Motor Activity in Children With Autism a Review.2

R E V I E W A R T I C L E

Motor Activity in ChildrenWith Autism: A Review ofCurrent LiteratureRebecca Downey, PT, DPT; Mary Jane K. Rapport, PT, DPT, PhD

Physical Therapy Program, School of Medicine, University of Colorado, Denver, Colorado.

Physical therapists have expanded their role and visibility in the treatment of children with autism spectrumdisorders (ASD). Limitations in motor activity have not been considered in the assessments of core deficitsof this population; however, physical therapists should be prepared to discuss and address these limitationsin children with ASD. Purpose: The primary purposes of this review were to summarize current evidence formotor activity limitations in children with ASD and suggest further areas of research in physical therapy andautism while considering how physical therapy may benefit children with autism. Method: A literature searchwas carried out in 2009 and 2010 by using multiple search engines. Results: Forty-nine articles met inclusioncriteria and were included in the review. Conclusion: Findings indicate that limitations in motor activity maybe present in individuals with ASD, and further research is needed to identify specific functional limitations.(Pediatr Phys Ther 2012;24:2–20) Key words: Asperger syndrome, autism spectrum disorder, child, female,male, motor activity, pervasive developmental disorder, systematic review

INTRODUCTION

Autism spectrum disorder (ASD) is a neurodevelop-mental disorder that is characterized by limitations in so-cial interactions and communication, restricted interest,and stereotyped or repetitive behaviors.1 The term autismspectrum disorder is often used to describe individuals whohave been diagnosed with autism disorder (AD), pervasivedevelopment disorder not otherwise specified (PDD-NOS),or Asperger syndrome (AS) on the basis of medical and de-velopmental history and clinical observations of behavior(see Table 1 for definitions).2,3 Recent research supported

0898-5669/110/2401-0002Pediatric Physical TherapyCopyright C© 2012 Wolters Kluwer Health | Lippincott Williams &Wilkins and Section on Pediatrics of the American Physical TherapyAssociation

Correspondence: Mary Jane K. Rapport, PT, DPT, PhD, Physical Ther-apy Program, School of Medicine, University of Colorado, 13121 E 17thAvenue, C244, Aurora, CO 80045 ([email protected]).

Grant Support: This study was funded by Leadership Education in Neu-rodevelopmental and Related Disorders Training (LEND), Health Re-sources and Service Administration (HRSA), and Maternal Child HealthBureau (MCHB), Award T73 MC11044.

The authors declare no conflict of interest.

DOI: 10.1097/PEP.0b013e31823db95f

by the Centers for Disease Control and Prevention4 indi-cates that based on parent report, the incidence of autismis 110 per 10 000, with a higher incidence in males thanin females.5

The approach to evaluation and treatment of childrenwith ASD is frequently a multidisciplinary team approach.6

Although the term clumsiness has been used in describingindividuals with AS, limitations in motor activity are notconsidered to be core deficits of individuals with ASD.7,8

Historically, physical therapists have not been involved inthe evaluation and treatment process of individuals withASD; the role of the physical therapist is evolving withregard to both the evaluation and intervention processes.There is an increasing body of literature related to ASD,but evidence regarding physical therapy (PT) and inter-vention with this population continues to be limited inscope. To best understand PT intervention in childrenwith autism, physical therapists need to assess the effectof motor activity limitations including motor anomalies,delays, or weaknesses on a child’s ability to fully par-ticipate in daily activities and routines. These deficits inmotor activity inform physical therapists’ clinical decisionmaking.

The purposes of this review were to summarize cur-rent evidence for limitations in motor activity in childrenwith ASD and suggest further areas of research relatedto PT and autism while considering how PT may benefit

Copyright © 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins and the Section on Pediatrics of the American Physical TherapyAssociation. Unauthorized reproduction of this article is prohibited.

2 Downey and Rapport Pediatric Physical Therapy

TABLE 1Definitions and Abbreviations of Developmental Disorders Identified in the Literature and Associated With Autism

Classification Definition

Autism spectrum disorder(ASD)1

ASD is a neurodevelopmental disorder that is characterized by limitations in social interactions andcommunication, restricted interest, and stereotyped or repetitive behaviors. There is a continuum ofbehaviors represented within the ASD diagnosis

Pervasive developmentaldisorder (PDD)a

The PDD diagnosis includes impaired social interaction and communication skills or the presence ofstereotyped behaviors or restricted interests that are not congruent with developmental or cognitive ages.PDD encompasses several disorders including autistic disorder, Rett’s disorder, childhood disintegrativedisorder, Asperger’s disorder, and PDD not otherwise specified.

Autism disordera The diagnosis of autistic disorder is based on impaired social interaction and communication and the presenceof repetitive or stereotyped behavior. There must also have been a delay in social interaction, social orcommunicative language, and play prior to the age of 3 years.

Asperger syndromea,b The diagnosis of Asperger syndrome is based on impaired social interaction and restricted or stereotypedinterests that interfere with daily functioning. There is no delay in language, cognitive development, oradaptive behaviors and activities of daily living skills.

Pervasive developmentaldisorder—not otherwisespecified (PDD-NOS)a,b

The diagnosis of PDD-NOS is used when there is impairment in social interaction that is associated withcommunication skills or is present with stereotyped behavior and restricted interest. These symptoms shouldnot be accounted for by PDD, schizophrenia, schizotypical personality disorder, or avoidant personalitydisorder. PDD-NOS includes “atypical autism” (when the criteria have not been met for autism disorder).

aAdapted from the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition (DSM-IV).1bBased on proposed revisions of the DSM, these currently used diagnoses may be incorporated into one diagnosis (ASD) when the fifth edition is

published.3

children with autism. The term motor activity was selectedby the authors in an attempt to capture and describe mo-tor abnormalities, delays, and general motor function ofchildren with ASD.

METHODS

A literature search was carried out using OVID,PubMed, and Google Scholar search engines between

January 1, 2009, and October 31, 2009. Seventeen searchterms were used in an attempt to best capture the broadrange of articles addressing children with autism or re-lated diagnoses, motor activity, and PT (see Figure forsearch terms). Care was given to be certain that each searchyielded all possible evidence in the published literature.The search was then updated during May 2010 using thesame strategies with each of the 17 search terms. In all,90 searches were completed (Figure). Further review of

Fig. Search strategy.


Pediatric Physical Therapy Motor Activity in Autism 3

an article was dependent on the appearance of the searchterms in the abstract of the article as identified by the firstauthor (Figure). Articles were then graded according tothe level of evidence, on the basis of criteria from Sackettet al.9,10 Only articles that were found to be at Sackett levels1-3B were used for this analysis. Other exclusion criteriaincluded systematic reviews that did not include specificresearch methods, articles in which the search terms werepresent in the abstract but did not include analysis of motorfunction, and articles that were not accessible in English(see Figure).

RESULTS

The search strategy and inclusion criteria resulted ina total of 49 articles that were acceptable for this literaturereview of motor activity in children with ASD. No level1 articles appropriate for analysis were identified duringthe search. Two level 2 articles and 47 level 3 articles wereidentified and included in the analysis. Articles classified aslevels 4 and 5 were identified but not used in the literaturereview. Articles that originally met the inclusion criteriabut did not address motor activities as part of study criteriaor outcomes also were not included. Review articles thatdid not include specific methods and inclusion criteria alsowere excluded from the analysis (see Figure). As is thecase with all searches of the literature, additional articlesthat would meet the search criteria may be found if a newsearch following the same strategy was conducted today.The search strategy used here yielded all articles meetinginclusion criteria as of May 2010.

Studies that inform health care providers about chil-dren with ASD and addressed motor activity were ana-lyzed and reviewed (see Appendix). During the reviewprocess, various themes related to the characteristics andconcerns related to ASD arose in the literature. The articleswere grouped and analyzed on the basis of these themes.The themes were not derived through a specific qualita-tive methodological approach; rather, they were generatedthrough the work of the first author, and corroborated withthe second author, during the process of sorting articlesthat met the search criteria. Based on recurring similaritiesin the content of the available literature as reviewed, thefollowing thematic categories emerged: early motor find-ings, gestures and motor imitation, postural control, anddyspraxia.

Early Motor Findings

Several recent studies sought evidence of motor im-pairments to provide early motor identification markersand characteristics of ASD. For example, in a study byProvost et al,11 68% of children with ASD who were testedon the Bayley Scales of Infant Development II Motor Scaleand 63% of children who were tested on the Peabody De-velopmental Motor Scales, Second Edition, would havequalified for early intervention services. Landa and Garrett-Mayer12 also prospectively studied 2 groups of infants:

1 group was classified as low-risk for development ofautism and the other as high-risk for development ofautism. Children were examined with the Mullen Scalesof Early Learning at 6, 14, and 24 months. At the 24-month visit, further testing was administered to classifychildren who had typical development, ASD, or languagedelay. No differences on the Mullen Scales of Early Learn-ing were found in children with ASD and typical childrenat 6 months; however, by 14 months, the children withASD began to demonstrate a slowing in development com-pared with the other groups. By 24 months, significantdifferences were found between the group with ASD andthe group developing typically in all domains, as well asbetween the group with ASD and the group with languagedelay. The children with ASD demonstrated the slowestrate of increase in developmental skills over time.

