Hindawi Publishing CorporationThe Scientific World JournalVolume 2013 Article ID 716890 9 pageshttpdxdoiorg1011552013716890

Research ArticleUsing the Virtual Reality-Cognitive Rehabilitation Approach toImprove Contextual Processing in Children with Autism

Michelle Wang1 and Denise Reid2

1 Office of Undergraduate Medical Education Queenrsquos University 80 Barrie Street KingstonON Canada K7L 3N6

2Virtual Reality and Neurorehabilitation Laboratory University of Toronto 160-500 University AvenueToronto ON Canada M5G 1V7

Correspondence should be addressed to Michelle Wang mwangqmedca

Received 6 August 2013 Accepted 17 September 2013

Academic Editors S Duport and B Unver

Copyright copy 2013 M Wang and D Reid This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Background This pilot study investigated the efficacy of a novel virtual reality-cognitive rehabilitation (VR-CR) interventionto improve contextual processing of objects in children with autism Previous research supports that children with autism showdeficits in contextual processing as well as deficits in its elementary components abstraction and cognitive flexibility MethodsFour children with autism participated in a multiple-baseline single-subject study The children were taught how to see objectsin context by reinforcing attention to pivotal contextual information Results All children demonstrated statistically significantimprovements in contextual processing and cognitive flexibility Mixed results were found on the control test and changesin context-related behaviours Conclusions Larger-scale studies are warranted to determine the effectiveness and usability incomprehensive educational programs

1 Introduction

Autism or autism spectrum disorders (ASD) refer to a groupof neurodevelopmental disorders that are characterized indiffering degrees by core deficits in social and communi-cation skills as well as distinct abnormal behaviours [1]The prevalence of autism in children is approaching 1 orapproximately 1 in 110 children [2] Although autism can bediagnosed as young as 18 months of age [3] the symptoms ofthis disorder last throughout an individualrsquos lifetime

An innovative combination of traditional cognitive reha-bilitation and virtual reality technology offers an interactivecognitive approach to autism intervention [4]This emergingframework focuses on remediating underlying cognitive im-pairments of disorders and uses virtual reality technology tomaintain a high level of engagement and attention from thechildren The current study used this approach to improvea specific cognitive impairment in children with autismcontextual processing of objects

11 The VR-CR Approach The virtual reality-cognitive reha-bilitation (VR-CR) framework was first proposed by Rizzoand Buckwater [4] for children with attention-deficit hyper-activity disorder (ADHD) Traditional cognitive rehabilita-tion employs specific exercises designed to improve cog-nitive functions through repetitive training exercises thatspecifically target the impaired component processes Intenserepetition of specific exercises is necessary to reorganizethe brain in a particular area however it places immensedemands on both the child and the instructor [5] Thispresents a greater problem when extending the cognitiverehabilitation approach to children with autism as they areparticularly difficult to engage Virtual reality makes cogni-tive programs accessible to children with autism through itscapacity to maintain their attention provide structured andindividualized activities and address their weaknesses whilebuilding on their strengths [6]

Virtual Reality is defined as a simulation of the real worldusing computer graphics [7] The defining features of a VR

2 The Scientific World Journal

program or application include interaction and immersionHuman-computer interaction is achieved through multiplesensory channels that allow children to explore virtual envi-ronments through sight sound and touch [7] Immersion isconsidered the degree to which the child feels engrossed orenveloped within the virtual environment [8]

A variety of display devices offer differing degrees ofimmersion and interactivity The current study employs atwo-dimensional flat screen projection system This systemhas motion-capture capabilities where a tracking camera isable to capture and project a childrsquos image and motions on-screen in real-time Being able to see oneself within the virtualenvironmentmay contribute to a greater sense of engagementand motivation during the task

Although VR systems have not yet targeted cognitiveimpairments in childrenwith autism these systemshave beensuccessful in improving specific behaviours such as followingdirections [9] crossing the street [10] finding a seat on thebus [11] ordering coffee in a cafe [11] and exiting a buildingduring a fire alarm [12] Overall VR systems provide theinstructor with a balance between flexibility and control Inthe context of a cognitive rehabilitation program VR systemsallow repetitive exercises to be presented in a motivatingengaging and naturally reinforcing way A more thoroughdescription of using the VR-CR approach with children withautism is provided by Wang and Reid [13] The current studyemployed the combined VR-CR approach to address andimprove a specific cognitive impairment in children withautism an impairment in contextual processing of objects

12 Contextual Processing of Objects Contextual processingof objects is defined as the ability to determine an objectrsquosmeaning or relevance in a particular context [14] Objectsare inherently multidimensional each encompasses simpleconcrete qualities such as colour as well as more complexabstract dimensions such as roles and spatial arrangements[14] To determine an objectrsquos meaning or significance ina multiobject context one must take into considerationthe relationships that make a target object relevant withina context as well as adapt flexibly to changing contextsThe three major types of dimensional relationships betweenobjects and their contexts are perceptual (colour shape size)spatial (location) and functional (role or use) [14 15] Therelevant relationships between an object and its context arein large part determined by top-down attentional controlthat is formed from a personrsquos expectations and storedmentalrepresentations of that object [14]

Contextual processing can be deconstructed into twoexecutive functions abstraction the ability to identify thecontext by extracting the relevant qualities and relationshipswithin the environment and cognitive flexibility the abilityto switch between multiple mental representations of a singleobject in response to changing contextual factors

13 Impaired Contextual Processing in Children with AutismCategorization tasks are typically used to assess contextualprocessing These tasks require the child to abstract a partic-ular object dimension by assigning objects to groups based

on internally generated or spontaneous sorting criteria Thecategorization task evaluates cognitive flexibility when thechild is asked to change sorting criteria Changing criteriarequires the child to make a mental switch between multipleobject representations

Ropar and Peebles [16] investigated the spontaneoussorting preferences of children with autism and providedinitial evidence that children with autism have difficultyabstracting higher-level object categories (eg types ofsports games) Klinger and Dawson [17] reported that dur-ing rule-based category learning tasks (the children withautism performed as well as typical controls) Howeverthe performance of the children with autism significantlydecreased during prototype-based category learning tasksAs prototype creation involves abstracting and integratingrelevant information from members of a category theseresults further the notion that children with autism fail tointegrate information at an abstract level

Alternatives to the categorization test include two novelassessments created by Jolliffe and Baron-Cohen [18] toevaluate contextual abstraction in adults with autism Thefirst test is the Object Integration test which requires theparticipant to identify which of five objects is incongruentby establishing a common context between the displayedobjects The commonalities between the objects includeeither spatial or functional relationships The second test isthe Scenic Integration test in which a complex visual scene ispresented where one object is incongruent The participantsare required to identify the incongruent object based onthe established context Performance results on these testsshowed that adults with autism made significantly moreerrors and tookmore time to complete both tests as comparedto typical controlsThe authors concluded that the individualswith autism failed to use context to complete the tests [18]Overall the Object Integration test and Scenic Integrationtest provide alternatives to evaluate abstraction abilities inindividuals with autism

Research has also shown pervasive impairments in cogni-tive flexibility in individualswith autismTheWisconsinCardSorting Task (WCST) [19 20] is a derivative of the simplecategorization task however in addition to sorting cardsbased on particular dimensions (eg colour shape size) thechild must change sorting criterion according to feedbackreceived by the experimenter The degree of perseverationor failure to switch sorting criteria is an effective indicatorof cognitive flexibility A high degree of perseveration on theoriginal and computerizedWCST in children and adults withautism has been well-replicated [21ndash24] Although theWCSTtest does not evaluate cognitive flexibility beyond simpleconcrete object dimensions it does provide evidence thateven at this simple categorization level cognitive flexibilityin the realm of object processing is impaired

In summary the evidence strongly supports the exis-tence of contextual processing impairments in children withautismThese studies demonstrate impairments in the abilityto abstract relevant contextual information and the abilityto flexibly switch mental representations as a function ofchanging contextsThe purpose of the current pilot study wasto evaluate a virtual reality-cognitive rehabilitation (VR-CR)

The Scientific World Journal 3

Table 1 Participant demographic information

Child 1 Child 2 Child 3 Child 4Gender Boy Girl Boy BoyAge 6 years 7 months 8 years 11 months 6 years 1 month 7 years 11 monthsGrade 1 4 1 2Diagnosis Autism Autism Autism PDD-NOSCARS score 325 325 30 33Nonverbal IQ 98 111 139 119

Siblings Brother (9 years 5 months) Sister (6 years 1 month) None Sister (9 years 11months)

Motherrsquos education level Bachelor degree Doctoral degree Bachelor degree Bachelor degreeFatherrsquos education level Postgraduate degree Bachelor degree Bachelor degree Bachelor degreePrimary languagespoken at home English Mandarin Chinese Mandarin Chinese English

Extracurricular activities Weekly 1 1 tutoring foracademic subjects

Weekly swimming classcooking class and

therapeutic horsebackriding

Weekly social skillsgroup for children with

autismNone

Baseline length 3 sessions 4 sessions 5 sessions 6 sessionsAbbreviations CARS Childhood Autism Rating Scale [25] ASD Autism Spectrum Disorder PDD-NOS Pervasive Developmental Disorder not otherwisespecified Note nonverbal IQ scores are derived from the Brief-IQ screener from the Leiter International Performance Scale [26]

intervention to improve contextual processing of objects inchildren with autism

14 Objectives The objectives of the pilot study were(1) to explore the efficacy of a novel VR-CR interven-

tion for improving contextual processing of objectsabstraction and cognitive flexibility in four childrenwith autism

(2) to explore parent perceptions of behavioural changesthat may have occurred during the intervention

2 Material and Methods

21 Experimental Design The current study used a single-subject design with nonconcurrent multiple baselines acrosssubjects Four children were enrolled in the pilot study Eachchild was studied over four to six weeks The study consistedof a baseline phase training phase and a two-week follow-up session with outcomes assessed throughout Prior toenrolment the children were randomized to baseline lengthsof 3 4 5 and 6 sessions

22 Participants Ethical approval for the current study wasobtained from the local hospital and university researchethics boards The inclusion criteria were as follows (1) diag-nosis of an autism spectrum disorder (ASD) by a pediatricneurologist pediatrician clinical psychologist or psychiatrist(copy of childrsquos diagnostic report required) (2) chronologicalage between 5 and 10 years old (3) autism severity classifi-cation of mild to moderate (Childhood Autism Rating Scalescore of 30ndash36) and (4) average or above-average nonverbalIQ (Leiter Brief-IQ score of 85 or greater) A summary ofthe participantsrsquo demographic information is provided in

Table 1 The childrenrsquos names have been changed to protectconfidentiality

23 Setting and Equipment The study was conducted ina quiet room in the childrenrsquos homes The virtual realitytraining programs and virtual reality tests were displayed ona 1510158401015840 Acer TravelMate 8204 laptop computerMotion-capturetechnology was incorporated using a tracking webcam (Log-itech QuickCam Pro 9000) to capture and project the childrsquosimage and movements into the virtual environment Allsoftware programs were programmed using Flash 8 withthe programming language Actionscript 20 The programswere run using Macromedia Flash Player The VR programrequired the children to drag virtual objects to specificlocations on the laptop screen

24 Outcome Measures

241 Virtual Reality Test of Contextual Processing of Objects(VR Test) The VR test is an adaptation of the Object Inte-gration test by Jolliffe and Baron-Cohen [18] which evaluatescontextual processing of objects in children between ages5 and 10 years To complete the task the child is requiredto make a similarity judgment between a movable targetobject and a multiobject context displayed on the screenThe context contains three objects that highlight a particularobject dimension perceptual (eg colour shape) spatial(eg kitchen bathroom) or functional (eg objects used tocut) The purpose of the task is to determine if the targetobject is meaningful in the given context (ie if it shared thesame relevant dimension)

There are two equivalent versions of the VR test Eachversion is composed of 18 test items (object-context pairs)Six pairs are matched based on perceptual relationships six

4 The Scientific World Journal

pairs are matched based on the spatial dimension and sixpairs are matched based on functional characteristics Halfof the pairs are matched correctly and half are matchedincorrectly The 18 test items are randomized differently foreach test version The VR test was administered during eachsession The software program and the researcher indepen-dently record the correct and incorrect responses for eachtest item The child does not receive feedback for his or herresponses on the VR test thus minimizing the occurrence oflearning effects as a result of repeated administration

242 Modified Version of the Flexible Item Selection Task(FIST-m) The Flexible Item Selection Task (FIST) [27] wasused to assess executive functioning On each test item threeobjects are shown (eg red fish blue fish and red telephone)On the first part of the task (Selection 1) the child is asked topoint to two objects that ldquogo togetherrdquoThese two itemsmatchon one relevant dimension (eg common object red fishand blue fish) Then the child is asked to point to a differentpair of objects that ldquogo togetherrdquo (Selection 2) This new pairmatches on a different dimension (eg common colour redfish and red telephone)The common item in both pairs is theldquopivot itemrdquo (eg red fish) Selection 1 involves the ability tointernally abstract a relevant dimension to identify the pairsAlthough Selection 2 also requires abstraction it is a goodmeasure of cognitive flexibility as the child must be able toswitch between different mental representations of the pivotobject

As the original FIST was developed for preschoolers theitems of the FIST-m have been adapted for the older 5-to-10age group Similar to the original the FIST-m comprises 12test items in total and includes items from the original testThe FIST-m was administered at prebaseline posttrainingand follow-up sessions

