomic indices, and specific hypothesis-driven cognitive operations, are crucial in order to more clearly explicate brain-behavior relationships in autism.

Many brain systems have been hypothesized to be involved in autism; however, to date, the cerebellum has been the most consistently reported site of neuroanatomic abnormality. For example, postmortem evidence has revealed Purkinje neuron loss in 17 of the 19 autism cases examined to date, and in 13 of these autism cases Purkinje neuron loss or abnormality was present in both the vermis and hemispheres [Ref. [8]; for review see Ref. [17]]. Magnetic resonance imaging (MRI) studies from multiple laboratories have identified hypoplasia of the neocerebellar vermis in a majority of individuals, and hyperplasia of this region in a small minority [for review see Ref. [17]]. In a MRI study of the cerebellar hemispheres in autism, hypoplasia again was identified, along with an association between the degree of cerebellar hemisphere and vermis abnormality [42]. Although postmortem and MRI studies have yielded evidence of cerebellar damage in both the vermis and hemispheres in autism, the majority of MRI studies have used midsagittal vermis area as an index of cerebellar pathology, due to the relative ease and accuracy with which it can be measured.

A number of recent investigations have explored the potential contribution of cerebellar abnormalities to autistic behavior. In contrast to patients with cortical lesions or other developmental disorders, individuals with autism, as well as patients with focal cerebellar lesions, have been shown to be slow at making voluntary shifts of attention 1, 2, 22, and slow to orient spatial attention to a cued location 16, 52, 53, 54. In spatial cueing paradigms, for example, patients with autism or acquired cerebellar damage show exaggerated `orienting delays', such that compared to controls, reaction times to validly cued targets are significantly slower when there is a shorter cue-to-target delay than when there is a longer cue-to-target delay 13, 16, 52, 53, 54, 55. Furthermore, a recent study found that degree of slowed orienting in autistic adolescents and adults was associated with severity of cerebellar abnormality [53]. Several new functional MRI studies also have reported that focused visual attention [4]and shifts in visual attention [35]activate the lateral cerebellum in normal volunteers. Taken together, these findings add to a growing body of evidence that the cerebellum may be involved in the rapid deployment of attentional resources, perhaps as part of its role in prediction and preparation for learning 3, 18. To date, however, in autism these attentional behaviors have been studied only in adolescents and adults.

The current study extends this line of research to children with autism. As part of a larger longitudinal investigation of brain growth, behavior, and diagnostic outcome in autism, we examined attentional behavior in a group of children who had met diagnostic criteria for autism when they were between the ages of two and six. All of the children underwent MRI at the time of initial diagnosis. These children were considered to be `provisionally autistic', due to the fact that at young ages the diagnosis of autism is somewhat unreliable (i.e., approximately 15–25% of children who meet diagnostic criteria for autism at young ages no longer meet research diagnostic criteria for autism when they are several years older, depending on the criteria used; C. Lord, personal communication). The provisionally-autistic children returned to the laboratory to participate in the attention experiment an average of 3 years after initial diagnosis and imaging. The primary question was whether early neuroanatomic profile in these children would be related to later behavioral outcome. Following completion of the attention experiment, the children were re-diagnosed as part of the follow-up phase of the larger longitudinal investigation. As anticipated, a subgroup met diagnostic criteria for autism, enabling examination of the relationship between early neuroanatomy, later diagnostic outcome, and attentional behavior.

A number of different neuroanatomic regions have been measured in this group of provisionally-autistic children. Area measures of neocerebellar vermian lobules VI–VII are of particular interest for the reasons described above: (1) in autistic individuals this region on average is consistently found to be smaller than normal; (2) individuals with autism, like patients with acquired cerebellar lesions, are slow to orient and shift attention; and (3) this region recently was found to be associated with speed of attentional orienting. Other neuroanatomic regions—frontal lobe volume, total brain volume, hippocampus area, and corpus callosum area—also proposed to be abnormal in autistic individuals, were used as contrast sites for testing the hypothesis about specific associations between cerebellar integrity and attentional orienting.

The provisionally-autistic children and a group of chronological age-matched controls were administered a spatial attention paradigm modeled after Posner [44]. It was hypothesized that children with confirmed diagnoses of autism would have the largest orienting delays (i.e., slowest attention orienting), followed by children who no longer met criteria for autism, with chronological age-matched controls showing the smallest orienting delays. In the provisionally-autistic group as a whole, orienting delays were hypothesized to be larger in children with more severe neocerebellar vermian hypoplasia (regardless of diagnostic outcome), and smaller in children with neocerebellar vermis areas closer to normal. Orienting deficits were predicted to be unrelated to any of the contrast neuroanatomic sites examined. Due to slower orienting to cue information, children with more severe neocerebellar hypoplasia were expected to be more affected by the cue (i.e., have larger validity effects) at longer cue-to-target delays relative to short cue-to-target delays, than children with less hypoplasia.