In another study, Esposito et al13 retrospectively ex-amined videotapes of 3 groups of infants aged 12 to 21weeks: those diagnosed with ASD, those with develop-mental delays not associated with ASD, and children withtypical development. These researchers found that thegroup with ASD had significantly less static and dynamicsymmetry in the supine position than the other groups.Teitelbaum et al14 also suggested that motor abnormalities,including asymmetrical movement patterns, are present atbirth in children with ASD and may aid in the early identi-fication of ASD. On the basis of video analysis, Baraneck15

suggested that symptoms of autism, including sensory-motor symptoms, might be present and identifiable be-tween 9 and 12 months of age. In another study, Dewrangand Sandberg16 used retrospective parent reports to com-pare individuals with AS with a group of young adults whowere developing typically. They found that during the first2 years of life, individuals with AS demonstrated impairedimitation, increased clumsiness, and poor coordination.These findings suggest that evaluation of motor activitymay play a role in early prediction of ASD.

In contrast, Ozonoff et al17 reported that infants whoare later diagnosed with ASD do not demonstrate an in-creased number of movement abnormalities or a lack ofprotective reactions when compared with a group of chil-dren who are developing typically. They did, however, finda slower rate of development in reaching mature motor pat-terns. In addition, they suggested that more comprehen-sive motor evaluations might be useful in early detection ofautism. In another study by Loh et al,18 stereotyped behav-iors and postures found in children with ASD were similarto those in the comparison group. These authors also sug-gested that more sensitive testing might be required toidentify motor impairments.

Gestures and Motor Imitation

The ability to use gestures and motor imitation relieson motor activity to communicate with others. Interest-ingly, children with ASD have difficulty with communica-tion as well as difficulty using motor activity (eg, gesturesand imitation) as forms of communication to support social



interaction. Motor imitation has been identified as a signif-icant impairment in previous literature on individuals withASD, particularly in relation to social communication.19-21

In a literature review by Williams et al,22 the authors sug-gested that imitation impairments are present in childrenwith ASD and are more apparent in younger age groups(below the age of 4 years) when compared with otherchildren.

Stone et al21 suggested that imitation impairments inchildren with ASD are due to a delay in acquiring imita-tion skills, rather than disordered sequencing. Althoughsome improvement may be seen between the ages of 2 and3 years, this delay was apparent in young children, aswell as in preschool-aged children.22,23 In another study,Rogers et al20 found that children with ASD have impairedimitation skills on sequential imitation tasks when com-pared with a group of children with developmental delays.No support was found for a relationship between imitationimpairment and play skills, language skills, or dyspraxia inchildren with ASD. In a more recent study, Rogers et al24

noted that during a simple task, children with autism failmore imitation tasks than a group of children with devel-opmental delays and a group of children who were devel-oping typically. These findings were especially meaningfulin children with autism who were younger than 14 months.In the older age group (older than 30 months), childrenwith regressive type autism continued to fail more tasks,especially nonfunctional imitation tasks.

Other researchers have suggested a possible link be-tween imitation impairments and the presence of motor ac-tivity abnormalities. For example, Van Vuchelen et al25 re-ported the presence of an imitation impairment, especiallyin nonmeaningful gestures, in children with ASD whohave low-functioning autism and high-functioning autism(HFA). They also found increased impairment on motortesting. Green et al26 also found an association betweenmotor and imitation scores, with lower and more variablescores in children with AS than in children with specificdevelopmental delays of motor function. Spatiotemporalerrors were more common in the group with AS. In astudy by Mostofsky et al,27 children with ASD had moreimitation errors on the Florida Apraxia Screening Test thanchildren who were developing typically. Spatial errors werethe most common in both groups. In the groups with ASD,errors were increased when gesturing on command andimitation, rather than tool use. Notably, no significant dif-ference was found in the number of total errors betweenindividuals diagnosed with HFA and those with AS.

In another study, Smith and Bryson28 found that chil-dren with ASD have increased difficulty performing andnaming both meaningful and nonmeaningful gestures butno difficulty understanding or identifying gestures. Theauthors suggested that these impairments might be relatedto dyspraxia.28 Dewey et al29 also found that when com-pared with children with both developmental coordinationdisorder (DCD) and attention-deficit/hyperactive disorder(ADHD), DCD only, or ADHD only, children with ASDhave significantly lower motor and gestural performance

scores. Although this may be related to praxis, these au-thors suggested that it might also be related to alteredneural substrates or language deficits.

More recently, Ben-Sasson et al30 examined gesturalrepresentations in 3 groups of children: children with HFA,children with language impairment, and children who aredeveloping typically. The authors concluded that childrenwith HFA have increased difficulty with gestural repre-sentations, which they suggest may be related to a motorplanning deficit. There appeared to be decreased qualityof gesture performance as well as a discrepancy betweengestures and verbal descriptions, when compared with theother 2 groups. The authors hypothesized that perform-ing motor actions might be more difficult when the taskis combined with verbal description. This may be relatedto a lack of integration between motor and language. Ben-Sasson et al30 suggest that individuals with ASD may ben-efit from further testing to examine motor planning andsequencing.

Postural Control

Postural control requires a level of stability neces-sary prior to executing additional motor skills or activities.Without this control, motor activity may be limited to morestatic positions. Individuals with autism tend to have de-creased postural control.31,32 Minshew et al31 found thatindividuals with autism have decreased postural stability,particularly in circumstances where there is sensory con-flict. Compared to a group of children who were develop-ing typically and adults, development of postural stabilityappeared to be delayed in children with autism. Posturalstability did not appear to improve in individuals withautism until the age of 12 years. At the age of 15 years,the group that was developing typically appeared to havea plateau in postural stability; however, this same level ofcontrol was not achieved in the group with ASD. Basedon the data from a bimanual lift task by Schmitz et al,32

children with ASD rely on reactive postural control ratherthan on the typical anticipatory postural control seen inthe comparison group when performing lifting tasks.

In another study, Kohen-Raz et al33 examined postu-ral stability in various standing positions between a groupof children with ASD and a group of children who weredeveloping typically. Children with ASD demonstrated in-creased sway, abnormal weight distribution, and the ab-sence of typical ankle strategies in standing. A “paradoxi-cal stress response” was noted in individuals with autism,indicated by an increase in postural stability in stressfulconditions (defined as removal of vision). In a follow-upstudy, Molloy et al34 also found that children with ASDhad significantly more sway in standing than a compari-son group of children who are developing typically. Insteadof a “paradoxical stress response,” they found that chil-dren with ASD experienced a larger increase in sway whenvisual input was removed and somatosensory input modi-fied, indicating that children with ASD rely on visual inputfor balance. This finding points to impaired processing



abilities with sensory conflict in individuals with ASD.Molloy et al34 argued that the presence of a paradoxicalresponse found in the Kohen-Raz et al33 study might havebeen the result of additional visual and auditory input usedin their methods. Further support for impaired posturalcontrol in children with ASD was provided by Fournieret al,35 who reported that children with ASD have increasedpostural sway in quiet stance without manipulation of sen-sory input as well as altered center of pressure shifts duringgait initiation.

Dyspraxia

Quality of movement may be altered in children withASD, and dyspraxia has been noted when comparing chil-dren with ASD with children who are developing typically.Motor delay and motor variability have also been noted anddescribed in some studies and refuted in others. Althoughmotor abnormalities were noted in individuals who werefirst described with AS,8 these limitations have not beenconsistently identified in individuals with ASD. While nolevel 1 research supporting the presence of limitations inmotor activities was found, several lower-level studies indi-cate that impairments in motor activity may be common inchildren with ASD. Researchers have sought to distinguishdifferences across individuals with ASD, as well as to com-pare those with ASD with individuals who are developingtypically or have other developmental concerns.