243 Attention Sustained Subtest [26] The Attention Sus-tained Subtest from the Leiter International PerformanceScale-Revised is a cancellation testwhichmeasures prolongedvisual attention visual scanning and visuomotor inhibitionIt consists of four separate test trials that require the childto colour a target shape or pattern (eg a square) withina complex array of different shapes None of the cognitiveconstructs evaluated by this subtest were explicitly taught inthis study thus it was used as a control test The AttentionSustained Subtest was administered at prebaseline posttrain-ing and follow-up sessions

244 Final Feedback Questionnaire The Final FeedbackQuestionnaire was developed to provide insight into parentalperceptions of behaviour changes in their children Thebehaviours on the form are those associated with con-textual processing and included seven general categoriesof behaviour in particular behaviour in public contextslanguage and communication in social contexts and flexibleuse of objects These categories of behaviours were chosenbased on reported correlations between these behavioursand cognitive impairments [28ndash31] The specific examplesfor each category were taken from the Vineland Adaptive

Behaviour Scale (VABS) [32] The Final Feedback Question-naire was administered once at the end of the study

25 Procedure

251 Baseline Phase Before initial testing the children wereprovided with prebaseline training to ensure that they under-stood the instructions associated with the VR test Using sim-plified pretest items (eg shapes) the childrsquos responses weremodelled prompted and reinforced during this pretrainingtask

252 Training Phase The training program was comprisedof three discrete lessons with one lesson taught per sessionEach lesson focused on one class of object characteristicsperceptual spatial or functional The goal of each lesson wasto teach the child to flexibly attend to object dimensionsof that particular class The lessons were taught in theorder as specified above and mastery on the precedinglesson was required to advance to the next Each trainingsession involved a 10-minute teaching protocol during whichthe same researcher provided each child with one-on-oneinstruction within the virtual environment The trainingsessions were designed to support the childrsquos understandingand performance on the task Verbal instructions modellingprompting repetition and reinforcement were the teachingstrategies that were incorporated

The virtual reality program presented each set of 10training items in a predictable sequence First themultiobjectcontext screen was shown (eg three objects displayed on atable) then the target object appeared The motion-capturevirtual technology allowed the child to see himherself on-screen and to indicate responses through gestures The childwas able to ldquograb and dragrdquo the target object across the screenDragging the target object towards the other items indicateda positive ldquomatchrdquo while dragging the target to the garbagebin indicated a negative ldquononmatchrdquo Visual reinforcementwas built into the training program with correct responsesrewarded with a happy face while incorrect responses werediscouragedwith a sad faceThere was no overlap in the itemsused on the VR test and those used in the training sessions

After completing each lesson the VR test was adminis-tered After each test administration theVR test was analyzedaccording to its separate components perceptual spatial andfunctional items A score of 80 or above was consideredmastery For example if the child scored over 80 on the VRtest perceptual items after Lesson 1 shehe would progress toLesson 2 A child who failed to achieve 80 on the VR testperceptual items after Lesson 1 would repeat that lesson Thetraining phase was ended when the child achieved over 80in all three components of the VR test (see Figure 1)

3 Results

31 VR Test The VR test percentage accuracy scores arepresented for each child across each session (Figure 2) Theaverage baseline scores ranged from 47 (Child 4) to 83(Child 2) Child 2 Child 3 and Child 4 showed marked

The Scientific World Journal 5

Phases

Week

Session

Baseline Training Follow-up

1 2 3 4 5

1 2 3 4 5 6 7 8 9 10 11 12

Figure 1 Overall study design for a hypothetical participant with a 5-session baseline phase 6-session training phase and 2-week follow-upsession

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11 12

Baseline Training Follow-upPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cyPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cy

Child

3

Session number

Child

1Ch

ild 2

Child

4

Figure 2 Percentage accuracy on the VR test demonstrated by each child across all phases of the study

improvements after the first training session which weremaintained throughout the training phase and at follow-up Child 1 showed significant improvements after the thirdtraining session and maintained this at follow-up

The PercentageNonoverlappingData statistic was used toanalyze the data [33] Child 2 Child 3 and Child 4 all demon-strated 100nonoverlapping data Child 1rsquos performance datashowed 60 nonoverlapping data

Overall the results demonstrate improvements in con-textual processing ability from baseline to treatment for

each child with average increases from 15 (Child 2) to46 (Child 4) All children maintained a high level ofperformance at the two-week follow-up assessment

32 FIST-m Percentage accuracy scores were calculated forpretraining and posttraining administrations of the FIST-m Table 2 shows these data for Selection 1 and Selection 2separately

All four children displayed ceiling or close to ceilingscores on Selection 1 (abstraction) scores at pretraining

6 The Scientific World Journal

Table 2 Percentage accuracy scores for Selections 1 and 2 of theFIST-m at pretraining posttraining and follow-up for each child

Selection 1 Selection 2Child 1

Pretraining 92 55Posttraining 92 73Follow-up 75 100

Child 2Pretraining 100 50Posttraining 100 100Follow-up 92 100

Child 3Pretraining 100 83Posttraining 100 92Follow-up 100 92

Child 4Pretraining 100 33Posttraining 92 100Follow-up 100 100

Table 3 Scaled scores for correct responses and error responsesof the attention sustained subtest at pretraining posttraining andfollow-up for each child

Scaled scores ofcorrect responses

Scaled scores of errorresponses

Child 1Pretraining 1 10Posttraining 1 10Follow-up 1 10

Child 2Pretraining 1 12Posttraining 1 12Follow-up 1 12

Child 3Pretraining 10 8Posttraining 16 7Follow-up 16 7

Child 4Pretraining 1 3Posttraining 1 3Follow-up 1 3

These high scores were generally maintained at posttrainingand at follow-up

All children showed improvements on Selection 2 (cog-nitive flexibility) Child 1 made progressive improvements onSelection 2 from 55 at baseline to 73 at posttraining and100 at follow-up Child 2 doubled her baseline score of 50and maintained 100 at follow-up Child 3 scored 83 atbaseline He showed improvement to 92 at posttraining andmaintained this at follow-up Child 4 more than tripled his

baseline score from 33 to 100 and maintained the highscore at follow-up

33 Control Test Attention Sustained Table 3 shows thescaled scores for the childrenrsquos performance on the AttentionSustained subtest Scaled scores for Correct Responses andError Responses are shown separately

Child 1 Child 2 and Child 4 showed no changes inperformance on the scaled scores for both correct responsesand error responses Child 3 showed an increase of 6 pointson scaled correct responses from baseline to posttrainingand a decrease in 1 point in scaled error responses atposttraining

34 Final Feedback Questionnaire The parents of Child 1and Child 4 reported no changes on any of the items onthe Final Feedback Questionnaire Both Child 2rsquos motherand Child 3rsquos mother reported changes in the category ofappropriate language and communication in social contextsChild 2rsquosmother reported thatChild 2 ldquoseems to initiate socialinteraction with more appropriate language (eg Hi Whatrsquosyour name I like your hair style rather than what colouris your nail polish)rdquo Child 3rsquos mother reported that Child3 ldquocan answer question(s) appropriatelyrdquo and that since thebeginning of the study Child 3 ldquois more flexible and like(s) totry something newrdquo

4 Discussion

Overall three of the four children showed 100 nonover-lapping data on the VR-test The fourth child showed 60nonoverlapping data No changes were found on the controltest for three children while the fourth child showed sig-nificant improvements on this test Finally parents of twochildren reported changes in the presence of appropriatesocial behaviours

The primary objective of the study was to explore if thenovel VR-CR intervention could improve contextual process-ing in children with autism According to the standards setby Logan and colleagues [34] the highest level of evidencefor single-subject research designs (Level 1) can be achievedby ldquoconcurrent or non-concurrent multiple baseline designs(MBD) with clear-cut results generalizability if the MBDdesign consists of a minimum of three subjects behavioursor settingsrdquo In determining ldquoclear-cut resultsrdquo Scruggs andMastropieri [33] assert that a treatment outcome with over90 nonoverlapping data can be considered a ldquohighly effec-tive treatmentrdquo The results of this study demonstrate clear100 nonoverlapping data for three children with threedifferent nonconcurrent baseline lengths Thus the currentstudy fulfills the criteria for a Level 1 single-subject designstudy

Changes in abstraction and cognitive flexibility were alsomeasured to verify improvements in contextual processingAs expected initial performance on Selection 2 (cognitiveflexibility) was low Three children performed at chance(50) or below After the training phase all children dis-played substantial improvements andmaintained a high level

The Scientific World Journal 7

of performance at follow-up Both Child 2 and Child 4showed notable improvements on Selection 2 on the FIST-m doubling and tripling their scores respectively Child 3also showed improvements however his performance onSelection 2 at baseline was initially high Although there is nonormative data available for comparison studies performedon the original FIST showed that during a crucial phase ofcognitive flexibility development in typical children (between4 and 5 years) 5-year-olds performed almost 18 better than4-year-olds on Selection 2 [27] The difference reflects therapid development in cognitive flexibility within that one-year span Three children Child 1 Child 2 and Child 4matched or exceeded this spike in development within onlytwo weeks Thus this is emerging evidence that cognitiveflexibility can be improved in this subgroup of children

Although impairments in abstraction have been demon-strated consistently in the literature [16] the children in thecurrent study demonstrated unexpectedly high abstractionperformance (FIST-m Selection 1) at baseline Findings fromRopar and Peebles [16] indicate that autistic children havedifficulty accessing ldquohigh-levelrdquo abstract categories such assports or games Although the current VR-CR interventionrequired the children to abstract qualities above the salientor perceptual the demands of the current task were demon-strated to be within the childrenrsquos range of capabilities

The control test was used to verify that the VR-CRintervention could specifically target contextual processingof objects rather than overall cognitive functioning Onthe Attention Sustained Subtest three of the four childrenshowed no changes over the course of the study supportingthe specificity of the intervention However one child Child3 showed significant improvements on one component of thesubtest from pretraining to posttraining Child 3 improved by6 points on the correct responses score which resulted in a10ndash13 increase in his ranked percentile This change is sig-nificant and may have been influenced by a couple of factorsincluding maturational effects and the level of rapport withinthe child-instructor therapeutic relationship Overall basedon the results of the control test the VR-CR interventiondemonstrated specificity to contextual processing for three offour of the children

The final objective of the pilot study was to explorechanges in context-related behaviours that may haveoccurred during the course of the study The commentsobtained by parents on the Final Feedback Questionnairewere mixed Child 1 and Child 4rsquos parents reported nobehavioural changes that occurred between the start andcompletion of the study Child 2 and Child 3rsquos mothersreported changes in the category of appropriate languageand communication in social contexts However it is notedthat both Child 2 and Child 3 were actively engaged inweekly social activities such as swimming and horsebackriding while Child 1 and Child 4 were not Thus the positivebehavioural changes reported are likely due to factors outsideof the intervention itself

The major limitation of this study is the high baselinelevels achieved by all the children prior to intervention Thissuggests that the participating children were already showinghigh levels of contextual processing ability before the study

particularly in abstraction Assessing performance on theVR-test prior to enrolment would help to exclude childrenfrom the study who are already performing at high levels onthe task In addition determining language levels or verbalintelligence at baseline may help to clarify those childrenwho would benefit maximally from this type of cognitiveintervention

A second limitation of this study is the lack of multipleindependent assessors The degree of therapeutic effective-ness and test performance was likely influenced by the devel-opment of rapport between the researcher and each child Tominimize this effect of researcher bias in the current studythere were no subjective rating scales involved In additionscoring of the VR test was verified through computer records

5 Conclusion

This pilot study evaluated a novel intervention that com-bined virtual reality technology with traditional cognitiverehabilitation methods to address impairments in contextualprocessing of objects in children with autism All childrenwho participated in the VR-CR program demonstratedstatistical improvements in overall contextual processingability and improvements in cognitive flexibility How-ever a larger-scale study will need to evaluate the mag-nitude of change resulting from the intervention and thedegree that cognitive improvements generalize to behaviourchanges

Overall this pilot study provides initial supporting evi-dence for the efficacy of this particular VR-CR interventionContinued research may further generate support for thebroader use of the VR-CR approach with children withautismThe hope is that comprehensive interventions will bedeveloped to target both cognitive and behavioural levels ofimpairment in children with autismThis may lead to greateroverall improvements in the daily functioning and quality oflife of these children

Conflict of Interests

The authors declare no conflict of interests

Acknowledgments

This study has been supported by the following Grantsawarded to the first author Canadian Institutes in HealthResearch 2008-09Masterrsquos Award Hilda andWilliam Court-neyClayton Paediatric ResearchAward 2008-09 andOntarioCouncil of Graduate Studies Autism Scholarrsquos Award 2009-10 This paper was prepared from a Masters thesis by thefirst author The authors give thanks to Bloorview KidsRehab Autism Ontario and the Geneva Centre for helpwith recruitment Special thanks to Fan (Mimi) Zhang fortechnical support and Luigi Girolametto and Tom Chaufor their mentorship and valuable feedback throughout theproject

8 The Scientific World Journal

References

[1] American Psychology Association Diagnostic and StatisticalManual ofMental Disorders DSM-IV-TR American PsychiatricAssociation Washington DC USA 4th edition 2000

[2] C Rice ldquoPrevalence of autism spectrum disordersmdashautism anddevelopmental disabilities monitoring network United States2006rdquo Morbidity and Mortality Weekly Report vol 58 no SS-10 pp 1ndash20 2009

[3] P A Filipek P J Accardo G T Baranek et al ldquoThe screeningand diagnosis of autistic spectrum disordersrdquo Journal of Autismand Developmental Disorders vol 29 pp 439ndash484 1999

[4] A Rizzo and J Buckwater ldquoVirtual reality and cognitiveassessment and rehabilitation the state of the artrdquo in VirtualReality in Neuro-Psycho-Physiology G Riva Ed vol 44 pp123ndash146 IOS Press Washington DC USA 1997