Section snippets

Children provisionally-diagnosed with autism (PA)

Seventeen children, who were diagnosed with autism and underwent MRI between the ages of 2.6 and 6.3 years, participated in the current study. All are part of a prospective longitudinal study of brain development, behavior, and diagnostic outcome in children who meet diagnostic criteria for autism at an early age. Average length of time between initial diagnosis/imaging and participation in the behavioral components of the current study was 3 years. At time of initial diagnosis and imaging,

Diagnosis

When re-diagnosed following completion of the current study, 11 of the 17 PA children continued to meet diagnostic criteria for autism (AUT). Of the six children who no longer met diagnostic criteria for autism (PDD-NOS), five met diagnostic criteria for Pervasive Developmental Disorder-NOS, and one received the non-PDD spectrum diagnosis of Attention Deficit Hyperactivity Disorder (ADHD). The two diagnostic subgroups differed significantly in CARS (t(15)=5.28, p<0.001), ADI-R Nonverbal

Speed of attentional orienting

Results from this study provide further evidence for a role of the cerebellum in the rapid deployment of cognitive and attentional resources. In a behaviorally and neuroanatomically heterogeneous group of children who all met diagnostic criteria for autism at young ages, an index of cerebellar abnormality (area of cerebellar vermian lobules VI–VII) (a) was significantly correlated with speed of attentional orienting to visual cues in the Posner paradigm, and (b) was the only neuroanatomic site

Acknowledgements

We would like to thank the children who participated in this work. We gratefully acknowledge Heather Chisum and Angilene Cowles for assistance with neuroanatomic measurement, Alan Lincoln and Laura Schreibman for their contribution to initial diagnosis, and Jim Covington and Senia Pizzo for technical assistance. We also thank several anonymous reviewers for helpful comments. This manuscript is based in part on the first author's dissertation submitted to the University of California, San Diego

References (58)

  • E Courchesne

    Brainstem, cerebellar and limbic neuroanatomical abnormalities in autism

    Curr. Opin. Neurobiol.

    (1997)
  • N.J Minshew et al.

    A preliminary 31P MRS study of autism: evidence for undersynthesis and increased degradation of brain membranes

    Biol. Psychiatry

    (1993)
  • N.A Akshoomoff et al.

    A new role for the cerebellum in cognitive operations

    Behav. Neurosci.

    (1992)
  • N.A Akshoomoff et al.

    ERP evidence for a shifting attention deficit in patients with damage to the cerebellum

    J. Cogn. Neurosci.

    (1994)
  • N.A. Akshoomoff, E. Courchesne, J. Townsend, Attention coordination and anticipatory control, in: J.D. Schmahmann...
  • G Allen et al.

    Attentional activation of the cerebellum independent of motor involvement

    Science

    (1997)
  • American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders Third Edition-Revised, American...
  • American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders Fourth Edition, American...
  • G. Arthur, Adaptation of the Leiter International Performance Scale, Stoelting, Chicago,...
  • A Bailey et al.

    A clinicopathological study of autism

    Brain

    (1998)
  • S.M. Blinkov, I.I. Glezer, The Human Brain in Figures and Tables: A Quantitative Handbook, Plenum, New York,...
  • M.M Bristol et al.

    State of the science in autism: report to the national institutes of health

    J. Autism Dev. Disord.

    (1996)
  • M.S Buchsbaum et al.

    Brief report: attention performance in autism and regional brain metabolic rate assessed by positron emission tomography

    J. Autism Dev. Disord.

    (1992)
  • R.A. Carper, E. Courchesne, Inverse correlation between frontal lobe and cerebellum sizes in children with autism,...
  • B.J Casey et al.

    Dysfunctional attention in autistic savants

    J. Clin. Exp. Neuropsychol.

    (1993)
  • M.L Cheal et al.

    Central and peripheral precueing of forced-choice discrimination

    Q. J. Exp. Psychol. A

    (1991)
  • E. Courchesne, H.J. Chisum, J. Townsend, A. Cowles, J. Covington, B. Egaas, S. Hinds, M. Harwood, G.A. Press,...
  • E Courchesne

    Infantile autism: Part 2. A new neurodevelopmental model

    Int. Pediatr.

    (1995)
  • E Courchesne et al.