Manjiviona and Prior36 found limits in motor activ-ity and function in children with HFA and AS. Ghaziuddinand Butler37 noted that children with autism, AS, and PDD-NOS have motor impairments. Statistically significant dif-ferences were found only between those with AS and thosewith AD (children with AD were noted to be more clumsy)as measured by the Bruinink-Oseretsky Test of Motor Pro-ficiency. Individuals diagnosed with AD had higher levelsof motor activity impairment, while those with AS demon-strated less impairment on gross motor, fine motor, andtotal battery scores. A strong correlation existed betweenintelligence quotient (IQ) scores and test results. Whenadjusted for level of intelligence, no significant differenceremained between groups. Green et al38 found a similarcorrelation between IQ and motor scores in individualswith ASD, indicating that motor impairments might be re-lated to IQ level. Current research has demonstrated thatthere is no significant difference in level of motor impair-ments among children with AS, AD, or PDD-NOS; how-ever, there is a trend toward higher cognitive limitationscorrelated to lower motor scores.

Several researchers have suggested that when com-pared with a comparison group developing typically, chil-dren with ASD have dyspraxia. Mostofsky et al27 suggestedthat based on the Florida Apraxia Screening test, motor im-itation may be linked to dyspraxia, particularly to a delayin spatial mapping. Dzuik et al39 proposed that dyspraxiamay be separate from other motor skills in children withASD and may be strongly correlated to the core deficits as-sociated with autism.1 According to Dowell et al,40 praxis

score was correlated with the Autism Diagnostic Obser-vation Schedule score, also suggesting that dyspraxia maybe a core symptom of ASD. In addition, Dowell et al40

reported that children with ASD have slower timed move-ments and score significantly lower on postural knowledgetesting.

Glazebrook et al41 noted that individuals with ASDare able to use advance information; however, more timeis required to plan movements. Glazebrook et al42 and Naz-zarali et al43 noted that individuals with ASD required moretime to plan and execute goal-directed movements. Rine-hart et al44 also noted a slower preparation time in childrenwith HFA and AS when compared with a cohort developingtypically. This was further supported in another study byRinehart et al,45 where children with HFA demonstratedincreased preparation time compared with a cohort de-veloping typically, and children with AS demonstrated atrend toward a motor preparation deficit. On the basis ofan analysis of goal-directed gait, Vernazza-Martin et al46

suggested that when compared with a typical comparisongroup, children with ASD have impaired motor planningand execution. Whereas differences in motor planning arepresent in some individuals with ASD, the alterations inpatterns are unclear. Hughes47 suggested that individualswith autism demonstrate difficulty executing simple goal-directed motor tasks that might be related to sequencing,vision, or consequence prediction.

Staples and Reid48 compared a group of children di-agnosed with ASD to 3 groups developing typically. The3 typical groups were matched with children in the groupwith ASD by chronological age, cognitive development,or movement skill development. They found that childrenwith ASD have significantly poorer motor scores than chil-dren who are developing typically and who are chronolog-ically age matched and cognitively age matched. Specifi-cally, children with ASD had difficulty with bilateral coor-dination and performed at a similar motor level as childrenapproximately of half their chronological age. The authorssuggested that by late childhood, motor skills in childrenwith ASD are significantly delayed.

Jansiewicz et al49 noted that boys with HFA and AShave increased difficulty with balance, gait, and dysrhyth-mia with timed hand and foot movements. In anotherstudy, Weimer et al50 examined tests of apraxia and ba-sic motor function in a group of children and young adultswith AS and a comparison group. The authors found thatdeficits were present on tests of apraxia, especially on mea-sures where visual input was removed, suggesting relianceon vision with a proprioceptive impairment. In addition,Freitag et al51 reported that when compared with a groupdeveloping typically, individuals with HFA and AS are“strongly impaired” in dynamic balance and diadochokine-sis, and integration between sensory and motor input maybe also impaired. They also noted a positive association ofmotor scores to the level of social withdrawal.

In a study by Fuentes et al,52 the authors found thatwhen compared with a group of children who were devel-oping typically, children with ASD had significantly poorer



motor and writing scores overall. More specifically, scoreswere decreased on gait, stance, and timed movement ac-tivities. Poor gross motor skills were correlated with poorhandwriting scores. They hypothesized that if therapiesaddress overall motor control, handwriting scores wouldimprove as a result of increased ability to control and ma-nipulate arm movements. In contrast, van Swieten et al53

sought to differentiate between motor and executive plan-ning abilities in 3 groups of children and adults: a groupwith ASD, a group with DCD, and a group with typicaldevelopment. Based on what the authors suggested to bea pure motor planning test, no difference was found be-tween the group with typical development and the groupwith ASD.

Miyahara et al54 found high rates of motor delay in2 groups of children: a group of children with AS and agroup of children with a learning disability. A significantdifference was found between groups for manual dexterityscores. The children with AS had a trend toward poorer ballskills, which, the authors hypothesized, might be relatedto the type of preferred play.

No difference in motor profile was found betweenchildren with ASD who were chronologically age matchedand children with other developmental delays.55 The au-thors did, however, note that scores of children with ASDwere more variable than those with developmental delays.In another study by Provost et al,11 the authors evaluated3 groups of children—children with ASD, developmen-tal delays, and developmental concerns—on the basis ofresults from the Bayley Scales of Infant Development. Inthis study, none of the children in the group with ASDtested within normal limits, and at least 68% of these chil-dren would have qualified for early intervention servicesbased on a delay of 25% or more. Motor scores of childrenwith ASD did not differ when compared with childrenwith developmental delay. These results suggest that thereis a limitation in motor function in children with ASD.Further research is needed to identify the specific type ofdysfunction.

Matson et al56 also noted gross and fine motor impair-ments in toddlers with AD when compared with toddlerswith atypical development between 18 and 36 months ofage. No significant differences were found between tod-dlers with PDD-NOS and children who are developingtypically. The authors suggested that motor impairmentsare present at a young age and benefit may be obtainedfrom early intervention. In another study, children withASD were compared with children with specific speechand language disorders and a comparison group of chil-dren who were developing typically. The children withspeech disorders had lower scores on all gross and fine mo-tor domains except coordination, whereas children withautism had significantly poorer scores on all fine and grossmotor scores (including balance), except oral motor andcoordination.57 Morin and Reid58 noted that although in-dividuals with autism have poor motor performance, theycan obtain higher balance scores than those with intel-

lectual delay. Balance in individuals with ASD may bedecreased; however, continued research is necessary toidentify the severity and the pattern of deficit.

Although impaired motor skills and function are nota core deficit of ASD, they are considered a core deficitof DCD and ADHD.1 In a study by Dewey et al,29 the au-thors found that when tested on the Bruinink-OseretskyTest of Motor Proficiency (short form) and a gestural per-formance test, children with ASD had significantly lowerscores than children with DCD, ADHD, or ADHD andDCD. This supports the presence of difficulty with motoractivity in children with ASD. This finding was further sup-ported by Pan et al,59 who found that children with ASDperformed significantly lower on motor tests than childrenwith ADHD and children who are developing typically.Specific limitations were noted on tests of locomotion andobject manipulation. They suggested that poor motor per-formance might be a sign of autism, with poor skills beingrelated to a lack of social skills as well as lack of motivationto practice. The authors encouraged clinicians to screen formotor impairments as poor motor skills were found to becorrelated with poor self-esteem, increased anxiety, anddecreased social function.

In a rare study with an all-female sample, Kopp et al60

compared several groups of girls: girls who were develop-ing typically, girls with ASD, and girls with ADHD. Theyfound that a large percentage of girls with ASD also fit thediagnosis for DCD, especially those in the preschool-agedgroup. Predictors of poor motor scores included youngerage, presence and severity of ASD symptoms, and low IQ.Poor motor scores were related to poor activities of dailyliving and physical education participation.

DISCUSSION

Based on this review of literature, evidence is emerg-ing that supports the identification of impaired motor ac-tivity in children who have the diagnosis of autism. Al-though impaired motor activity is not included in the di-agnosis, impaired motor activity appears to be an observ-able trend. The ability to understand and address the entireclinical picture of the child, including all areas of function,becomes an essential component of any intervention plan.The majority of current evidence does support the pres-ence of motor activity abnormalities prior to 2 years of agein children who are later diagnosed with ASD that per-sists into early childhood. As children are being diagnosedearlier with ASD and receiving early intervention services,physical therapists should consistently be part of the teamaddressing all the needs of the child.

Although we still have much to learn about the timingof motor development and the patterns of motor activity inchildren with ASD, evidence supports the presence of spe-cific difficulties related to motor activity in individuals withASD. The link between imitation and motor activity is stillunclear; however, research indicates that there are limita-tions in motor imitation in children with ASD. Questions



still remain as to whether restrictions in social behaviorlimit imitation, or whether limitations in motor activityrestrict social participation and adversely affect imitation.The presence of postural instability is also supported inthe literature. Decreased postural stability can significantlylimit participation in activities since the simplest of move-ments require complex control61,62 and further research isneeded to examine the severity, cause, and functional out-comes related to postural control. The literature also lendssupport to difficulty with motor planning in children withASD.