[5] R W Butler ldquoCognitive rehabilitationrdquo in Pediatric Neuropsy-chological Intervention S J Hunter and J Donders EdsCambridge University Press New York NY USA 2007

[6] A A RizzoM Schultheis K A Kerns andCMateer ldquoAnalysisof assets for virtual reality applications in neuropsychologyrdquoNeuropsychological Rehabilitation vol 14 no 1 pp 207ndash2392004

[7] G Burdea and P CoiffetVirtual Reality Technology JohnWileyamp Sons Hoboken NJ USA 2003

[8] BGWitmer C J Jerome andM J Singer ldquoThe factor structureof the presence questionnairerdquo Presence vol 14 no 3 pp 298ndash312 2005

[9] F D Rose E A Attree B M Brooks D M Parslow P RPenn and N Ambihaipahan ldquoTraining in virtual environ-ments transfer to real world tasks and equivalence to real tasktrainingrdquo Ergonomics vol 43 no 4 pp 494ndash511 2000

[10] D Strickland L M Marcus G B Mesibov and K HoganldquoBrief report two case studies using virtual reality as a learningtool for autistic childrenrdquo Journal of Autism and DevelopmentalDisorders vol 26 no 6 pp 651ndash659 1996

[11] P Mitchell S Parsons and A Leonard ldquoUsing virtual environ-ments for teaching social understanding to 6 adolescents withautistic spectrum disordersrdquo Journal of Autism and Develop-mental Disorders vol 37 no 3 pp 589ndash600 2007

[12] T Self R R Scudder G Weheba and D Crumrine ldquoA virtualapproach to teaching safety skills to children with autismspectrum disorderrdquo Topics in Language Disorders vol 27 no 3pp 242ndash253 2007

[13] M Wang and D Reid ldquoThe virtual reality-cognitive rehabili-tation (VR-CR) approach for children with autismrdquo Journal ofCyberTherapy and Rehabilitation vol 2 no 2 pp 95ndash102 2009

[14] M Bar ldquoVisual objects in contextrdquo Nature Reviews Neuro-science vol 5 no 8 pp 617ndash629 2004

[15] A Blaye and F Bonthoux ldquoThematic and taxonomic relationsin preschoolers the development of flexibility in categorizationchoicesrdquoBritish Journal of Developmental Psychology vol 19 no3 pp 395ndash412 2001

[16] D Ropar and D Peebles ldquoSorting preference in children withautism the dominance of concrete featuresrdquo Journal of Autismand Developmental Disorders vol 37 no 2 pp 270ndash280 2007

[17] L G Klinger and G Dawson ldquoPrototype formation in autismrdquoDevelopment and Psychopathology vol 13 no 1 pp 111ndash1242001

[18] T Jolliffe and S Baron-Cohen ldquoA test of central coherencetheory can adults with high-functioning autism or Asperger

syndrome integrate objects in contextrdquo Visual Cognition vol8 no 1 pp 67ndash101 2001

[19] E A Berg ldquoA simple objective technique formeasuring flexibil-ity in thinkingrdquo Journal of General Psychology vol 39 pp 15ndash221948

[20] D A Grant and E Berg ldquoA behavioral analysis of degree ofreinforcement and ease of shifting to new responses in aWeigl-type card-sorting problemrdquo Journal of Experimental Psychologyvol 38 no 4 pp 404ndash411 1948

[21] N Kaland L Smith and E L Mortensen ldquoBrief reportcognitive flexibility and focused attention in children and ado-lescents with Asperger syndrome or high-functioning autism asmeasured on the computerized version of the Wisconsin cardsorting testrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1161ndash1165 2008

[22] N J Minshew J Meyer and G Goldstein ldquoAbstract reasoningin autism a dissociation between concept formation andconcept identificationrdquo Neuropsychology vol 16 no 3 pp 327ndash334 2002

[23] S Ozonoff Specifying the Nature of the Primary Deficit inAutism an Investigation of Neuropsychological and Social-Cognitive Abilities in High-Functioning Autistic Children Uni-versity of Denver Denver Colo USA edition 1991

[24] M Prior and W Hoffmann ldquoBrief report neuropsychologicaltesting of autistic children through an exploration with frontallobe testsrdquo Journal of Autism and Developmental Disorders vol20 no 4 pp 581ndash590 1990

[25] E Schopler R J Reichler and B R Renner The ChildhoodAutism Rating Scale (CARS) Western Psychological ServicesLos Angeles Calif USA 1988

[26] G H Roid and L J Miller Leiter International PerformanceScale Revised Stoelting Wood Dale Ill USA 1997

[27] S Jacques The Roles of Labelling and Abstraction in the Devel-opment of Cognitive Flexibility University of Toronto TorontoCanada 2001

[28] H J C Berger F H T M Aerts K P M van SpaendonckA R Cools and J Teunisse ldquoCentral coherence and cognitiveshifting in relation to social improvement in high-functioningyoung adults with autismrdquo Journal of Clinical and ExperimentalNeuropsychology vol 25 no 4 pp 502ndash511 2003

[29] H J C Berger K P M van Spaendonck M W I MHorstink E L Buytenhuijs P W J M Lammers and A RCools ldquoCognitive shifting as a predictor of progress in socialunderstanding in high-functioning adolescents with autisma prospective studyrdquo Journal of Autism and DevelopmentalDisorders vol 23 no 2 pp 341ndash359 1993

[30] F Happe and U Frith ldquoThe weak coherence account detail-focused cognitive style in autism spectrum disordersrdquo Journalof Autism and Developmental Disorders vol 36 no 1 pp 5ndash252006

[31] B R Lopez A J Lincoln S Ozonoff and Z Lai ldquoExaminingthe relationship between executive functions and restrictedrepetitive symptoms ofAutisticDisorderrdquo Journal of Autism andDevelopmental Disorders vol 35 no 4 pp 445ndash460 2005

[32] S S Sparrow D A Balla andD V CicchettiVineland AdaptiveBehavior Scales Interview Edition Survey form Manual Amer-ican Guidance Service Circle Pines Minn USA 1984

[33] T E Scruggs and M A Mastropieri ldquoSummarizing single-subject research Issues and applicationsrdquo Behavior Modifica-tion vol 22 no 3 pp 221ndash242 1998

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

2 The Scientific World Journal

program or application include interaction and immersionHuman-computer interaction is achieved through multiplesensory channels that allow children to explore virtual envi-ronments through sight sound and touch [7] Immersion isconsidered the degree to which the child feels engrossed orenveloped within the virtual environment [8]

A variety of display devices offer differing degrees ofimmersion and interactivity The current study employs atwo-dimensional flat screen projection system This systemhas motion-capture capabilities where a tracking camera isable to capture and project a childrsquos image and motions on-screen in real-time Being able to see oneself within the virtualenvironmentmay contribute to a greater sense of engagementand motivation during the task

Although VR systems have not yet targeted cognitiveimpairments in childrenwith autism these systemshave beensuccessful in improving specific behaviours such as followingdirections [9] crossing the street [10] finding a seat on thebus [11] ordering coffee in a cafe [11] and exiting a buildingduring a fire alarm [12] Overall VR systems provide theinstructor with a balance between flexibility and control Inthe context of a cognitive rehabilitation program VR systemsallow repetitive exercises to be presented in a motivatingengaging and naturally reinforcing way A more thoroughdescription of using the VR-CR approach with children withautism is provided by Wang and Reid [13] The current studyemployed the combined VR-CR approach to address andimprove a specific cognitive impairment in children withautism an impairment in contextual processing of objects

12 Contextual Processing of Objects Contextual processingof objects is defined as the ability to determine an objectrsquosmeaning or relevance in a particular context [14] Objectsare inherently multidimensional each encompasses simpleconcrete qualities such as colour as well as more complexabstract dimensions such as roles and spatial arrangements[14] To determine an objectrsquos meaning or significance ina multiobject context one must take into considerationthe relationships that make a target object relevant withina context as well as adapt flexibly to changing contextsThe three major types of dimensional relationships betweenobjects and their contexts are perceptual (colour shape size)spatial (location) and functional (role or use) [14 15] Therelevant relationships between an object and its context arein large part determined by top-down attentional controlthat is formed from a personrsquos expectations and storedmentalrepresentations of that object [14]

Contextual processing can be deconstructed into twoexecutive functions abstraction the ability to identify thecontext by extracting the relevant qualities and relationshipswithin the environment and cognitive flexibility the abilityto switch between multiple mental representations of a singleobject in response to changing contextual factors

13 Impaired Contextual Processing in Children with AutismCategorization tasks are typically used to assess contextualprocessing These tasks require the child to abstract a partic-ular object dimension by assigning objects to groups based

on internally generated or spontaneous sorting criteria Thecategorization task evaluates cognitive flexibility when thechild is asked to change sorting criteria Changing criteriarequires the child to make a mental switch between multipleobject representations

Ropar and Peebles [16] investigated the spontaneoussorting preferences of children with autism and providedinitial evidence that children with autism have difficultyabstracting higher-level object categories (eg types ofsports games) Klinger and Dawson [17] reported that dur-ing rule-based category learning tasks (the children withautism performed as well as typical controls) Howeverthe performance of the children with autism significantlydecreased during prototype-based category learning tasksAs prototype creation involves abstracting and integratingrelevant information from members of a category theseresults further the notion that children with autism fail tointegrate information at an abstract level

Alternatives to the categorization test include two novelassessments created by Jolliffe and Baron-Cohen [18] toevaluate contextual abstraction in adults with autism Thefirst test is the Object Integration test which requires theparticipant to identify which of five objects is incongruentby establishing a common context between the displayedobjects The commonalities between the objects includeeither spatial or functional relationships The second test isthe Scenic Integration test in which a complex visual scene ispresented where one object is incongruent The participantsare required to identify the incongruent object based onthe established context Performance results on these testsshowed that adults with autism made significantly moreerrors and tookmore time to complete both tests as comparedto typical controlsThe authors concluded that the individualswith autism failed to use context to complete the tests [18]Overall the Object Integration test and Scenic Integrationtest provide alternatives to evaluate abstraction abilities inindividuals with autism

Research has also shown pervasive impairments in cogni-tive flexibility in individualswith autismTheWisconsinCardSorting Task (WCST) [19 20] is a derivative of the simplecategorization task however in addition to sorting cardsbased on particular dimensions (eg colour shape size) thechild must change sorting criterion according to feedbackreceived by the experimenter The degree of perseverationor failure to switch sorting criteria is an effective indicatorof cognitive flexibility A high degree of perseveration on theoriginal and computerizedWCST in children and adults withautism has been well-replicated [21ndash24] Although theWCSTtest does not evaluate cognitive flexibility beyond simpleconcrete object dimensions it does provide evidence thateven at this simple categorization level cognitive flexibilityin the realm of object processing is impaired

In summary the evidence strongly supports the exis-tence of contextual processing impairments in children withautismThese studies demonstrate impairments in the abilityto abstract relevant contextual information and the abilityto flexibly switch mental representations as a function ofchanging contextsThe purpose of the current pilot study wasto evaluate a virtual reality-cognitive rehabilitation (VR-CR)

The Scientific World Journal 3

Table 1 Participant demographic information

Child 1 Child 2 Child 3 Child 4Gender Boy Girl Boy BoyAge 6 years 7 months 8 years 11 months 6 years 1 month 7 years 11 monthsGrade 1 4 1 2Diagnosis Autism Autism Autism PDD-NOSCARS score 325 325 30 33Nonverbal IQ 98 111 139 119

Siblings Brother (9 years 5 months) Sister (6 years 1 month) None Sister (9 years 11months)

Motherrsquos education level Bachelor degree Doctoral degree Bachelor degree Bachelor degreeFatherrsquos education level Postgraduate degree Bachelor degree Bachelor degree Bachelor degreePrimary languagespoken at home English Mandarin Chinese Mandarin Chinese English

Extracurricular activities Weekly 1 1 tutoring foracademic subjects

Weekly swimming classcooking class and

therapeutic horsebackriding

Weekly social skillsgroup for children with

autismNone

Baseline length 3 sessions 4 sessions 5 sessions 6 sessionsAbbreviations CARS Childhood Autism Rating Scale [25] ASD Autism Spectrum Disorder PDD-NOS Pervasive Developmental Disorder not otherwisespecified Note nonverbal IQ scores are derived from the Brief-IQ screener from the Leiter International Performance Scale [26]

intervention to improve contextual processing of objects inchildren with autism

14 Objectives The objectives of the pilot study were(1) to explore the efficacy of a novel VR-CR interven-

tion for improving contextual processing of objectsabstraction and cognitive flexibility in four childrenwith autism

(2) to explore parent perceptions of behavioural changesthat may have occurred during the intervention

2 Material and Methods

21 Experimental Design The current study used a single-subject design with nonconcurrent multiple baselines acrosssubjects Four children were enrolled in the pilot study Eachchild was studied over four to six weeks The study consistedof a baseline phase training phase and a two-week follow-up session with outcomes assessed throughout Prior toenrolment the children were randomized to baseline lengthsof 3 4 5 and 6 sessions

22 Participants Ethical approval for the current study wasobtained from the local hospital and university researchethics boards The inclusion criteria were as follows (1) diag-nosis of an autism spectrum disorder (ASD) by a pediatricneurologist pediatrician clinical psychologist or psychiatrist(copy of childrsquos diagnostic report required) (2) chronologicalage between 5 and 10 years old (3) autism severity classifi-cation of mild to moderate (Childhood Autism Rating Scalescore of 30ndash36) and (4) average or above-average nonverbalIQ (Leiter Brief-IQ score of 85 or greater) A summary ofthe participantsrsquo demographic information is provided in