    Prediction and preparation, fundamental functions of the cerebellum

    Learn. Mem.

    (1997)
  • E Courchesne et al.

    Neural activity-dependent brain changes in development: implications for psychopathology

    Dev. Psychopathol.

    (1994)
  • E Courchesne et al.

    The cerebellum in sagittal plane: anatomical MR correlation

    Am. J. Neuroradiol.

    (1989)
  • E Courchesne et al.

    Abnormality of cerebellar vermian lobules VI and VII in patients with infantile autism: identification of hypoplastic and hyperplastic subtypes by MR imaging

    Am. J. Roentgenol.

    (1994)
  • E Courchesne et al.

    Impairment in shifting attention in autistic and cerebellar patients

    Behav. Neurosci.

    (1994)
  • E. Courchesne, R. Yeung-Courchesne, K. Pierce, Biological and behavioral heterogeneity in autism: role of pleiotropy...
  • F Curcio

    Sensorimotor functioning and communication in mute autistic children

    J. Aut. Child. Schiz.

    (1978)
  • G. Dawson, A psychological perspective on the early socio-emotional development of children with autism, in: D....
  • G. Dawson, A. Lewy, Arousal, attention, and the socioemotional impairments of individuals with autism, in: G. Dawson...
  • P DiLavore et al.

    Pre-linguistic autism diagnostic observation schedule (PL-ADOS)

    J. Autism Dev. Disord.

    (1995)
  • J Dobbing et al.

    Quantitative growth and development of human brain

    Arch. Dis. Child.

    (1973)
  • Cited by (127)

    • Autism and arousal

      2019, Arousal in Neurological and Psychiatric Diseases
    • Characterizing Attention with Predictive Network Models

      2017, Trends in Cognitive Sciences
      Citation Excerpt :

      In concert with a wide body of behavioral evidence (e.g., [26–28]) this result suggests that a fundamental organizing principle of what we call ‘attention’ is whether it is ‘pushed’ towards a goal-relevant stimulus or ‘pulled’ towards a stimulus with low-level salience. Studies have related attentional orienting to cerebellar volume in children with autism [29] and performance on a perceptual task requiring orienting to functional connectivity between visual and prefrontal areas [30], but so far these relationships have not been leveraged to predict behavior in novel individuals. Recently attention researchers have described another pair of large-scale brain networks that reveal an important dimension along which attention can vary: the so-called ‘task-negative’ and ‘task-positive’ networks [31].

    • Ocular motor disturbances in autism spectrum disorders: Systematic review and comprehensive meta-analysis

      2016, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      With respect to intrinsic control of visually guided saccades, Johnson et al. (2013) had overlapping participants with Johnson et al. (2012), therefore results from Johnson et al. (2013) were excluded from our analysis. Eight studies used attentional orienting/cueing tasks (Greene et al., 2011; Harris et al., 1999; Iarocci et al., 2004; Kuhn et al., 2010; Pruett et al., 2011; Senju et al., 2004; Stanley-Cary et al., 2011; Vlamings et al., 2005) however fundamental differences in study design, specifically different types of cueing stimuli (symbol vs. social cue), differences in stimulus onset asynchrony (SOA), number of conditions and dissimilarities between variables reported, meant that we were unable to pool results across these studies. Three studies examined memory guided saccades (Luna et al., 2007; Goldberg et al., 2002; Minshew et al., 1999) however differences in timing of stimulus onset meant that results from these studies were unable to be pooled.

    • Impairments to visual disengagement in autism spectrum disorder: A review of experimental studies from infancy to adulthood

      2014, Neuroscience and Biobehavioral Reviews
      Citation Excerpt :

      First, several of the studies report ISI durations of longer than 1000 ms between the presentation of the central and peripheral stimulus (Chawarska et al., 2010; Fischer et al., 2013; Kikuchi et al., 2011; Kawakubo et al., 2004; Mosconi et al., 2009; Pruett et al., 2011; data not available for Green et al., 2011). Examination of the effects of variable ISI durations indicates that ASD are slower to disengage at short ISIs (under 500 ms) but are no different from TD at longer ISIs (800 ms or longer; Harris et al., 1999; Casey et al., 1993; Townsend et al., 1996a,b, 1999; Wainwright-Sharp and Bryson, 1993; also replicated in recent meta-analysis by Landry and Parker, 2013). The delay in disengagement at short ISIs suggests that top-down preparation of the saccadic response is impaired in ASD and requires additional time to shift attention to a peripheral stimulus in the visual field (Kingstone and Klein, 1993a,b; Taylor et al., 1998).

    View all citing articles on Scopus
    View full text