Limitations in motor activity in children with ASDmight decrease the opportunity for social interactions andlearning opportunities. Although limited research relatedto interventions for motor activity impairments in childrenwith ASD was found, Travers et al63 did address motor-linked implicit learning in children with ASD. In a studycomparing a group of individuals with HFA, ASD, and agroup developing typically, Travers et al63 reported thatfor simple tasks, motor-linked implicit learning might beintact in children with ASD, which might influence thera-peutic approach.

Motor activity delays have been observed in infantsand toddlers with autism and may affect future motor de-velopment. As with all developmental delays, early iden-tification leading to the initiation of early service deliverymight have a positive benefit on motor skills and long-termdisability in individuals diagnosed with ASD.64,65 Sincechildren with ASD are ambulatory, they previously wouldnot have been thought to benefit from PT. Physical ther-apists need to consider how to address these impairmentsin motor activity within the child’s daily routines.

Although the results of this literature review do in-dicate that motor activity impairments may be present inchildren with ASD, there are limitations to this analysis.An updated literature search may provide increased evi-dence supporting motor activity impairments as well asprovide documentation for intervention strategies for chil-dren with ASD. No consistent measure was used in eachof the reviewed studies, which limits generalization of thefindings.

CONCLUSION

Although there may be limitations in motor activ-ity present in children with ASD, much research is stillneeded to identify the age at which these limitations inmotor activity are present and to what extent they differfrom children who are developing typically. As with manyother diagnoses, lack of evidence has not prohibited the PTprofession from forging ahead with new research and on-going intervention. Future research is necessary to identifyto what extent functional activities are limited in childrenwith ASD. Research is also necessary to determine the un-derlying causes as well as the most appropriate interven-tions. As physical therapists move toward consistent useof the ICF,66 it may be useful to implement interventionsand research their efficacy based on activity limitation and

participation restrictions rather than solely by impairmentand disability. Physical therapists can and should play aunique role in promoting functionally based interventionstrategies to enhance motor activity and improve functionin children with ASD.

ACKNOWLEDGMENT

The authors thank Stephanie Lyle, PT, DPT, for herearly contributions to this work.

REFERENCES

1. American Psychiatric Association. Diagnostic and Statistical Manualof Mental Disorders: DSM-IV. 4th ed. Washington, DC: AmericanPsychiatric Association; 1994.

2. Rapin I, Tuchman R. Autism: definition, neurobiology, screening,diagnosis. Pediatr Clin N Am. 2008;55:1129-1146.

3. DSM-5: The future of psychiatric diagnosis. American Psy-chiatric Association DSM-5 Development. http://www.dsm5.org/Pages/Default.aspx. Published 2010. Accessed December 2010.

4. Centers for Disease Control and Prevention. Autism and de-velopment disabilities monitoring (ADDM) network. http://www.cdc.gov/ncbddd/autism/addm.html. Published March 2011. Ac-cessed June 2011.

5. Kogan MD, Blumberg SJ, Schieve LA, et al. Prevalence of parent-reported diagnosis of autism spectrum disorder among children inthe US, 2007. Pediatrics. 2009;124:1395-1403.

6. Rogers E. Functional behavioral assessment and children withautism: working as a team. Focus Autism Dev Disord. 2001;16:228-231.

7. Wing L. Asperger syndrome: a clinical account. Psychol Med.1981;11:115-129.

8. Tantam D. Annotation Asperger syndrome. J Child Psychol Psychiat.1988;29:245-255.

9. Sackett DL, Strauss SE, Richardson WS, et al. Evidence-BasedMedicine: How to Practice and Teach EBM. Philadelphia, PA:Churchill-Livingstone; 2000.

10. Glaros S. All evidence is not created equal: a discussion of levels ofevidence. PT: Magazine of Phys Ther. 2003;11:42-52.

11. Provost B, Lopez BR, Heimerl S. A comparison of motor delays inyoung children: autism spectrum disorder, developmental delay, anddevelopmental concerns. J Autism Dev Disord. 2007;37:321-328.

12. Landa R, Garrett-Mayer E. Development in infants with autismspectrum disorders: a prospective study. J Child Psychol Psychiat.2006;47:629-638.

13. Esposito G, Venuti P, Maestro S, Muratori F. An exploration of sym-metry in early autism spectrum disorders: analysis of lying. BrainDev-JPN. 2009;31:131-138.

14. Teitelbaum P, Teitelbaum O, Nye J, Fryman J, Maurer RG. Movementanalysis in infancy may be useful for early diagnosis of autism. ProcNatl Acad Sci U S A. 1998;95:13982-13987.

15. Baraneck GT. Autism during infancy: a retrospective video analysisof sensory-motor and social behaviors at 9-12 months of age. J AutismDev Disord. 1999;29:213-224.

16. Dewrang P, Sandberg AD. Parental retrospective assessment of devel-opment and behavior in Asperger syndrome during the first 2 yearsof life. Res Autism Spectrum Disord. 2010;4:461-473.

17. Ozonoff S, Young GS, Goldring S, et al. Gross motor development,movement abnormalities, and early identification of autism. J AutismDev Disord. 2008;38:644-656.

18. Loh A, Soman T, Brian J, et al. Stereotyped motor behaviors associatedwith autism in high-risk infants: a pilot videotape analysis of a siblingsample. J Autism Dev Disord. 2007;37:25-36.

19. Ingersoll B. The social role of imitation in autism implications for thetreatment of imitation deficits. Infant Young Child. 1998;21:107-119.



20. Rogers SJ, Hepburn SL, Stackhouse T, Wehner E. Imitation perfor-mance in toddlers with autism and those with other developmentaldisorders. J Child Psychol Psychiat. 2003;44:763-781.

21. Stone WL, Ousley OY, Littleford CD. Motor imitation in youngchildren with autism: what’s the object? J Abnorm Child Psych.1997;25:475-485.

22. Williams JHG, Whiten A, Singh T. A systematic review of action imi-tation in autism spectrum disorder. J Autism Dev Disord. 2004;34:285-299.

23. Stone WL, Lemanek KL, Fishel PT, Fernandez MC, Altemeier WA.Play and imitation skills in the diagnosis of autism in young children.Pediatrics. 1990;86:267-272.

24. Rogers SK, Young GS, Cook I, Giolzetti A, Ozonoff S. Imitation ac-tions on objects in early-onset and regressive autism: effects and im-plications of task characteristics on performance. Dev Psychopathol.2010;22:71-85.

25. Van Vuchelen M, Roeyers H, Weerdt WD. Nature of motor imitationproblems in school-aged males with autism: how congruent are theerror types? Dev Med Child Neurol. 2007;49:6-12.

26. Green D, Baird G, Barnett AL, Henderson L, Huber J, Henderson SE.The severity and nature of motor impairment in Asperger syndrome: acomparison with specific developmental disorder of motor function.J Child Psychol Psychiatry. 2002;43:655-668.

27. Mostofsky SH, Dubey P, Jerath VK. Developmental dyspraxia is notlimited to imitation in children with autism spectrum disorders. J IntNeuropsych Soc. 2006;12:314-326.

28. Smith IM, Bryson SE. Gesture imitation in autism: II. Symbolic ges-tures and pantomimed object use. Cogn Neuropsychol. 2007;24:679-700.

29. Dewey D, Cantell M, Crawford SG. Motor and gestural performancein children with autism spectrum disorders, developmental coordi-nation disorder, and/or attention deficit hyperactivity disorder. J IntNeuropsych Soc. 2007;13:246-256.

30. Ben-Sasson A, Stimmell KE, Cermak SA. Sequence of gestural repre-sentations in children with high functioning autism. Israeli J OccupTher. 2009;18:E57-E73.

31. Minshew NJ, Sung MB, Jones BL, Furman JM. Underdevelopmentof the postural control system in autism. Neurology. 2004;63:2056-2061.

32. Schmitz C, Martineau J, Barthelemy C, Assaiante C. Motor controland children with autism: a deficit of anticipatory function? NeurosciLett. 2003;348:17-20.

33. Kohen-Raz R, Volkmar FR, Cohen DJ. Postural control in childrenwith autism. J Autism Dev Disord. 1992;22:419-432.

34. Molloy CA, Dietrich KN, Bhattacharya A. Postural stability inchildren with autism spectrum disorder. J Autism Dev Disord.2003;33:643-652.

35. Fournier KA, Kimberg CI, Radonovich KL, et al. Decreased static anddynamic postural control in children with autism spectrum disorders.Gait Posture. 2010;32: 6-9.

36. Manjiviona J, Prior M. Comparison of Asperger syndrome and high-functioning autistic children on a test of motor impairment. J AutismDev Disord. 1995;25:23-39.

37. Ghaziuddin M, Butler E. Clumsiness in autism and Asperger syn-drome: a further report. J Intell Disabil Res. 1998;42:43-48.