Table 1 The childrenrsquos names have been changed to protectconfidentiality

23 Setting and Equipment The study was conducted ina quiet room in the childrenrsquos homes The virtual realitytraining programs and virtual reality tests were displayed ona 1510158401015840 Acer TravelMate 8204 laptop computerMotion-capturetechnology was incorporated using a tracking webcam (Log-itech QuickCam Pro 9000) to capture and project the childrsquosimage and movements into the virtual environment Allsoftware programs were programmed using Flash 8 withthe programming language Actionscript 20 The programswere run using Macromedia Flash Player The VR programrequired the children to drag virtual objects to specificlocations on the laptop screen

24 Outcome Measures

241 Virtual Reality Test of Contextual Processing of Objects(VR Test) The VR test is an adaptation of the Object Inte-gration test by Jolliffe and Baron-Cohen [18] which evaluatescontextual processing of objects in children between ages5 and 10 years To complete the task the child is requiredto make a similarity judgment between a movable targetobject and a multiobject context displayed on the screenThe context contains three objects that highlight a particularobject dimension perceptual (eg colour shape) spatial(eg kitchen bathroom) or functional (eg objects used tocut) The purpose of the task is to determine if the targetobject is meaningful in the given context (ie if it shared thesame relevant dimension)

There are two equivalent versions of the VR test Eachversion is composed of 18 test items (object-context pairs)Six pairs are matched based on perceptual relationships six

4 The Scientific World Journal

pairs are matched based on the spatial dimension and sixpairs are matched based on functional characteristics Halfof the pairs are matched correctly and half are matchedincorrectly The 18 test items are randomized differently foreach test version The VR test was administered during eachsession The software program and the researcher indepen-dently record the correct and incorrect responses for eachtest item The child does not receive feedback for his or herresponses on the VR test thus minimizing the occurrence oflearning effects as a result of repeated administration

242 Modified Version of the Flexible Item Selection Task(FIST-m) The Flexible Item Selection Task (FIST) [27] wasused to assess executive functioning On each test item threeobjects are shown (eg red fish blue fish and red telephone)On the first part of the task (Selection 1) the child is asked topoint to two objects that ldquogo togetherrdquoThese two itemsmatchon one relevant dimension (eg common object red fishand blue fish) Then the child is asked to point to a differentpair of objects that ldquogo togetherrdquo (Selection 2) This new pairmatches on a different dimension (eg common colour redfish and red telephone)The common item in both pairs is theldquopivot itemrdquo (eg red fish) Selection 1 involves the ability tointernally abstract a relevant dimension to identify the pairsAlthough Selection 2 also requires abstraction it is a goodmeasure of cognitive flexibility as the child must be able toswitch between different mental representations of the pivotobject

As the original FIST was developed for preschoolers theitems of the FIST-m have been adapted for the older 5-to-10age group Similar to the original the FIST-m comprises 12test items in total and includes items from the original testThe FIST-m was administered at prebaseline posttrainingand follow-up sessions

243 Attention Sustained Subtest [26] The Attention Sus-tained Subtest from the Leiter International PerformanceScale-Revised is a cancellation testwhichmeasures prolongedvisual attention visual scanning and visuomotor inhibitionIt consists of four separate test trials that require the childto colour a target shape or pattern (eg a square) withina complex array of different shapes None of the cognitiveconstructs evaluated by this subtest were explicitly taught inthis study thus it was used as a control test The AttentionSustained Subtest was administered at prebaseline posttrain-ing and follow-up sessions

244 Final Feedback Questionnaire The Final FeedbackQuestionnaire was developed to provide insight into parentalperceptions of behaviour changes in their children Thebehaviours on the form are those associated with con-textual processing and included seven general categoriesof behaviour in particular behaviour in public contextslanguage and communication in social contexts and flexibleuse of objects These categories of behaviours were chosenbased on reported correlations between these behavioursand cognitive impairments [28ndash31] The specific examplesfor each category were taken from the Vineland Adaptive

Behaviour Scale (VABS) [32] The Final Feedback Question-naire was administered once at the end of the study

25 Procedure

251 Baseline Phase Before initial testing the children wereprovided with prebaseline training to ensure that they under-stood the instructions associated with the VR test Using sim-plified pretest items (eg shapes) the childrsquos responses weremodelled prompted and reinforced during this pretrainingtask

252 Training Phase The training program was comprisedof three discrete lessons with one lesson taught per sessionEach lesson focused on one class of object characteristicsperceptual spatial or functional The goal of each lesson wasto teach the child to flexibly attend to object dimensionsof that particular class The lessons were taught in theorder as specified above and mastery on the precedinglesson was required to advance to the next Each trainingsession involved a 10-minute teaching protocol during whichthe same researcher provided each child with one-on-oneinstruction within the virtual environment The trainingsessions were designed to support the childrsquos understandingand performance on the task Verbal instructions modellingprompting repetition and reinforcement were the teachingstrategies that were incorporated

The virtual reality program presented each set of 10training items in a predictable sequence First themultiobjectcontext screen was shown (eg three objects displayed on atable) then the target object appeared The motion-capturevirtual technology allowed the child to see himherself on-screen and to indicate responses through gestures The childwas able to ldquograb and dragrdquo the target object across the screenDragging the target object towards the other items indicateda positive ldquomatchrdquo while dragging the target to the garbagebin indicated a negative ldquononmatchrdquo Visual reinforcementwas built into the training program with correct responsesrewarded with a happy face while incorrect responses werediscouragedwith a sad faceThere was no overlap in the itemsused on the VR test and those used in the training sessions

After completing each lesson the VR test was adminis-tered After each test administration theVR test was analyzedaccording to its separate components perceptual spatial andfunctional items A score of 80 or above was consideredmastery For example if the child scored over 80 on the VRtest perceptual items after Lesson 1 shehe would progress toLesson 2 A child who failed to achieve 80 on the VR testperceptual items after Lesson 1 would repeat that lesson Thetraining phase was ended when the child achieved over 80in all three components of the VR test (see Figure 1)

3 Results

31 VR Test The VR test percentage accuracy scores arepresented for each child across each session (Figure 2) Theaverage baseline scores ranged from 47 (Child 4) to 83(Child 2) Child 2 Child 3 and Child 4 showed marked

The Scientific World Journal 5

Phases

Week

Session

Baseline Training Follow-up

1 2 3 4 5

1 2 3 4 5 6 7 8 9 10 11 12

Figure 1 Overall study design for a hypothetical participant with a 5-session baseline phase 6-session training phase and 2-week follow-upsession

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11 12

Baseline Training Follow-upPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cyPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cy

Child

3

Session number

Child

1Ch

ild 2

Child

4

Figure 2 Percentage accuracy on the VR test demonstrated by each child across all phases of the study

improvements after the first training session which weremaintained throughout the training phase and at follow-up Child 1 showed significant improvements after the thirdtraining session and maintained this at follow-up

The PercentageNonoverlappingData statistic was used toanalyze the data [33] Child 2 Child 3 and Child 4 all demon-strated 100nonoverlapping data Child 1rsquos performance datashowed 60 nonoverlapping data

Overall the results demonstrate improvements in con-textual processing ability from baseline to treatment for

each child with average increases from 15 (Child 2) to46 (Child 4) All children maintained a high level ofperformance at the two-week follow-up assessment

32 FIST-m Percentage accuracy scores were calculated forpretraining and posttraining administrations of the FIST-m Table 2 shows these data for Selection 1 and Selection 2separately

All four children displayed ceiling or close to ceilingscores on Selection 1 (abstraction) scores at pretraining

6 The Scientific World Journal

Table 2 Percentage accuracy scores for Selections 1 and 2 of theFIST-m at pretraining posttraining and follow-up for each child

Selection 1 Selection 2Child 1

Pretraining 92 55Posttraining 92 73Follow-up 75 100

Child 2Pretraining 100 50Posttraining 100 100Follow-up 92 100

Child 3Pretraining 100 83Posttraining 100 92Follow-up 100 92

Child 4Pretraining 100 33Posttraining 92 100Follow-up 100 100

Table 3 Scaled scores for correct responses and error responsesof the attention sustained subtest at pretraining posttraining andfollow-up for each child

Scaled scores ofcorrect responses

Scaled scores of errorresponses

Child 1Pretraining 1 10Posttraining 1 10Follow-up 1 10

Child 2Pretraining 1 12Posttraining 1 12Follow-up 1 12

Child 3Pretraining 10 8Posttraining 16 7Follow-up 16 7

Child 4Pretraining 1 3Posttraining 1 3Follow-up 1 3

These high scores were generally maintained at posttrainingand at follow-up

All children showed improvements on Selection 2 (cog-nitive flexibility) Child 1 made progressive improvements onSelection 2 from 55 at baseline to 73 at posttraining and100 at follow-up Child 2 doubled her baseline score of 50and maintained 100 at follow-up Child 3 scored 83 atbaseline He showed improvement to 92 at posttraining andmaintained this at follow-up Child 4 more than tripled his

baseline score from 33 to 100 and maintained the highscore at follow-up

33 Control Test Attention Sustained Table 3 shows thescaled scores for the childrenrsquos performance on the AttentionSustained subtest Scaled scores for Correct Responses andError Responses are shown separately

Child 1 Child 2 and Child 4 showed no changes inperformance on the scaled scores for both correct responsesand error responses Child 3 showed an increase of 6 pointson scaled correct responses from baseline to posttrainingand a decrease in 1 point in scaled error responses atposttraining

34 Final Feedback Questionnaire The parents of Child 1and Child 4 reported no changes on any of the items onthe Final Feedback Questionnaire Both Child 2rsquos motherand Child 3rsquos mother reported changes in the category ofappropriate language and communication in social contextsChild 2rsquosmother reported thatChild 2 ldquoseems to initiate socialinteraction with more appropriate language (eg Hi Whatrsquosyour name I like your hair style rather than what colouris your nail polish)rdquo Child 3rsquos mother reported that Child3 ldquocan answer question(s) appropriatelyrdquo and that since thebeginning of the study Child 3 ldquois more flexible and like(s) totry something newrdquo

4 Discussion

Overall three of the four children showed 100 nonover-lapping data on the VR-test The fourth child showed 60nonoverlapping data No changes were found on the controltest for three children while the fourth child showed sig-nificant improvements on this test Finally parents of twochildren reported changes in the presence of appropriatesocial behaviours

The primary objective of the study was to explore if thenovel VR-CR intervention could improve contextual process-ing in children with autism According to the standards setby Logan and colleagues [34] the highest level of evidencefor single-subject research designs (Level 1) can be achievedby ldquoconcurrent or non-concurrent multiple baseline designs(MBD) with clear-cut results generalizability if the MBDdesign consists of a minimum of three subjects behavioursor settingsrdquo In determining ldquoclear-cut resultsrdquo Scruggs andMastropieri [33] assert that a treatment outcome with over90 nonoverlapping data can be considered a ldquohighly effec-tive treatmentrdquo The results of this study demonstrate clear100 nonoverlapping data for three children with threedifferent nonconcurrent baseline lengths Thus the currentstudy fulfills the criteria for a Level 1 single-subject designstudy

Changes in abstraction and cognitive flexibility were alsomeasured to verify improvements in contextual processingAs expected initial performance on Selection 2 (cognitiveflexibility) was low Three children performed at chance(50) or below After the training phase all children dis-played substantial improvements andmaintained a high level

The Scientific World Journal 7

of performance at follow-up Both Child 2 and Child 4showed notable improvements on Selection 2 on the FIST-m doubling and tripling their scores respectively Child 3also showed improvements however his performance onSelection 2 at baseline was initially high Although there is nonormative data available for comparison studies performedon the original FIST showed that during a crucial phase ofcognitive flexibility development in typical children (between4 and 5 years) 5-year-olds performed almost 18 better than4-year-olds on Selection 2 [27] The difference reflects therapid development in cognitive flexibility within that one-year span Three children Child 1 Child 2 and Child 4matched or exceeded this spike in development within onlytwo weeks Thus this is emerging evidence that cognitiveflexibility can be improved in this subgroup of children

Although impairments in abstraction have been demon-strated consistently in the literature [16] the children in thecurrent study demonstrated unexpectedly high abstractionperformance (FIST-m Selection 1) at baseline Findings fromRopar and Peebles [16] indicate that autistic children havedifficulty accessing ldquohigh-levelrdquo abstract categories such assports or games Although the current VR-CR interventionrequired the children to abstract qualities above the salientor perceptual the demands of the current task were demon-strated to be within the childrenrsquos range of capabilities

The control test was used to verify that the VR-CRintervention could specifically target contextual processingof objects rather than overall cognitive functioning Onthe Attention Sustained Subtest three of the four childrenshowed no changes over the course of the study supportingthe specificity of the intervention However one child Child3 showed significant improvements on one component of thesubtest from pretraining to posttraining Child 3 improved by6 points on the correct responses score which resulted in a10ndash13 increase in his ranked percentile This change is sig-nificant and may have been influenced by a couple of factorsincluding maturational effects and the level of rapport withinthe child-instructor therapeutic relationship Overall basedon the results of the control test the VR-CR interventiondemonstrated specificity to contextual processing for three offour of the children

The final objective of the pilot study was to explorechanges in context-related behaviours that may haveoccurred during the course of the study The commentsobtained by parents on the Final Feedback Questionnairewere mixed Child 1 and Child 4rsquos parents reported nobehavioural changes that occurred between the start andcompletion of the study Child 2 and Child 3rsquos mothersreported changes in the category of appropriate languageand communication in social contexts However it is notedthat both Child 2 and Child 3 were actively engaged inweekly social activities such as swimming and horsebackriding while Child 1 and Child 4 were not Thus the positivebehavioural changes reported are likely due to factors outsideof the intervention itself