38. Green D, Charman T, Pickles A. Impairment in movement skillsof children with autistic spectrum disorders. Dev Med Child Neurol.2009;51:311-316.

39. Dzuik MA, Larson JC, Apostu A, Mahone EM, Denckla MB, Mostof-sky SH. Dyspraxia in autism: association with motor, social, andcommunicative deficits. Dev Med Child Neurol. 2007;49:734-739.

40. Dowell LR, Mahone EM, Mostofsky SH. Associations of posturalknowledge and basic motor skill with dyspraxia in autism: implica-tion for abnormalities in distributed connectivity and motor learning.Neuropsychology. 2009;23:563-570.

41. Glazebrook CM, Elliott D, Szatmari P. How do individuals withautism plan their movements? J Autism Dev Disord. 2008;38:114-126.

42. Glazebrook CM, Elliott D, Lyons J. A kinematic analysis of howyoung adults with and without autism plan and control goal-directedmovements. Motor Control. 2006;10:244-264.

43. Nazzarali N, Glazebrook CM, Elliott D. Movement planning andreprogramming in individuals with autism. J Autism Dev Disord.2009;39:1401-1411.

44. Rinehart NJ, Bradshaw JL, Bereton AV, Tonge BJ. Movement prepa-ration in high-functioning autism and Asperger disorder: a serialchoice reaction time task involving motor reprogramming. J AutismDev Disord. 2001;31:79-88.

45. Rinehart NJ, Bellgrove MA, Tonge BJ, Brereton AV, Howells-RankinD, Bradshaw JL. An examination of movement kinematics in youngpeople with high-functioning autism and Asperger’s disorder: fur-ther evidence for a motor planning deficit. J Autism Dev Disord.2006;36:757-767.

46. Vernazza-Martin S, Martin N, Vernazza A, et al. Goal directed loco-motion and balance control in autistic children. J Autism Dev Disord.2005;35:91-102.

47. Hughes C. Brief report: planning problems in autism at the level ofmotor control. J Autism Dev Disord. 1996;26:99-107.

48. Staples KL, Reid G. Fundamental movement skills and autism spec-trum disorders. J Autism Dev Disord. 2010;40:209-217.

49. Jansiewicz EM, Goldberg MC, Newschaffer CJ, Denckla MB, LandaR, Mostofsky SH. Motor signs distinguish children with high func-tioning autism and Asperger syndrome from controls. J Autism DevDisord. 2006;36:613-621.

50. Weimer AK, Schatz AM, Lincoln A, Ballantyne AO, Trauner DA.“Motor” impairment in Asperger syndrome: evidence for a deficit inproprioception. Dev Behav Ped. 2001;22:92-101.

51. Freitag CM, Kleser C, Schnieder M, von Gontard A. Quantitative as-sessment of neuromotor function in adolescents with high function-ing autism and Asperger syndrome. J Autism Dev Disord. 2007;37:948-959.

52. Fuentes CT, Mostofsky SH, Bastian AJ. Children with autismshow specific handwriting impairments. Neurology. 2009;73:1532-1537.

53. van Swieten LM, van Bergen E, Williams JHG, et al. A test of motor(not executive) planning in developmental coordination disorder andautism. J Exp Psychol Human. 2001;36:493-499.

54. Miyahara M, Tsujii M, Hori M, Nakanishi K, Kageyama H, SugiyamaT. Brief report: motor incoordination in children with Asperger syn-drome and learning disabilities. J Autism Dev Disord. 1997;27:595-603.

55. Provost B, Heimerl S, Lopez BR. Levels of gross and fine motor de-velopment in young children with autism spectrum disorder. PhysOccup Ther Pediatr. 2007;27(3):21-36.

56. Matson JL, Mahan S, Fodstad JC, Hess JA, Neal D. Motor skillabilities in toddlers with autistic disorder, pervasive developmen-tal disorder—not otherwise specified, and atypical development. ResAutism Spectrum Dis. 2010;4:444-449.

57. Noterdaeme M, Mildenberger K, Minow F, Amorosa H. Evaluationof neuromotor deficits in children with autism and children with aspecific speech and language disorder. Eur Child Adolesc Psychiatry.2002;11:219-225.

58. Morin B, Reid G. A quantitative and qualitative assessment of autis-tic individuals on selected motor tasks. Adapt Phys Act Q. 1985;2:43-55.

59. Pan CY, Tsai CL, Chu CH. Fundamental movement skills in chil-dren diagnosed with autism spectrum disorders and attention deficithyperactivity disorder. J Autism Dev Disord. 2009;39:1964-1705.

60. Kopp S, Beckung E, Gillberg C. Developmental coordination disor-der and other motor control problems in girls with autism spectrumdisorder and/or attention-deficit/hyperactivity disorder. Res Dev Dis-abil. 2010;31:350-361.

61. Frank JS, Earl M. Coordination of posture and movement. Phys Ther.1990;70:855-863.

62. Wallman HV. The basics of balance and falls. Home Health Care ManagPract. 2009;21:436-439.



63. Travers BG, Klinger MR, Mussey JL, Klinger LG. Motor-linked im-plicit learning in persons with autism spectrum disorders. Autism Res.2010;3:68-77.

64. Gomby DS, Larner MB, Stevenson CS, Behrman RE. Long-term out-comes of early childhood programs: analysis and recommendations.Future Child. 1995;5:6-24.

65. Berlin LJ, Brooks-Gunn J, McCarton C, McCorminck MC. The effec-tiveness of early intervention: examining risk factors and pathwaysto enhanced development. Prev Med. 1998;27:238-245.

66. Jette A. Toward a common language for function, disability, andhealth. Phys Ther. 2006; 86:726-734.

APPENDIXBrief Summary of Each Article Revieweda

Author(s)Study Design/Methodology Study Groups Examination Toolsb Summary of Findings

Provost et al11 Comparison Autism spectrum disorder(ASD) (n = 19)

Developmental delay (DD)with concerns for motordelay chronologicallyaged matched within 3months (n = 19)

Developmental concernswithout motor delaychronologically agedmatched within 3 months(NMD) (n = 18)

Bayley Scales of InfantDevelopment IIMotor Scale (BSID II)

PeadbodyDevelopmentalMotor Scales, 2ndEdition (PDMS-2)

According to scores on the BSID II and thePDMS-2, 63% and 68% (respectively) ofchildren with ASD would qualify for earlyintervention services on the basis of a 25%motor delay. These scores were similar tothose of a group of children diagnosedwith DD.

Landa andGarrett-Mayer12

Prospectivecomparison

High risk for autism(n = 60)

Low risk for autism (n = 27)

Mullen Scales of EarlyLeaning (MSEL)

Participants were initially identified from2 groups: infants considered to be at highrisk of autism because they were siblingsof children with autism and infantsconsidered at low risk because there wasno family history of autism. Review of testscores and clinical judgment led tocategorization of these infants asunaffected, ASD, or language delayed. Onthe basis of MSEL scores, children withASD had slowed in development in alldomains except visual reception by 14months. By 24 months, significantdifferences were found between the groupwith ASD and the group of childrendeveloping typically on all domains of theMSEL.

Esposito et al13 Retrospectivecomparison(videoanalysis)

ASD (n = 18)Typical development (TD)

(n = 18)DD (n = 12)

Eschkol-WachmanMovement Notationstatic and dynamicsymmetry

Based on retrospective video analysis, infants(12-21 weeks) who were later diagnosedwith ASD had higher rates of asymmetryin supine static and dynamic lyingpostures. Symmetry was noted in somechildren with ASD; however, childrenwith early onset ASD were more likely todemonstrate lower levels of symmetry.

Teitelbaum et al14 Retrospectivecomparison(videoanalysis)

ASD (n = 17)TD (n = 15)

Eschkol-WachmanMovement Notation

Based on retrospective video analysis ofinfants, most of the children with ASDdemonstrated altered movement patternsin mouth shape and lying, righting, sitting,crawling, and walking that could beidentified within the first few months oflife.

Baraneck15 Retrospectivecomparison(videoanalysis)

Autism disorder (AD)(n = 11)

DD (n = 10)TD (n = 11)

Video analysis andcoding of behavioralcategories: looking,affect, response toname, anticipatorypostures,motor/objectstereotypies, socialtouch, sensorymodulation

Based on retrospective video analysis ofinfants between 9 and 12 months’corrected chronological age, subtlesensory-motor deficits were present ininfants who were later diagnosed with AD.Social deficits were also noted.

(continued)



APPENDIXBrief Summary of Each Article Revieweda (Continued)


Dewrang andSandberg16

Retrospectivecomparison(parentquestionnaire)

Asperger syndrome (AS)(n = 23)

TD (n = 12)

Parent questionnaire-symptoms of autismbefore the age of2 years (SAB-2)

According to results from retrospectiveparent questionnaire, individualswith AS demonstrated difficulties inthe first2 years of life with several areas ofdevelopment, including motorskills. Parents did report difficultywith imitation of motor skills andcoordination.

Ozonoff et al17 Retrospectivecomparison(videoanalysis)

AD (n = 54) including-Autism: no regression

(At(NR)) (n = 26)-Autism: regression (At(R))

(n = 28)DD (n = 25)TD (n = 24)

Infant Motor Maturityand AtypicallyCoding Scales

On the basis of retrospective videoanalysis, children who were laterdiagnosed with AD did notdemonstrate higher rates ofmovement abnormalities or fewerprotective responses before the ageof 2. There were slower rates ofmotor development noted in thegroup with At(NR) in early skillsand in the group with At(R) inwalking.

Loh et al18 Retrospectivecomparison(videoanalysis)

ASD (from a population ofchildren with siblingsdiagnosed with ASD)(n = 8)

Nondiagnosed siblings ofchildren with ASD(n = 9)

TD (n = 15)

Coding of motormannerisms duringstandardized testing

Videos were analyzed of children at 12and 18 months of age. The arm waveposture was more commonly seen inchildren with ASD in both agegroups. At 18 months, thehand-to-ear posture was noted inboth the group with ASD and thenondiagnosed siblings of childrenwith ASD. Overlap between allgroups was present for stereotypedbehaviors.

Rogers et al20 Comparison AD (n = 24)DD (mixed etiology)

(n = 20)Fragile × Syndrome (FXS)

(n = 18, not included ingroup comparisonanalysis) including:-FXS without AD(n = 13)-FXS with AD (n = 5)

Imitation batteryPraxis battery

Children with AD had decreasedimitation performance whencompared with children with DD orchildren developing typically. Nodifferences were found in motorskills between children with AD,DD, or children developingtypically, and no correlation wasfound between motor skills andimitation abilities in childrendiagnosed with AD.TD (n = 15)

Stone et al21 Part 1ASD (n = 18)DD (n = 18)TD (n = 18)

Part 1: Motor ImitationScale

Part 1: Children with ASD under 31/2years old have poorer imitationskills than children without ASD butwith developmental delays whenmatched on mental age,chronological age, and languageability. Difficulties were notedspecifically with imitation of bodymovements and nonmeaningfulactions. Difficulties were similaracross all groups, suggesting thatmotor imitation skills in childrenwith ASD may be delayed inacquisition and not disordered.

Part 2ASD (n = 26)

Part 2: Motor ImitationScale (only total,body, and objectscores)

Part 2: Motor imitation improved inchildren with ASD between the ageof 2 and 3 years.

(continued)





Williams et al22 Systematicreview

Based on a literature review of 21 studies,imitation deficits were present inchildren with ASD and more apparentin younger children (under the age of4) and with nonmeaningful tasks. Moreresearch is necessary to furtherdelineate autism and dyspraxia.

Stone et al23 Comparison AD (n = 22)Intellectually delayed with

an IQ less than 70(n = 15)

Hearing-impaired (n = 15)Language- impaired

(n = 19)TD (n = 20)

Imitation motor tasks(12 total)

Preschool-aged children with AD hadsignificantly lower motor imitationscores compared with preschool-agedchildren who have an intellectual delay,are hearing or language impaired, andchildren developing typically. Motorimitation scores strongly differentiatedchildren with AD from children withother developmental delays and may bea useful screening tool.

Rogers et al24 Comparison Early onset AD (n = 17)Regressive-onset AD

(n = 24)DD (n = 22)TD (n = 22)

Motor imitation in 2conditions:functional andnonfunctional

Children younger than 14 months in thecombined AD group demonstratedincreased errors on imitation taskswhen compared with the other groups,to which they were matched on thebasis of nonverbal skill age. All groupsof older aged children demonstratedsimilar imitation skills except for thegroup with regressive-onset AD, whodemonstrated impaired nonfunctionalimitation skills. No differences werefound between groups on error type orpattern, nor was there a difference inemulation of the task between groups.The results suggest that imitationdifferences are not due to a motorimpairment; rather, they may be due tomirror neuron network impairments oratypical brain mechanisms associatedwith the mirror neuron system.

Van Vuchelenet al25

Comparison Cognitive impairment(n = 21)

with IQ < 80 including-Low functioning ASD

(LFA) (n = 8)-Cognitive impairment

without ASD (n = 13)No cognitive impairment

with IQ > 80 (n = 34)including

-High functioning ASD(HFA) (n = 17)

-TD (n = 17)

PDMS-2- administeredto those in the groupwith a cognitiveimpairment

Movement assessmentbattery for children(MABC)-administered to thosewith highfunctioning IQ

Motor imitation test

Children with ASD who have beendiagnosed with LFA or HFA, whenmatched for age, sex, anddevelopmental level with a comparisoncohort, demonstrated increased errorswith imitation tasks. Children in thegroup with ASD also demonstratedsignificantly poorer motor scores thanthe comparison group. Based on theresults, the authors suggested thatdifficulties with imitation arise from adelayed “action production system.”

Green et al26 Comparison AS (n = 11)Specific developmental

disorder of motorfunction (SDD-MF)(n = 9)

MABCThe Gesture Test

Although not statistically significant,children with AS performed morepoorly and variably on the MABC andthe Gesture Test when compared withchildren with SDD-MF of similar age.All children with AS tested below the15th percentile on the MABC. In thegroup with AS, poor scores on theMABC were correlated with lowerscores on the Gesture Test.

(continued)





Mostofsky et al27 Comparison ASD (n = 21) including-HFA (n = 13)-AS (n = 8)Gender and age matched

TD (n = 24)Inclusion for all subjects:

IQ > 80

Florida ApraxiaScreening Test(Revised)

While error type was similar between the 2groups, children with ASD demonstratedsignificantly higher errors on the FloridaApraxia Screening Test than thecomparison group. In the group with ASD,children with HFA had significantly moreerrors on the body-part-for-tool than thegroup with AS; however, no othersignificant differences existed. The authorssuggested that these findings are indicativeof a praxis issue in children with ASD, notjust difficulty with imitation.

Smith andBryson28

Comparison AD (n = 20)Language impairment

chronologically andreceptive age matched(n = 20)

TD receptive age matched(n = 20)

Tests of:1. Memory and

comprehension ofgestures

2. Gesture productionand imitation

Children with AD had no difficultyunderstanding or recognizing motorgestures. They demonstrated increaseddifficulty naming and imitating gestures.The authors suggested that this might bedue to a praxis impairment with difficultyin mapping movements as well as therepresentation of movements.

Dewey et al29 Comparison ASD (n = 49)Developmental coordination

disorder (DCD) (n = 46)Attention-deficit/

hyperactivity disorder(ADHD) (n = 27)

ADHD and DCD (n = 38)TD (n = 78)

Bruininks-OseretskyTest of MotorProficiency ShortForm (BOT-SF)

The Gestures Test

On testing to assess gesturing, children withASD had significantly higher rates oferrors than children in the other groups.Overall, on motor testing, children withASD demonstrated significantly poorerscores with increased variability in scoresthan children in any of the other groups(41% of children with ASD did not meetcriteria for motor impairment based on theBOT-SF). Although this may be related topraxis, these authors suggested that itmight also be related to altered neuralsubstrates or language deficits, as errorswere still present when motor deficitswere accounted for.

Ben-Sasson et al30 Comparison HFA (n = 23) with IQ > 70including

-AD (n = 15)-Pervasive developmental

disorder–not otherwisespecified (PDD-NOS)(n = 11)

Language impairment(n = 23)

TD (n = 30)

Demonstration taskportion of the AutismDiagnosticObservationSchedule (ADOS)

Children with HFA or language impairmentdemonstrated significantly lower levels ofgesture representation than childrendeveloping typically on a demonstrationtask. The authors suggested that in thegroup with HFA, this might be attributedto motor planning or languageimpairments. Difficulties may have alsobeen exacerbated by the requirement tospeak and gesture, as well as the lack ofnatural environment.

Minshew et al31 Comparison HFA (n = 79)TD “group matched”

(n = 61)Inclusion for all subjects: full

scale and verbal IQ > 70

Dynamicposturography(EquiTest)

Individuals between the ages of 5 and52 years with HFA demonstrateddecreased postural control when tested onthe EquiTest compared to a sample withTD. Increased difficulty was noted duringthe conditions of sensory conflict. Posturalcontrol did not improve until the age of 12and individuals with HFA never achievedadult levels (plateau seen at approximately20 years of age in comparison group).