The major limitation of this study is the high baselinelevels achieved by all the children prior to intervention Thissuggests that the participating children were already showinghigh levels of contextual processing ability before the study

particularly in abstraction Assessing performance on theVR-test prior to enrolment would help to exclude childrenfrom the study who are already performing at high levels onthe task In addition determining language levels or verbalintelligence at baseline may help to clarify those childrenwho would benefit maximally from this type of cognitiveintervention

A second limitation of this study is the lack of multipleindependent assessors The degree of therapeutic effective-ness and test performance was likely influenced by the devel-opment of rapport between the researcher and each child Tominimize this effect of researcher bias in the current studythere were no subjective rating scales involved In additionscoring of the VR test was verified through computer records

5 Conclusion

This pilot study evaluated a novel intervention that com-bined virtual reality technology with traditional cognitiverehabilitation methods to address impairments in contextualprocessing of objects in children with autism All childrenwho participated in the VR-CR program demonstratedstatistical improvements in overall contextual processingability and improvements in cognitive flexibility How-ever a larger-scale study will need to evaluate the mag-nitude of change resulting from the intervention and thedegree that cognitive improvements generalize to behaviourchanges

Overall this pilot study provides initial supporting evi-dence for the efficacy of this particular VR-CR interventionContinued research may further generate support for thebroader use of the VR-CR approach with children withautismThe hope is that comprehensive interventions will bedeveloped to target both cognitive and behavioural levels ofimpairment in children with autismThis may lead to greateroverall improvements in the daily functioning and quality oflife of these children

Conflict of Interests

The authors declare no conflict of interests

Acknowledgments

This study has been supported by the following Grantsawarded to the first author Canadian Institutes in HealthResearch 2008-09Masterrsquos Award Hilda andWilliam Court-neyClayton Paediatric ResearchAward 2008-09 andOntarioCouncil of Graduate Studies Autism Scholarrsquos Award 2009-10 This paper was prepared from a Masters thesis by thefirst author The authors give thanks to Bloorview KidsRehab Autism Ontario and the Geneva Centre for helpwith recruitment Special thanks to Fan (Mimi) Zhang fortechnical support and Luigi Girolametto and Tom Chaufor their mentorship and valuable feedback throughout theproject

8 The Scientific World Journal

References

[1] American Psychology Association Diagnostic and StatisticalManual ofMental Disorders DSM-IV-TR American PsychiatricAssociation Washington DC USA 4th edition 2000

[2] C Rice ldquoPrevalence of autism spectrum disordersmdashautism anddevelopmental disabilities monitoring network United States2006rdquo Morbidity and Mortality Weekly Report vol 58 no SS-10 pp 1ndash20 2009

[3] P A Filipek P J Accardo G T Baranek et al ldquoThe screeningand diagnosis of autistic spectrum disordersrdquo Journal of Autismand Developmental Disorders vol 29 pp 439ndash484 1999

[4] A Rizzo and J Buckwater ldquoVirtual reality and cognitiveassessment and rehabilitation the state of the artrdquo in VirtualReality in Neuro-Psycho-Physiology G Riva Ed vol 44 pp123ndash146 IOS Press Washington DC USA 1997

[5] R W Butler ldquoCognitive rehabilitationrdquo in Pediatric Neuropsy-chological Intervention S J Hunter and J Donders EdsCambridge University Press New York NY USA 2007

[6] A A RizzoM Schultheis K A Kerns andCMateer ldquoAnalysisof assets for virtual reality applications in neuropsychologyrdquoNeuropsychological Rehabilitation vol 14 no 1 pp 207ndash2392004

[7] G Burdea and P CoiffetVirtual Reality Technology JohnWileyamp Sons Hoboken NJ USA 2003

[8] BGWitmer C J Jerome andM J Singer ldquoThe factor structureof the presence questionnairerdquo Presence vol 14 no 3 pp 298ndash312 2005

[9] F D Rose E A Attree B M Brooks D M Parslow P RPenn and N Ambihaipahan ldquoTraining in virtual environ-ments transfer to real world tasks and equivalence to real tasktrainingrdquo Ergonomics vol 43 no 4 pp 494ndash511 2000

[10] D Strickland L M Marcus G B Mesibov and K HoganldquoBrief report two case studies using virtual reality as a learningtool for autistic childrenrdquo Journal of Autism and DevelopmentalDisorders vol 26 no 6 pp 651ndash659 1996

[11] P Mitchell S Parsons and A Leonard ldquoUsing virtual environ-ments for teaching social understanding to 6 adolescents withautistic spectrum disordersrdquo Journal of Autism and Develop-mental Disorders vol 37 no 3 pp 589ndash600 2007

[12] T Self R R Scudder G Weheba and D Crumrine ldquoA virtualapproach to teaching safety skills to children with autismspectrum disorderrdquo Topics in Language Disorders vol 27 no 3pp 242ndash253 2007

[13] M Wang and D Reid ldquoThe virtual reality-cognitive rehabili-tation (VR-CR) approach for children with autismrdquo Journal ofCyberTherapy and Rehabilitation vol 2 no 2 pp 95ndash102 2009

[14] M Bar ldquoVisual objects in contextrdquo Nature Reviews Neuro-science vol 5 no 8 pp 617ndash629 2004

[15] A Blaye and F Bonthoux ldquoThematic and taxonomic relationsin preschoolers the development of flexibility in categorizationchoicesrdquoBritish Journal of Developmental Psychology vol 19 no3 pp 395ndash412 2001

[16] D Ropar and D Peebles ldquoSorting preference in children withautism the dominance of concrete featuresrdquo Journal of Autismand Developmental Disorders vol 37 no 2 pp 270ndash280 2007

[17] L G Klinger and G Dawson ldquoPrototype formation in autismrdquoDevelopment and Psychopathology vol 13 no 1 pp 111ndash1242001

[18] T Jolliffe and S Baron-Cohen ldquoA test of central coherencetheory can adults with high-functioning autism or Asperger

syndrome integrate objects in contextrdquo Visual Cognition vol8 no 1 pp 67ndash101 2001

[19] E A Berg ldquoA simple objective technique formeasuring flexibil-ity in thinkingrdquo Journal of General Psychology vol 39 pp 15ndash221948

[20] D A Grant and E Berg ldquoA behavioral analysis of degree ofreinforcement and ease of shifting to new responses in aWeigl-type card-sorting problemrdquo Journal of Experimental Psychologyvol 38 no 4 pp 404ndash411 1948

[21] N Kaland L Smith and E L Mortensen ldquoBrief reportcognitive flexibility and focused attention in children and ado-lescents with Asperger syndrome or high-functioning autism asmeasured on the computerized version of the Wisconsin cardsorting testrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1161ndash1165 2008

[22] N J Minshew J Meyer and G Goldstein ldquoAbstract reasoningin autism a dissociation between concept formation andconcept identificationrdquo Neuropsychology vol 16 no 3 pp 327ndash334 2002

[23] S Ozonoff Specifying the Nature of the Primary Deficit inAutism an Investigation of Neuropsychological and Social-Cognitive Abilities in High-Functioning Autistic Children Uni-versity of Denver Denver Colo USA edition 1991

[24] M Prior and W Hoffmann ldquoBrief report neuropsychologicaltesting of autistic children through an exploration with frontallobe testsrdquo Journal of Autism and Developmental Disorders vol20 no 4 pp 581ndash590 1990

[25] E Schopler R J Reichler and B R Renner The ChildhoodAutism Rating Scale (CARS) Western Psychological ServicesLos Angeles Calif USA 1988

[26] G H Roid and L J Miller Leiter International PerformanceScale Revised Stoelting Wood Dale Ill USA 1997

[27] S Jacques The Roles of Labelling and Abstraction in the Devel-opment of Cognitive Flexibility University of Toronto TorontoCanada 2001

[28] H J C Berger F H T M Aerts K P M van SpaendonckA R Cools and J Teunisse ldquoCentral coherence and cognitiveshifting in relation to social improvement in high-functioningyoung adults with autismrdquo Journal of Clinical and ExperimentalNeuropsychology vol 25 no 4 pp 502ndash511 2003

[29] H J C Berger K P M van Spaendonck M W I MHorstink E L Buytenhuijs P W J M Lammers and A RCools ldquoCognitive shifting as a predictor of progress in socialunderstanding in high-functioning adolescents with autisma prospective studyrdquo Journal of Autism and DevelopmentalDisorders vol 23 no 2 pp 341ndash359 1993

[30] F Happe and U Frith ldquoThe weak coherence account detail-focused cognitive style in autism spectrum disordersrdquo Journalof Autism and Developmental Disorders vol 36 no 1 pp 5ndash252006

[31] B R Lopez A J Lincoln S Ozonoff and Z Lai ldquoExaminingthe relationship between executive functions and restrictedrepetitive symptoms ofAutisticDisorderrdquo Journal of Autism andDevelopmental Disorders vol 35 no 4 pp 445ndash460 2005

[32] S S Sparrow D A Balla andD V CicchettiVineland AdaptiveBehavior Scales Interview Edition Survey form Manual Amer-ican Guidance Service Circle Pines Minn USA 1984

[33] T E Scruggs and M A Mastropieri ldquoSummarizing single-subject research Issues and applicationsrdquo Behavior Modifica-tion vol 22 no 3 pp 221ndash242 1998

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

The Scientific World Journal 3

Table 1 Participant demographic information

Child 1 Child 2 Child 3 Child 4Gender Boy Girl Boy BoyAge 6 years 7 months 8 years 11 months 6 years 1 month 7 years 11 monthsGrade 1 4 1 2Diagnosis Autism Autism Autism PDD-NOSCARS score 325 325 30 33Nonverbal IQ 98 111 139 119

Siblings Brother (9 years 5 months) Sister (6 years 1 month) None Sister (9 years 11months)

Motherrsquos education level Bachelor degree Doctoral degree Bachelor degree Bachelor degreeFatherrsquos education level Postgraduate degree Bachelor degree Bachelor degree Bachelor degreePrimary languagespoken at home English Mandarin Chinese Mandarin Chinese English

Extracurricular activities Weekly 1 1 tutoring foracademic subjects

Weekly swimming classcooking class and

therapeutic horsebackriding

Weekly social skillsgroup for children with

autismNone

Baseline length 3 sessions 4 sessions 5 sessions 6 sessionsAbbreviations CARS Childhood Autism Rating Scale [25] ASD Autism Spectrum Disorder PDD-NOS Pervasive Developmental Disorder not otherwisespecified Note nonverbal IQ scores are derived from the Brief-IQ screener from the Leiter International Performance Scale [26]

intervention to improve contextual processing of objects inchildren with autism

14 Objectives The objectives of the pilot study were(1) to explore the efficacy of a novel VR-CR interven-

tion for improving contextual processing of objectsabstraction and cognitive flexibility in four childrenwith autism

(2) to explore parent perceptions of behavioural changesthat may have occurred during the intervention

2 Material and Methods

21 Experimental Design The current study used a single-subject design with nonconcurrent multiple baselines acrosssubjects Four children were enrolled in the pilot study Eachchild was studied over four to six weeks The study consistedof a baseline phase training phase and a two-week follow-up session with outcomes assessed throughout Prior toenrolment the children were randomized to baseline lengthsof 3 4 5 and 6 sessions

22 Participants Ethical approval for the current study wasobtained from the local hospital and university researchethics boards The inclusion criteria were as follows (1) diag-nosis of an autism spectrum disorder (ASD) by a pediatricneurologist pediatrician clinical psychologist or psychiatrist(copy of childrsquos diagnostic report required) (2) chronologicalage between 5 and 10 years old (3) autism severity classifi-cation of mild to moderate (Childhood Autism Rating Scalescore of 30ndash36) and (4) average or above-average nonverbalIQ (Leiter Brief-IQ score of 85 or greater) A summary ofthe participantsrsquo demographic information is provided in

Table 1 The childrenrsquos names have been changed to protectconfidentiality

23 Setting and Equipment The study was conducted ina quiet room in the childrenrsquos homes The virtual realitytraining programs and virtual reality tests were displayed ona 1510158401015840 Acer TravelMate 8204 laptop computerMotion-capturetechnology was incorporated using a tracking webcam (Log-itech QuickCam Pro 9000) to capture and project the childrsquosimage and movements into the virtual environment Allsoftware programs were programmed using Flash 8 withthe programming language Actionscript 20 The programswere run using Macromedia Flash Player The VR programrequired the children to drag virtual objects to specificlocations on the laptop screen

24 Outcome Measures

241 Virtual Reality Test of Contextual Processing of Objects(VR Test) The VR test is an adaptation of the Object Inte-gration test by Jolliffe and Baron-Cohen [18] which evaluatescontextual processing of objects in children between ages5 and 10 years To complete the task the child is requiredto make a similarity judgment between a movable targetobject and a multiobject context displayed on the screenThe context contains three objects that highlight a particularobject dimension perceptual (eg colour shape) spatial(eg kitchen bathroom) or functional (eg objects used tocut) The purpose of the task is to determine if the targetobject is meaningful in the given context (ie if it shared thesame relevant dimension)

There are two equivalent versions of the VR test Eachversion is composed of 18 test items (object-context pairs)Six pairs are matched based on perceptual relationships six

4 The Scientific World Journal

pairs are matched based on the spatial dimension and sixpairs are matched based on functional characteristics Halfof the pairs are matched correctly and half are matchedincorrectly The 18 test items are randomized differently foreach test version The VR test was administered during eachsession The software program and the researcher indepen-dently record the correct and incorrect responses for eachtest item The child does not receive feedback for his or herresponses on the VR test thus minimizing the occurrence oflearning effects as a result of repeated administration

242 Modified Version of the Flexible Item Selection Task(FIST-m) The Flexible Item Selection Task (FIST) [27] wasused to assess executive functioning On each test item threeobjects are shown (eg red fish blue fish and red telephone)On the first part of the task (Selection 1) the child is asked topoint to two objects that ldquogo togetherrdquoThese two itemsmatchon one relevant dimension (eg common object red fishand blue fish) Then the child is asked to point to a differentpair of objects that ldquogo togetherrdquo (Selection 2) This new pairmatches on a different dimension (eg common colour redfish and red telephone)The common item in both pairs is theldquopivot itemrdquo (eg red fish) Selection 1 involves the ability tointernally abstract a relevant dimension to identify the pairsAlthough Selection 2 also requires abstraction it is a goodmeasure of cognitive flexibility as the child must be able toswitch between different mental representations of the pivotobject

As the original FIST was developed for preschoolers theitems of the FIST-m have been adapted for the older 5-to-10age group Similar to the original the FIST-m comprises 12test items in total and includes items from the original testThe FIST-m was administered at prebaseline posttrainingand follow-up sessions

243 Attention Sustained Subtest [26] The Attention Sus-tained Subtest from the Leiter International PerformanceScale-Revised is a cancellation testwhichmeasures prolongedvisual attention visual scanning and visuomotor inhibitionIt consists of four separate test trials that require the childto colour a target shape or pattern (eg a square) withina complex array of different shapes None of the cognitiveconstructs evaluated by this subtest were explicitly taught inthis study thus it was used as a control test The AttentionSustained Subtest was administered at prebaseline posttrain-ing and follow-up sessions

244 Final Feedback Questionnaire The Final FeedbackQuestionnaire was developed to provide insight into parentalperceptions of behaviour changes in their children Thebehaviours on the form are those associated with con-textual processing and included seven general categoriesof behaviour in particular behaviour in public contextslanguage and communication in social contexts and flexibleuse of objects These categories of behaviours were chosenbased on reported correlations between these behavioursand cognitive impairments [28ndash31] The specific examplesfor each category were taken from the Vineland Adaptive

Behaviour Scale (VABS) [32] The Final Feedback Question-naire was administered once at the end of the study

25 Procedure

251 Baseline Phase Before initial testing the children wereprovided with prebaseline training to ensure that they under-stood the instructions associated with the VR test Using sim-plified pretest items (eg shapes) the childrsquos responses weremodelled prompted and reinforced during this pretrainingtask

252 Training Phase The training program was comprisedof three discrete lessons with one lesson taught per sessionEach lesson focused on one class of object characteristicsperceptual spatial or functional The goal of each lesson wasto teach the child to flexibly attend to object dimensionsof that particular class The lessons were taught in theorder as specified above and mastery on the precedinglesson was required to advance to the next Each trainingsession involved a 10-minute teaching protocol during whichthe same researcher provided each child with one-on-oneinstruction within the virtual environment The trainingsessions were designed to support the childrsquos understandingand performance on the task Verbal instructions modellingprompting repetition and reinforcement were the teachingstrategies that were incorporated

The virtual reality program presented each set of 10training items in a predictable sequence First themultiobjectcontext screen was shown (eg three objects displayed on atable) then the target object appeared The motion-capturevirtual technology allowed the child to see himherself on-screen and to indicate responses through gestures The childwas able to ldquograb and dragrdquo the target object across the screenDragging the target object towards the other items indicateda positive ldquomatchrdquo while dragging the target to the garbagebin indicated a negative ldquononmatchrdquo Visual reinforcementwas built into the training program with correct responsesrewarded with a happy face while incorrect responses werediscouragedwith a sad faceThere was no overlap in the itemsused on the VR test and those used in the training sessions

After completing each lesson the VR test was adminis-tered After each test administration theVR test was analyzedaccording to its separate components perceptual spatial andfunctional items A score of 80 or above was consideredmastery For example if the child scored over 80 on the VRtest perceptual items after Lesson 1 shehe would progress toLesson 2 A child who failed to achieve 80 on the VR testperceptual items after Lesson 1 would repeat that lesson Thetraining phase was ended when the child achieved over 80in all three components of the VR test (see Figure 1)

3 Results

31 VR Test The VR test percentage accuracy scores arepresented for each child across each session (Figure 2) Theaverage baseline scores ranged from 47 (Child 4) to 83(Child 2) Child 2 Child 3 and Child 4 showed marked

The Scientific World Journal 5

Phases

Week

Session

Baseline Training Follow-up

1 2 3 4 5

1 2 3 4 5 6 7 8 9 10 11 12

Figure 1 Overall study design for a hypothetical participant with a 5-session baseline phase 6-session training phase and 2-week follow-upsession

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11 12

Baseline Training Follow-upPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cyPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cy

Child

3

Session number

Child

1Ch

ild 2

Child

4

Figure 2 Percentage accuracy on the VR test demonstrated by each child across all phases of the study

improvements after the first training session which weremaintained throughout the training phase and at follow-up Child 1 showed significant improvements after the thirdtraining session and maintained this at follow-up

The PercentageNonoverlappingData statistic was used toanalyze the data [33] Child 2 Child 3 and Child 4 all demon-strated 100nonoverlapping data Child 1rsquos performance datashowed 60 nonoverlapping data

Overall the results demonstrate improvements in con-textual processing ability from baseline to treatment for

each child with average increases from 15 (Child 2) to46 (Child 4) All children maintained a high level ofperformance at the two-week follow-up assessment

32 FIST-m Percentage accuracy scores were calculated forpretraining and posttraining administrations of the FIST-m Table 2 shows these data for Selection 1 and Selection 2separately

All four children displayed ceiling or close to ceilingscores on Selection 1 (abstraction) scores at pretraining

6 The Scientific World Journal

Table 2 Percentage accuracy scores for Selections 1 and 2 of theFIST-m at pretraining posttraining and follow-up for each child

Selection 1 Selection 2Child 1

Pretraining 92 55Posttraining 92 73Follow-up 75 100

Child 2Pretraining 100 50Posttraining 100 100Follow-up 92 100

Child 3Pretraining 100 83Posttraining 100 92Follow-up 100 92

Child 4Pretraining 100 33Posttraining 92 100Follow-up 100 100

Table 3 Scaled scores for correct responses and error responsesof the attention sustained subtest at pretraining posttraining andfollow-up for each child

Scaled scores ofcorrect responses

Scaled scores of errorresponses

Child 1Pretraining 1 10Posttraining 1 10Follow-up 1 10

Child 2Pretraining 1 12Posttraining 1 12Follow-up 1 12

Child 3Pretraining 10 8Posttraining 16 7Follow-up 16 7

Child 4Pretraining 1 3Posttraining 1 3Follow-up 1 3

These high scores were generally maintained at posttrainingand at follow-up

All children showed improvements on Selection 2 (cog-nitive flexibility) Child 1 made progressive improvements onSelection 2 from 55 at baseline to 73 at posttraining and100 at follow-up Child 2 doubled her baseline score of 50and maintained 100 at follow-up Child 3 scored 83 atbaseline He showed improvement to 92 at posttraining andmaintained this at follow-up Child 4 more than tripled his

baseline score from 33 to 100 and maintained the highscore at follow-up

33 Control Test Attention Sustained Table 3 shows thescaled scores for the childrenrsquos performance on the AttentionSustained subtest Scaled scores for Correct Responses andError Responses are shown separately

Child 1 Child 2 and Child 4 showed no changes inperformance on the scaled scores for both correct responsesand error responses Child 3 showed an increase of 6 pointson scaled correct responses from baseline to posttrainingand a decrease in 1 point in scaled error responses atposttraining

34 Final Feedback Questionnaire The parents of Child 1and Child 4 reported no changes on any of the items onthe Final Feedback Questionnaire Both Child 2rsquos motherand Child 3rsquos mother reported changes in the category ofappropriate language and communication in social contextsChild 2rsquosmother reported thatChild 2 ldquoseems to initiate socialinteraction with more appropriate language (eg Hi Whatrsquosyour name I like your hair style rather than what colouris your nail polish)rdquo Child 3rsquos mother reported that Child3 ldquocan answer question(s) appropriatelyrdquo and that since thebeginning of the study Child 3 ldquois more flexible and like(s) totry something newrdquo

4 Discussion

Overall three of the four children showed 100 nonover-lapping data on the VR-test The fourth child showed 60nonoverlapping data No changes were found on the controltest for three children while the fourth child showed sig-nificant improvements on this test Finally parents of twochildren reported changes in the presence of appropriatesocial behaviours

The primary objective of the study was to explore if thenovel VR-CR intervention could improve contextual process-ing in children with autism According to the standards setby Logan and colleagues [34] the highest level of evidencefor single-subject research designs (Level 1) can be achievedby ldquoconcurrent or non-concurrent multiple baseline designs(MBD) with clear-cut results generalizability if the MBDdesign consists of a minimum of three subjects behavioursor settingsrdquo In determining ldquoclear-cut resultsrdquo Scruggs andMastropieri [33] assert that a treatment outcome with over90 nonoverlapping data can be considered a ldquohighly effec-tive treatmentrdquo The results of this study demonstrate clear100 nonoverlapping data for three children with threedifferent nonconcurrent baseline lengths Thus the currentstudy fulfills the criteria for a Level 1 single-subject designstudy

Changes in abstraction and cognitive flexibility were alsomeasured to verify improvements in contextual processingAs expected initial performance on Selection 2 (cognitiveflexibility) was low Three children performed at chance(50) or below After the training phase all children dis-played substantial improvements andmaintained a high level

The Scientific World Journal 7

of performance at follow-up Both Child 2 and Child 4showed notable improvements on Selection 2 on the FIST-m doubling and tripling their scores respectively Child 3also showed improvements however his performance onSelection 2 at baseline was initially high Although there is nonormative data available for comparison studies performedon the original FIST showed that during a crucial phase ofcognitive flexibility development in typical children (between4 and 5 years) 5-year-olds performed almost 18 better than4-year-olds on Selection 2 [27] The difference reflects therapid development in cognitive flexibility within that one-year span Three children Child 1 Child 2 and Child 4matched or exceeded this spike in development within onlytwo weeks Thus this is emerging evidence that cognitiveflexibility can be improved in this subgroup of children

Although impairments in abstraction have been demon-strated consistently in the literature [16] the children in thecurrent study demonstrated unexpectedly high abstractionperformance (FIST-m Selection 1) at baseline Findings fromRopar and Peebles [16] indicate that autistic children havedifficulty accessing ldquohigh-levelrdquo abstract categories such assports or games Although the current VR-CR interventionrequired the children to abstract qualities above the salientor perceptual the demands of the current task were demon-strated to be within the childrenrsquos range of capabilities

The control test was used to verify that the VR-CRintervention could specifically target contextual processingof objects rather than overall cognitive functioning Onthe Attention Sustained Subtest three of the four childrenshowed no changes over the course of the study supportingthe specificity of the intervention However one child Child3 showed significant improvements on one component of thesubtest from pretraining to posttraining Child 3 improved by6 points on the correct responses score which resulted in a10ndash13 increase in his ranked percentile This change is sig-nificant and may have been influenced by a couple of factorsincluding maturational effects and the level of rapport withinthe child-instructor therapeutic relationship Overall basedon the results of the control test the VR-CR interventiondemonstrated specificity to contextual processing for three offour of the children

The final objective of the pilot study was to explorechanges in context-related behaviours that may haveoccurred during the course of the study The commentsobtained by parents on the Final Feedback Questionnairewere mixed Child 1 and Child 4rsquos parents reported nobehavioural changes that occurred between the start andcompletion of the study Child 2 and Child 3rsquos mothersreported changes in the category of appropriate languageand communication in social contexts However it is notedthat both Child 2 and Child 3 were actively engaged inweekly social activities such as swimming and horsebackriding while Child 1 and Child 4 were not Thus the positivebehavioural changes reported are likely due to factors outsideof the intervention itself

The major limitation of this study is the high baselinelevels achieved by all the children prior to intervention Thissuggests that the participating children were already showinghigh levels of contextual processing ability before the study

particularly in abstraction Assessing performance on theVR-test prior to enrolment would help to exclude childrenfrom the study who are already performing at high levels onthe task In addition determining language levels or verbalintelligence at baseline may help to clarify those childrenwho would benefit maximally from this type of cognitiveintervention

A second limitation of this study is the lack of multipleindependent assessors The degree of therapeutic effective-ness and test performance was likely influenced by the devel-opment of rapport between the researcher and each child Tominimize this effect of researcher bias in the current studythere were no subjective rating scales involved In additionscoring of the VR test was verified through computer records

5 Conclusion

This pilot study evaluated a novel intervention that com-bined virtual reality technology with traditional cognitiverehabilitation methods to address impairments in contextualprocessing of objects in children with autism All childrenwho participated in the VR-CR program demonstratedstatistical improvements in overall contextual processingability and improvements in cognitive flexibility How-ever a larger-scale study will need to evaluate the mag-nitude of change resulting from the intervention and thedegree that cognitive improvements generalize to behaviourchanges

Overall this pilot study provides initial supporting evi-dence for the efficacy of this particular VR-CR interventionContinued research may further generate support for thebroader use of the VR-CR approach with children withautismThe hope is that comprehensive interventions will bedeveloped to target both cognitive and behavioural levels ofimpairment in children with autismThis may lead to greateroverall improvements in the daily functioning and quality oflife of these children