Schmitz et al32 Comparison AD (n = 8) (right-handdominance)

TD (n = 16) (right-handdominance)

Bimanual load lift taskwith kinematic andelectromyographicanalysis

During a bimanual load lift task, musclelatencies and increased unloading timefound in children with AD suggest adecreased use of anticipatory control seenin a group with TD. Children with ADdemonstrated an increased use of reactivepostural control.

(continued)





Kohen-Raz et al33 Comparison AD (n = 91)TD (n = 166)

Tetra-ataxiametrymethod for posturalcontrol duringposturographictesting

With posturographic testing, children with ADhad increased variability in performance andabnormal weight distribution with less use ofthe typical anteroposterior sway. Adolescentaged children with AD demonstrateddecreased stability when compared topreschool-aged children with TD. Notably,the authors also found a “paradoxical”response to stressful situations, as childrenhad increased postural stability in stressfulconditions (removal of vision).

Molloy et al34 Comparison ASD (n = 8)TD chronologically age

matched (n = 8)

Posturographic testing Children were tested in 4 balance positions. Onthe basis of the results, children with ASDhad less postural stability than the childrendeveloping typically with removal of visualcues and deviation of somatosensory cues.Children with ASD tended to rely on visualinput, demonstrated by increased sway withremoval of visual cues regardless ofsomatosensory input. A “paradoxical stressresponse” was not found. Authors reportedthat these results in a previous study mighthave been due to additional visual stimuli.

Fournier et al35 Comparison ASD (n = 13)TD chronologically age

matched (n = 12)

Posturographic testing Children with ASD demonstrated significantlyhigher levels of mediolateral andanteroposterior sway, as well as sway area,than children with TD during quiet stance.The authors also noted a decreaseddisplacement of the center of pressure towardthe swing leg during gait in the group withASD. This can cause a decrease in shift of thecenter of mass to the stance limb, creating anincreased need for postural control. Theauthors suggested that children with ASDdemonstrate postural instability.

Manjiviona andPrior36

Comparison AS (n = 12)HFA (n = 9)Inclusion for all subjects:

“normal or nearnormal IQ”

Test of MotorImpairment-Henderson Revision(TOMI-H)

On the basis of motor testing, children with ASand HFA have variability in motor activities.Fifty percent of the children with AS and66.7% of the children with HFAdemonstrated motor impairments whencompared with a normative sample data forthe TOMI-H. No significant differences werenoted between the 2 groups. A significantnegative correlation was found betweenTOMI-H scores and IQ. The authorssuggested that the lack of difference betweenthe group scores provides support for ASbeing included in ASD diagnoses, rather thanits own diagnostic classification.

Ghaziuddin andButler37

Comparison AD (n = 12)AS (n = 12)PDD-NOS (n = 12)

Bruininks-OseretskyTest (BOT)

Children in all 3 age-matched groupsdemonstrated motor activity impairmentswhen tested on the BOT. Children with ADscored significantly lower on the BOT thanthose with AS, and no other significantdifferences were noted between groups. Astrong correlation between IQ scores andmotor scores was found. The authors alsosuggested that while the BOT does test formotor impairments, a pattern of impairmentis not yet clear for individuals with ASD.

(continued)





Green et al38 Comparison AD (n = 45)ASD (broad) (n = 56)

MABCDevelopmental

CoordinationDisorderQuestionnaire

Motor impairments, as tested by theMABC (defined by <5th percentile),were present in 79.2% of all children(9-10 years old) in the study. Groupscores were similar with children inboth groups having definite motorimpairments. When IQ was accountedfor, 97.1% of the children with low IQ(<70) had definite motor problems, ascompared with 69.7% of children withtypical IQ (>70).

Dzuik et al39 Comparison ASD (n = 47)TD (n = 47)

Florida ApraxiaScreening Test(Revised)

Physical andNeurologicalAssessment of SubtleSigns (PANESS)

Children with ASD demonstratedsignificantly poorer scores on theFlorida Apraxia Screening Test and thePANESS, suggesting poorer basic motorskills as well as praxis. Scores on praxistesting significantly predicted scores onthe ADOS, whereas basic motor skilldid not. The authors suggested thatalthough individuals with ASD mayhave impaired basic motor skills,dyspraxia may actually be independentof motor skills and may be a coresymptom of ASD.

Dowell et al40 Comparison ASD (n = 37)TD (n = 50)

Florida ApraxiaScreening Test(modified forchildren)

PANESSPostural knowledge test

(modified forchildren)

The children in the group diagnosed withASD had significantly poorer scores onthe PANESS, the postural knowledgetest, and the Florida Apraxia ScreeningTest. No significant difference wasnoted between children with HFA andAS on the postural knowledge test,although age predicted praxis and IQdid not. When age and IQ wereaccounted for, postural knowledge andbasic motor score predicted praxisscore, however; when age, IQ, posturalknowledge, and basic motor score wereall accounted for, praxis performancewas significantly related to diagnosis.The authors found that praxisperformance was significantlyassociated with ADOS score, suggestingthat praxis may be a core symptom ofASD.

Glazebrooket al41

Comparison Part 1AD (n = 18)Without AD (n = 18)

Part 1: Calculation ofreaction andmovement timesduring an adaptedprecue paradigm

Part 1: Individuals with AD demonstratedlower and significantly more variablereaction times than those without AD.The group with AD also demonstratedsignificantly longer times to executemovement. Similar to individualswithout AD, those with AD were ableto use advanced visual cues to planmovements and decrease reaction time.

Part 2AD (n = 9)Without AD (n = 9)

Part 2: Calculation ofreaction time andmovement timesbased on a rapidaiming task adoptedfrom earlier research

Part 2: Individuals with AD demonstratedlower reaction times and longer timesto execute movement than thosewithout AD. Individuals with ADappeared to use results from the priortrial to assist in movement planning forthe current trial.

(continued)





Glazebrooket al42

Comparison AD (n = 9)Age-matched individuals

without AD (n = 9)

Calculation of reactiontime and movementtimes based on arapid aiming taskadopted from earlierresearch

The group with AD required moretime to plan movements. They alsodemonstrated significantly increasedtimes to perform movements withsignificantly decreased peakvelocities and peak accelerationsthan the group without AD. Verbalability was correlated to reactionand movement times and nonverbalability was correlated with reactiontimes. Overall, individuals in thegroup with AD demonstrated poorermotor ability than individualswithout AD.

Nazzarali et al43 Comparison Part 1AD (n = 12)Without AD (n = 12)

Part 1: Variation ofprotocol fromGlazebrooket al41

Part 1: Individuals with ADdemonstrated slower reaction andmovement times than the groupdeveloping typically. They were ableto use advance cues to planmovements and decrease reactiontimes.

Part 2AD (n = 12)Without AD (n = 12)

Part 2: Measurement ofreaction andmovement timesduring a reachingtask that was handmanipulated ordirectionmanipulated

Part 2: Individuals with ADdemonstrated increased difficultyreprogramming an already-plannedmovement. This was more difficultwhen the task required a change inhands than a change in directions.The authors suggested that thismight be due to a slowed visualresponsiveness for spatial attentionor inefficient connections betweenhemispheres of the brain.

Rinehart et al44 Comparison HFA with performance andverbal IQ >70 (n = 12)

TD (matched on age, sex,IQ) (n = 12)

AS (n = 12)TD (matched on age, sex,

IQ) (n = 12)

Measurement ofpreparation andmovement time witha serial-choicebutton-pressingapparatus

Although individuals with HFA and ASdemonstrated similar errors in aserial-choice task, both groupsdemonstrated increased preparationtime when compared with a cohortwith TD. The group with ASdemonstrated slower preparationmovements, while the group withHFA demonstrated a lack ofanticipation in preparation.

Rinehart et al45 Comparison HFA with performance andverbal IQ > 70 (n = 12)

TD (matched on age, sex,IQ) (n = 12)

AS (n = 12)TD (matched on age, sex,

IQ) (n = 12)

Upper extremitykinematic task tomeasure movementpreparation andmovement time

Individuals with HFA demonstratedsignificantly slowed preparationtimes when compared with a cohortwith TD. Although no significantdifference was found betweenindividuals with AD and a cohortwith TD, the authors did note atrend toward increased preparationtime in the group with AD. Theysuggested that there is a trueplanning deficit, rather than aslowed movement. This deficit ismore predominant in individualswith ASD. The authors furthersuggested that although motorimpairments may be present in bothgroups, the underlying cause may bedifferent and further research isnecessary to examine these causes.