Conflict of Interests

The authors declare no conflict of interests

Acknowledgments

This study has been supported by the following Grantsawarded to the first author Canadian Institutes in HealthResearch 2008-09Masterrsquos Award Hilda andWilliam Court-neyClayton Paediatric ResearchAward 2008-09 andOntarioCouncil of Graduate Studies Autism Scholarrsquos Award 2009-10 This paper was prepared from a Masters thesis by thefirst author The authors give thanks to Bloorview KidsRehab Autism Ontario and the Geneva Centre for helpwith recruitment Special thanks to Fan (Mimi) Zhang fortechnical support and Luigi Girolametto and Tom Chaufor their mentorship and valuable feedback throughout theproject

8 The Scientific World Journal

References

[1] American Psychology Association Diagnostic and StatisticalManual ofMental Disorders DSM-IV-TR American PsychiatricAssociation Washington DC USA 4th edition 2000

[2] C Rice ldquoPrevalence of autism spectrum disordersmdashautism anddevelopmental disabilities monitoring network United States2006rdquo Morbidity and Mortality Weekly Report vol 58 no SS-10 pp 1ndash20 2009

[3] P A Filipek P J Accardo G T Baranek et al ldquoThe screeningand diagnosis of autistic spectrum disordersrdquo Journal of Autismand Developmental Disorders vol 29 pp 439ndash484 1999

[4] A Rizzo and J Buckwater ldquoVirtual reality and cognitiveassessment and rehabilitation the state of the artrdquo in VirtualReality in Neuro-Psycho-Physiology G Riva Ed vol 44 pp123ndash146 IOS Press Washington DC USA 1997

[5] R W Butler ldquoCognitive rehabilitationrdquo in Pediatric Neuropsy-chological Intervention S J Hunter and J Donders EdsCambridge University Press New York NY USA 2007

[6] A A RizzoM Schultheis K A Kerns andCMateer ldquoAnalysisof assets for virtual reality applications in neuropsychologyrdquoNeuropsychological Rehabilitation vol 14 no 1 pp 207ndash2392004

[7] G Burdea and P CoiffetVirtual Reality Technology JohnWileyamp Sons Hoboken NJ USA 2003

[8] BGWitmer C J Jerome andM J Singer ldquoThe factor structureof the presence questionnairerdquo Presence vol 14 no 3 pp 298ndash312 2005

[9] F D Rose E A Attree B M Brooks D M Parslow P RPenn and N Ambihaipahan ldquoTraining in virtual environ-ments transfer to real world tasks and equivalence to real tasktrainingrdquo Ergonomics vol 43 no 4 pp 494ndash511 2000

[10] D Strickland L M Marcus G B Mesibov and K HoganldquoBrief report two case studies using virtual reality as a learningtool for autistic childrenrdquo Journal of Autism and DevelopmentalDisorders vol 26 no 6 pp 651ndash659 1996

[11] P Mitchell S Parsons and A Leonard ldquoUsing virtual environ-ments for teaching social understanding to 6 adolescents withautistic spectrum disordersrdquo Journal of Autism and Develop-mental Disorders vol 37 no 3 pp 589ndash600 2007

[12] T Self R R Scudder G Weheba and D Crumrine ldquoA virtualapproach to teaching safety skills to children with autismspectrum disorderrdquo Topics in Language Disorders vol 27 no 3pp 242ndash253 2007

[13] M Wang and D Reid ldquoThe virtual reality-cognitive rehabili-tation (VR-CR) approach for children with autismrdquo Journal ofCyberTherapy and Rehabilitation vol 2 no 2 pp 95ndash102 2009

[14] M Bar ldquoVisual objects in contextrdquo Nature Reviews Neuro-science vol 5 no 8 pp 617ndash629 2004

[15] A Blaye and F Bonthoux ldquoThematic and taxonomic relationsin preschoolers the development of flexibility in categorizationchoicesrdquoBritish Journal of Developmental Psychology vol 19 no3 pp 395ndash412 2001

[16] D Ropar and D Peebles ldquoSorting preference in children withautism the dominance of concrete featuresrdquo Journal of Autismand Developmental Disorders vol 37 no 2 pp 270ndash280 2007

[17] L G Klinger and G Dawson ldquoPrototype formation in autismrdquoDevelopment and Psychopathology vol 13 no 1 pp 111ndash1242001

[18] T Jolliffe and S Baron-Cohen ldquoA test of central coherencetheory can adults with high-functioning autism or Asperger

syndrome integrate objects in contextrdquo Visual Cognition vol8 no 1 pp 67ndash101 2001

[19] E A Berg ldquoA simple objective technique formeasuring flexibil-ity in thinkingrdquo Journal of General Psychology vol 39 pp 15ndash221948

[20] D A Grant and E Berg ldquoA behavioral analysis of degree ofreinforcement and ease of shifting to new responses in aWeigl-type card-sorting problemrdquo Journal of Experimental Psychologyvol 38 no 4 pp 404ndash411 1948

[21] N Kaland L Smith and E L Mortensen ldquoBrief reportcognitive flexibility and focused attention in children and ado-lescents with Asperger syndrome or high-functioning autism asmeasured on the computerized version of the Wisconsin cardsorting testrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1161ndash1165 2008

[22] N J Minshew J Meyer and G Goldstein ldquoAbstract reasoningin autism a dissociation between concept formation andconcept identificationrdquo Neuropsychology vol 16 no 3 pp 327ndash334 2002

[23] S Ozonoff Specifying the Nature of the Primary Deficit inAutism an Investigation of Neuropsychological and Social-Cognitive Abilities in High-Functioning Autistic Children Uni-versity of Denver Denver Colo USA edition 1991

[24] M Prior and W Hoffmann ldquoBrief report neuropsychologicaltesting of autistic children through an exploration with frontallobe testsrdquo Journal of Autism and Developmental Disorders vol20 no 4 pp 581ndash590 1990

[25] E Schopler R J Reichler and B R Renner The ChildhoodAutism Rating Scale (CARS) Western Psychological ServicesLos Angeles Calif USA 1988

[26] G H Roid and L J Miller Leiter International PerformanceScale Revised Stoelting Wood Dale Ill USA 1997

[27] S Jacques The Roles of Labelling and Abstraction in the Devel-opment of Cognitive Flexibility University of Toronto TorontoCanada 2001

[28] H J C Berger F H T M Aerts K P M van SpaendonckA R Cools and J Teunisse ldquoCentral coherence and cognitiveshifting in relation to social improvement in high-functioningyoung adults with autismrdquo Journal of Clinical and ExperimentalNeuropsychology vol 25 no 4 pp 502ndash511 2003

[29] H J C Berger K P M van Spaendonck M W I MHorstink E L Buytenhuijs P W J M Lammers and A RCools ldquoCognitive shifting as a predictor of progress in socialunderstanding in high-functioning adolescents with autisma prospective studyrdquo Journal of Autism and DevelopmentalDisorders vol 23 no 2 pp 341ndash359 1993

[30] F Happe and U Frith ldquoThe weak coherence account detail-focused cognitive style in autism spectrum disordersrdquo Journalof Autism and Developmental Disorders vol 36 no 1 pp 5ndash252006

[31] B R Lopez A J Lincoln S Ozonoff and Z Lai ldquoExaminingthe relationship between executive functions and restrictedrepetitive symptoms ofAutisticDisorderrdquo Journal of Autism andDevelopmental Disorders vol 35 no 4 pp 445ndash460 2005

[32] S S Sparrow D A Balla andD V CicchettiVineland AdaptiveBehavior Scales Interview Edition Survey form Manual Amer-ican Guidance Service Circle Pines Minn USA 1984

[33] T E Scruggs and M A Mastropieri ldquoSummarizing single-subject research Issues and applicationsrdquo Behavior Modifica-tion vol 22 no 3 pp 221ndash242 1998

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

4 The Scientific World Journal

pairs are matched based on the spatial dimension and sixpairs are matched based on functional characteristics Halfof the pairs are matched correctly and half are matchedincorrectly The 18 test items are randomized differently foreach test version The VR test was administered during eachsession The software program and the researcher indepen-dently record the correct and incorrect responses for eachtest item The child does not receive feedback for his or herresponses on the VR test thus minimizing the occurrence oflearning effects as a result of repeated administration

242 Modified Version of the Flexible Item Selection Task(FIST-m) The Flexible Item Selection Task (FIST) [27] wasused to assess executive functioning On each test item threeobjects are shown (eg red fish blue fish and red telephone)On the first part of the task (Selection 1) the child is asked topoint to two objects that ldquogo togetherrdquoThese two itemsmatchon one relevant dimension (eg common object red fishand blue fish) Then the child is asked to point to a differentpair of objects that ldquogo togetherrdquo (Selection 2) This new pairmatches on a different dimension (eg common colour redfish and red telephone)The common item in both pairs is theldquopivot itemrdquo (eg red fish) Selection 1 involves the ability tointernally abstract a relevant dimension to identify the pairsAlthough Selection 2 also requires abstraction it is a goodmeasure of cognitive flexibility as the child must be able toswitch between different mental representations of the pivotobject

As the original FIST was developed for preschoolers theitems of the FIST-m have been adapted for the older 5-to-10age group Similar to the original the FIST-m comprises 12test items in total and includes items from the original testThe FIST-m was administered at prebaseline posttrainingand follow-up sessions

243 Attention Sustained Subtest [26] The Attention Sus-tained Subtest from the Leiter International PerformanceScale-Revised is a cancellation testwhichmeasures prolongedvisual attention visual scanning and visuomotor inhibitionIt consists of four separate test trials that require the childto colour a target shape or pattern (eg a square) withina complex array of different shapes None of the cognitiveconstructs evaluated by this subtest were explicitly taught inthis study thus it was used as a control test The AttentionSustained Subtest was administered at prebaseline posttrain-ing and follow-up sessions

244 Final Feedback Questionnaire The Final FeedbackQuestionnaire was developed to provide insight into parentalperceptions of behaviour changes in their children Thebehaviours on the form are those associated with con-textual processing and included seven general categoriesof behaviour in particular behaviour in public contextslanguage and communication in social contexts and flexibleuse of objects These categories of behaviours were chosenbased on reported correlations between these behavioursand cognitive impairments [28ndash31] The specific examplesfor each category were taken from the Vineland Adaptive

Behaviour Scale (VABS) [32] The Final Feedback Question-naire was administered once at the end of the study

25 Procedure

251 Baseline Phase Before initial testing the children wereprovided with prebaseline training to ensure that they under-stood the instructions associated with the VR test Using sim-plified pretest items (eg shapes) the childrsquos responses weremodelled prompted and reinforced during this pretrainingtask

252 Training Phase The training program was comprisedof three discrete lessons with one lesson taught per sessionEach lesson focused on one class of object characteristicsperceptual spatial or functional The goal of each lesson wasto teach the child to flexibly attend to object dimensionsof that particular class The lessons were taught in theorder as specified above and mastery on the precedinglesson was required to advance to the next Each trainingsession involved a 10-minute teaching protocol during whichthe same researcher provided each child with one-on-oneinstruction within the virtual environment The trainingsessions were designed to support the childrsquos understandingand performance on the task Verbal instructions modellingprompting repetition and reinforcement were the teachingstrategies that were incorporated

The virtual reality program presented each set of 10training items in a predictable sequence First themultiobjectcontext screen was shown (eg three objects displayed on atable) then the target object appeared The motion-capturevirtual technology allowed the child to see himherself on-screen and to indicate responses through gestures The childwas able to ldquograb and dragrdquo the target object across the screenDragging the target object towards the other items indicateda positive ldquomatchrdquo while dragging the target to the garbagebin indicated a negative ldquononmatchrdquo Visual reinforcementwas built into the training program with correct responsesrewarded with a happy face while incorrect responses werediscouragedwith a sad faceThere was no overlap in the itemsused on the VR test and those used in the training sessions

After completing each lesson the VR test was adminis-tered After each test administration theVR test was analyzedaccording to its separate components perceptual spatial andfunctional items A score of 80 or above was consideredmastery For example if the child scored over 80 on the VRtest perceptual items after Lesson 1 shehe would progress toLesson 2 A child who failed to achieve 80 on the VR testperceptual items after Lesson 1 would repeat that lesson Thetraining phase was ended when the child achieved over 80in all three components of the VR test (see Figure 1)

3 Results

31 VR Test The VR test percentage accuracy scores arepresented for each child across each session (Figure 2) Theaverage baseline scores ranged from 47 (Child 4) to 83(Child 2) Child 2 Child 3 and Child 4 showed marked

The Scientific World Journal 5

Phases

Week

Session

Baseline Training Follow-up

1 2 3 4 5

1 2 3 4 5 6 7 8 9 10 11 12

Figure 1 Overall study design for a hypothetical participant with a 5-session baseline phase 6-session training phase and 2-week follow-upsession

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11 12

Baseline Training Follow-upPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cyPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cy

Child

3

Session number

Child

1Ch

ild 2

Child

4

Figure 2 Percentage accuracy on the VR test demonstrated by each child across all phases of the study

improvements after the first training session which weremaintained throughout the training phase and at follow-up Child 1 showed significant improvements after the thirdtraining session and maintained this at follow-up

The PercentageNonoverlappingData statistic was used toanalyze the data [33] Child 2 Child 3 and Child 4 all demon-strated 100nonoverlapping data Child 1rsquos performance datashowed 60 nonoverlapping data

Overall the results demonstrate improvements in con-textual processing ability from baseline to treatment for

each child with average increases from 15 (Child 2) to46 (Child 4) All children maintained a high level ofperformance at the two-week follow-up assessment