(continued)





Vernazza-Martinet al46

Comparison AD (n = 9)TD (n = 6)

Gait analysis The authors found no significant differencebetween the group with AD and the groupwith TD on gait parameters (stride duration,step length, velocity, cadence, and string andstride length), except for stride length, whichwas found to be shorter in children with AD.On the basis of analysis, children with ADdemonstrated increased oscillations of thehead, shoulder, and trunk causing less stableand more variable posture. Although childrenwith AD demonstrated increased oscillations,they were able to stabilize in the frontalplane. Gait parameters and stability appearedsimilar between groups; however, locomotionpattern was not maintained in the group withAD. The group with AD demonstrateddifficulty with gait trajectory based on animposed goal, suggesting difficulties withmotor planning.

Hughes47 Comparison AD (n = 36)Moderate learning

disabilities (n = 24)TD (n = 28)

Reach, grasp, andplace task (BarGame) withexaminer report ofhand positions

Based on hand positioning during areach-and-grasp task, the authors suggestedthat individuals with AD have increaseddifficulty with executing even simplegoal-directed tasks when compared withchildren with moderate learning disabilitiesor children with TD. The authors suggestedthat the pattern seen in older children withAD is similar to that seen in preschool-agedchildren with TD, suggesting thatdevelopment is delayed rather than altered.They also hypothesized that the differencesmay be due to sequencing, vision, orconsequence prediction.

Staples andReid48

Comparison ASD (n = 25)TD, age matched (n = 25)TD, movement skill

performance matched(n = 22)

TD, mental age matched(n = 19)

Test of Gross MotorDevelopment

Children with ASD demonstrated significantlylower scores on locomotor and object controlscores than children who werechronologically age matched and mental agematched. No difference was found betweenthe group with ASD and the developmentalage matched, as they were matched on thebasis of motor scores. In general, childrenwith ASD had increased difficultycoordinating both sides of the body for a task,as arm movements were noted to be awkward.On the basis of comparison with childrenwho were developmentally matched, childrenwith ASD appear to be delayed rather thandisordered, as skills aligned with childrenwith TD approximately half their age.

Jansiewicz et al49 Comparison ASD (n = 40)TD with no neurologic or

psychiatric diagnoses(n = 55)

PANESS In the group with ASD, no differences werefound on the PANESS between individualsdiagnosed with HFA or AS. The group withASD demonstrated significant differencesfrom the group with TD on all variables of thePANESS, except for impersistence andpatterned timed movements (a trend towardsignificant was present). Overall, childrenwith ASD tended to have difficulty withbalance, gait, and timed movement of thehands and feet.

(continued)





Weimer et al50 Comparison AS (n = 10)TD age matched (n = 10)

Motor testing: FingerTapping, GroovedPegboard, TrailMaking, Assessmentof Apraxia,Finger-ThumbApposition,Assessment of Ataxia,Assessment ofVisuomotorIntegration

Individuals with AS demonstrated poorerscores on tests of apraxia, balance(tandem and single leg), andfinger-thumb apposition, but did notdemonstrate significantly poorer scoreson tests of basic motor function than acohort with TD. The balance scores werepoorer with eyes closed. On the basis ofthese results and the lack of dizzinessusually found with a vestibulardysfunction, the authors suggested thatindividuals with AS may have a relianceon visual input and a proprioceptivedeficit.

Freitag et al51 Comparison ASD with Full Scale IQ >70(n = 16) including AS(n = 4) HFA (n = 12)

TD IQ matched (n = 16)

Zurich NeuromotorAssessment

The group with ASD demonstratedincreased difficulty with dynamic balanceand diadochokinesis. The authorssuggested that these impairments mightbe a result of poor integration of motor,sensory, and executive function. Anassociation between motor scores, thecore symptoms of ASD, and level ofwithdrawal was also found.

Fuentes et al52 Comparison ASD (n = 14)TD (n = 14)

Minnesota HandwritingAssessment RevisedPANESS

Children developing typically hadsignificantly higher scores on thePANESS when compared with childrenwith ASD, particularly for the sectionsexamining gait and timed movements.Handwriting scores were lower in thegroup with ASD; however, children withASD did not demonstrate difficultyaligning or sizing letters. Overall scoreson the PANESS, as well as scores on thetimed movement section, were found tobe predictive of handwriting scores. Theauthors suggested that when overallmotor skills are addressed in childrenwith ASD, handwriting abilities mayimprove as a result of increased controland ease of manipulation.

van Swietenet al53

Comparison DCD (n = 27)ASD (n = 20)TD (n = 70)

Grasp and turn task tomeasure preferredgrip

Children with ASD demonstrated similargrip selection to age-matched childrendeveloping typically on a task that theauthors used to test motor planning. Theauthors hypothesized that the need formotor planning was not strong enoughto elicit a difference in the children withASD.

Miyahara et al54 Comparison AS (n = 26)Learning disabilities

(n = 16)

MABC Based on motor testing, 85% of the childrenwith AS and 88% of the children withlearning disabilities qualified for thediagnosis of SDD-MF, which the authorsreported as being 42-44 times higherthan the typical population. There was asignificant difference between groups formanual dexterity skills. The childrenwith AS trended toward poorer ballskills. The authors suggested that thistrend may be due to type of preferredplay or decreased interpersonal skills inchildren with AS.

(continued)





Provost et al55 Comparison ASD (n = 19)DD (n = 19)

PDMS-2 When children with ASD were matched withchildren with DD based on age, gender,and cognitive age, similar gross and finemotor profiles were found.

Matson et al56 Comparison AD (n = 117)PDD-NOS (n = 112)Atypically developing

without ASD (n = 168)

Battelle DevelopmentalInventory, 2ndEdition

There was no significant difference foundbetween children with AD and PDD-NOS(differences did approach significance), orchildren with PDD-NOS and childrenatypically developing on gross and finemotor scores. Children with AD did havesignificantly lower fine and gross motorscores than children who were developingatypically. Motor impairments werepresent in all groups. The authorssuggested that motor impairments inchildren with AD are apparent at an earlyage and they may benefit from earlyintervention services.

Noterdaemeet al57

Comparison AD (n = 11)Expressive language

disorder age and IQmatched (n = 11)

Receptive language disorderage and IQ matched(n = 11)

TD age and IQ matched(n = 11)

Standardizedneurologicalexamination: finemotor, gross motor,coordination,balance, and oralmotor for globalneuromotorimpairment score,performance times

Qualitatively, statistically significantdifferences were found on motorperformance between the children withTD and the children with expressive andreceptive language disorders in all areasexcept for coordination. Children with ADhad significantly poorer scores thanchildren with TD for all sections except fororal motor and coordination tasks.

Morin and Reid58 Comparison AD (n = 8)Intellectually delayed

(n = 8)

5 test items: dynamicbalance, catching,standing long jump(adapted from theBOT), throwing,running

On motor testing, there was an overall trend(not significant) toward poorer motorscores in the group with AD than thegroup that was intellectually delayed. Thegroup with AD demonstrated significantlyhigher balance scores than the group thatwas intellectually delayed, which theauthors suggested may be due to slowedmovements seen in functional play. Thegroup that was intellectually delayeddemonstrated superior target throwingskills. The authors suggested that lowmotor scores in children with AD might berelated to level of cognitive impairment,rather than diagnosis alone.

Pan et al59 Comparison ASD (n = 28)ADHD (n = 29)TD (n = 34)

Test of Gross MotorDevelopment–Second Edition,calculation of theGross MotorDevelopmentQuotient (GMDQ)

Children with ASD demonstratedsignificantly poorer scores on locomotion,object control, and GMDQ than childrenwith ADHD and children with TD.Children with ADHD also demonstratedsignificantly poorer scores on measuresthan children with TD. When children inthe group with ASD who demonstratedattention deficits were omitted fromanalysis, results still indicated thatchildren with ASD have poorer scores,suggesting that motor ability is not relatedto attention. The authors suggested thatbased on GMDQ, differences might besecondary to limited social skills.

(continued)





Kopp et al60 Comparison ASD (n = 20)ADHD (n = 34)TD age and IQ matched

(n = 57)

EB-test (used forchildren older than 6years), Cailler-AsuzaScale (children below4 years), MABC,motor-neurological-perceptualassessment

All study participants were female. Highrates of DCD were found in the group withASD and ADHD. In school-aged girls, 25%of those with ASD and 32% of those withADHD were diagnosed with DCD. In thepreschool-aged girls, 80% of girls withASD were diagnosed with DCD. On theEB-test, girls with ASD and ADHD scoredsignificantly lower than girls with TD, andgirls with ADHD had lower overall scoresthan those with ASD. On the basis ofoverall clinical picture, the authorssuggested that younger age, low IQ, andautistic symptoms are predictors for lowermotor scores.

aArticles are listed in the same order in which they appear in the article.bTests related to motor and imitation skills were reported in the table. Studies may have used other testing (such as the ADOS), for diagnostic purposes,

and these tests were not included in the descriptions of each study in the table. See specific studies for more details on tests used.



Documents

Motor Activity in Children With Autism a Review.2