32 FIST-m Percentage accuracy scores were calculated forpretraining and posttraining administrations of the FIST-m Table 2 shows these data for Selection 1 and Selection 2separately

All four children displayed ceiling or close to ceilingscores on Selection 1 (abstraction) scores at pretraining

6 The Scientific World Journal

Table 2 Percentage accuracy scores for Selections 1 and 2 of theFIST-m at pretraining posttraining and follow-up for each child

Selection 1 Selection 2Child 1

Pretraining 92 55Posttraining 92 73Follow-up 75 100

Child 2Pretraining 100 50Posttraining 100 100Follow-up 92 100

Child 3Pretraining 100 83Posttraining 100 92Follow-up 100 92

Child 4Pretraining 100 33Posttraining 92 100Follow-up 100 100

Table 3 Scaled scores for correct responses and error responsesof the attention sustained subtest at pretraining posttraining andfollow-up for each child

Scaled scores ofcorrect responses

Scaled scores of errorresponses

Child 1Pretraining 1 10Posttraining 1 10Follow-up 1 10

Child 2Pretraining 1 12Posttraining 1 12Follow-up 1 12

Child 3Pretraining 10 8Posttraining 16 7Follow-up 16 7

Child 4Pretraining 1 3Posttraining 1 3Follow-up 1 3

These high scores were generally maintained at posttrainingand at follow-up

All children showed improvements on Selection 2 (cog-nitive flexibility) Child 1 made progressive improvements onSelection 2 from 55 at baseline to 73 at posttraining and100 at follow-up Child 2 doubled her baseline score of 50and maintained 100 at follow-up Child 3 scored 83 atbaseline He showed improvement to 92 at posttraining andmaintained this at follow-up Child 4 more than tripled his

baseline score from 33 to 100 and maintained the highscore at follow-up

33 Control Test Attention Sustained Table 3 shows thescaled scores for the childrenrsquos performance on the AttentionSustained subtest Scaled scores for Correct Responses andError Responses are shown separately

Child 1 Child 2 and Child 4 showed no changes inperformance on the scaled scores for both correct responsesand error responses Child 3 showed an increase of 6 pointson scaled correct responses from baseline to posttrainingand a decrease in 1 point in scaled error responses atposttraining

34 Final Feedback Questionnaire The parents of Child 1and Child 4 reported no changes on any of the items onthe Final Feedback Questionnaire Both Child 2rsquos motherand Child 3rsquos mother reported changes in the category ofappropriate language and communication in social contextsChild 2rsquosmother reported thatChild 2 ldquoseems to initiate socialinteraction with more appropriate language (eg Hi Whatrsquosyour name I like your hair style rather than what colouris your nail polish)rdquo Child 3rsquos mother reported that Child3 ldquocan answer question(s) appropriatelyrdquo and that since thebeginning of the study Child 3 ldquois more flexible and like(s) totry something newrdquo

4 Discussion

Overall three of the four children showed 100 nonover-lapping data on the VR-test The fourth child showed 60nonoverlapping data No changes were found on the controltest for three children while the fourth child showed sig-nificant improvements on this test Finally parents of twochildren reported changes in the presence of appropriatesocial behaviours

The primary objective of the study was to explore if thenovel VR-CR intervention could improve contextual process-ing in children with autism According to the standards setby Logan and colleagues [34] the highest level of evidencefor single-subject research designs (Level 1) can be achievedby ldquoconcurrent or non-concurrent multiple baseline designs(MBD) with clear-cut results generalizability if the MBDdesign consists of a minimum of three subjects behavioursor settingsrdquo In determining ldquoclear-cut resultsrdquo Scruggs andMastropieri [33] assert that a treatment outcome with over90 nonoverlapping data can be considered a ldquohighly effec-tive treatmentrdquo The results of this study demonstrate clear100 nonoverlapping data for three children with threedifferent nonconcurrent baseline lengths Thus the currentstudy fulfills the criteria for a Level 1 single-subject designstudy

Changes in abstraction and cognitive flexibility were alsomeasured to verify improvements in contextual processingAs expected initial performance on Selection 2 (cognitiveflexibility) was low Three children performed at chance(50) or below After the training phase all children dis-played substantial improvements andmaintained a high level

The Scientific World Journal 7

of performance at follow-up Both Child 2 and Child 4showed notable improvements on Selection 2 on the FIST-m doubling and tripling their scores respectively Child 3also showed improvements however his performance onSelection 2 at baseline was initially high Although there is nonormative data available for comparison studies performedon the original FIST showed that during a crucial phase ofcognitive flexibility development in typical children (between4 and 5 years) 5-year-olds performed almost 18 better than4-year-olds on Selection 2 [27] The difference reflects therapid development in cognitive flexibility within that one-year span Three children Child 1 Child 2 and Child 4matched or exceeded this spike in development within onlytwo weeks Thus this is emerging evidence that cognitiveflexibility can be improved in this subgroup of children

Although impairments in abstraction have been demon-strated consistently in the literature [16] the children in thecurrent study demonstrated unexpectedly high abstractionperformance (FIST-m Selection 1) at baseline Findings fromRopar and Peebles [16] indicate that autistic children havedifficulty accessing ldquohigh-levelrdquo abstract categories such assports or games Although the current VR-CR interventionrequired the children to abstract qualities above the salientor perceptual the demands of the current task were demon-strated to be within the childrenrsquos range of capabilities

The control test was used to verify that the VR-CRintervention could specifically target contextual processingof objects rather than overall cognitive functioning Onthe Attention Sustained Subtest three of the four childrenshowed no changes over the course of the study supportingthe specificity of the intervention However one child Child3 showed significant improvements on one component of thesubtest from pretraining to posttraining Child 3 improved by6 points on the correct responses score which resulted in a10ndash13 increase in his ranked percentile This change is sig-nificant and may have been influenced by a couple of factorsincluding maturational effects and the level of rapport withinthe child-instructor therapeutic relationship Overall basedon the results of the control test the VR-CR interventiondemonstrated specificity to contextual processing for three offour of the children

The final objective of the pilot study was to explorechanges in context-related behaviours that may haveoccurred during the course of the study The commentsobtained by parents on the Final Feedback Questionnairewere mixed Child 1 and Child 4rsquos parents reported nobehavioural changes that occurred between the start andcompletion of the study Child 2 and Child 3rsquos mothersreported changes in the category of appropriate languageand communication in social contexts However it is notedthat both Child 2 and Child 3 were actively engaged inweekly social activities such as swimming and horsebackriding while Child 1 and Child 4 were not Thus the positivebehavioural changes reported are likely due to factors outsideof the intervention itself

The major limitation of this study is the high baselinelevels achieved by all the children prior to intervention Thissuggests that the participating children were already showinghigh levels of contextual processing ability before the study

particularly in abstraction Assessing performance on theVR-test prior to enrolment would help to exclude childrenfrom the study who are already performing at high levels onthe task In addition determining language levels or verbalintelligence at baseline may help to clarify those childrenwho would benefit maximally from this type of cognitiveintervention

A second limitation of this study is the lack of multipleindependent assessors The degree of therapeutic effective-ness and test performance was likely influenced by the devel-opment of rapport between the researcher and each child Tominimize this effect of researcher bias in the current studythere were no subjective rating scales involved In additionscoring of the VR test was verified through computer records

5 Conclusion

This pilot study evaluated a novel intervention that com-bined virtual reality technology with traditional cognitiverehabilitation methods to address impairments in contextualprocessing of objects in children with autism All childrenwho participated in the VR-CR program demonstratedstatistical improvements in overall contextual processingability and improvements in cognitive flexibility How-ever a larger-scale study will need to evaluate the mag-nitude of change resulting from the intervention and thedegree that cognitive improvements generalize to behaviourchanges

Overall this pilot study provides initial supporting evi-dence for the efficacy of this particular VR-CR interventionContinued research may further generate support for thebroader use of the VR-CR approach with children withautismThe hope is that comprehensive interventions will bedeveloped to target both cognitive and behavioural levels ofimpairment in children with autismThis may lead to greateroverall improvements in the daily functioning and quality oflife of these children

Conflict of Interests

The authors declare no conflict of interests

Acknowledgments

This study has been supported by the following Grantsawarded to the first author Canadian Institutes in HealthResearch 2008-09Masterrsquos Award Hilda andWilliam Court-neyClayton Paediatric ResearchAward 2008-09 andOntarioCouncil of Graduate Studies Autism Scholarrsquos Award 2009-10 This paper was prepared from a Masters thesis by thefirst author The authors give thanks to Bloorview KidsRehab Autism Ontario and the Geneva Centre for helpwith recruitment Special thanks to Fan (Mimi) Zhang fortechnical support and Luigi Girolametto and Tom Chaufor their mentorship and valuable feedback throughout theproject

8 The Scientific World Journal

References

[1] American Psychology Association Diagnostic and StatisticalManual ofMental Disorders DSM-IV-TR American PsychiatricAssociation Washington DC USA 4th edition 2000

[2] C Rice ldquoPrevalence of autism spectrum disordersmdashautism anddevelopmental disabilities monitoring network United States2006rdquo Morbidity and Mortality Weekly Report vol 58 no SS-10 pp 1ndash20 2009

[3] P A Filipek P J Accardo G T Baranek et al ldquoThe screeningand diagnosis of autistic spectrum disordersrdquo Journal of Autismand Developmental Disorders vol 29 pp 439ndash484 1999

[4] A Rizzo and J Buckwater ldquoVirtual reality and cognitiveassessment and rehabilitation the state of the artrdquo in VirtualReality in Neuro-Psycho-Physiology G Riva Ed vol 44 pp123ndash146 IOS Press Washington DC USA 1997

[5] R W Butler ldquoCognitive rehabilitationrdquo in Pediatric Neuropsy-chological Intervention S J Hunter and J Donders EdsCambridge University Press New York NY USA 2007

[6] A A RizzoM Schultheis K A Kerns andCMateer ldquoAnalysisof assets for virtual reality applications in neuropsychologyrdquoNeuropsychological Rehabilitation vol 14 no 1 pp 207ndash2392004

[7] G Burdea and P CoiffetVirtual Reality Technology JohnWileyamp Sons Hoboken NJ USA 2003

[8] BGWitmer C J Jerome andM J Singer ldquoThe factor structureof the presence questionnairerdquo Presence vol 14 no 3 pp 298ndash312 2005

[9] F D Rose E A Attree B M Brooks D M Parslow P RPenn and N Ambihaipahan ldquoTraining in virtual environ-ments transfer to real world tasks and equivalence to real tasktrainingrdquo Ergonomics vol 43 no 4 pp 494ndash511 2000

[10] D Strickland L M Marcus G B Mesibov and K HoganldquoBrief report two case studies using virtual reality as a learningtool for autistic childrenrdquo Journal of Autism and DevelopmentalDisorders vol 26 no 6 pp 651ndash659 1996

[11] P Mitchell S Parsons and A Leonard ldquoUsing virtual environ-ments for teaching social understanding to 6 adolescents withautistic spectrum disordersrdquo Journal of Autism and Develop-mental Disorders vol 37 no 3 pp 589ndash600 2007

[12] T Self R R Scudder G Weheba and D Crumrine ldquoA virtualapproach to teaching safety skills to children with autismspectrum disorderrdquo Topics in Language Disorders vol 27 no 3pp 242ndash253 2007

[13] M Wang and D Reid ldquoThe virtual reality-cognitive rehabili-tation (VR-CR) approach for children with autismrdquo Journal ofCyberTherapy and Rehabilitation vol 2 no 2 pp 95ndash102 2009

[14] M Bar ldquoVisual objects in contextrdquo Nature Reviews Neuro-science vol 5 no 8 pp 617ndash629 2004

[15] A Blaye and F Bonthoux ldquoThematic and taxonomic relationsin preschoolers the development of flexibility in categorizationchoicesrdquoBritish Journal of Developmental Psychology vol 19 no3 pp 395ndash412 2001

[16] D Ropar and D Peebles ldquoSorting preference in children withautism the dominance of concrete featuresrdquo Journal of Autismand Developmental Disorders vol 37 no 2 pp 270ndash280 2007

[17] L G Klinger and G Dawson ldquoPrototype formation in autismrdquoDevelopment and Psychopathology vol 13 no 1 pp 111ndash1242001

[18] T Jolliffe and S Baron-Cohen ldquoA test of central coherencetheory can adults with high-functioning autism or Asperger

syndrome integrate objects in contextrdquo Visual Cognition vol8 no 1 pp 67ndash101 2001

[19] E A Berg ldquoA simple objective technique formeasuring flexibil-ity in thinkingrdquo Journal of General Psychology vol 39 pp 15ndash221948

[20] D A Grant and E Berg ldquoA behavioral analysis of degree ofreinforcement and ease of shifting to new responses in aWeigl-type card-sorting problemrdquo Journal of Experimental Psychologyvol 38 no 4 pp 404ndash411 1948

[21] N Kaland L Smith and E L Mortensen ldquoBrief reportcognitive flexibility and focused attention in children and ado-lescents with Asperger syndrome or high-functioning autism asmeasured on the computerized version of the Wisconsin cardsorting testrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1161ndash1165 2008

[22] N J Minshew J Meyer and G Goldstein ldquoAbstract reasoningin autism a dissociation between concept formation andconcept identificationrdquo Neuropsychology vol 16 no 3 pp 327ndash334 2002

[23] S Ozonoff Specifying the Nature of the Primary Deficit inAutism an Investigation of Neuropsychological and Social-Cognitive Abilities in High-Functioning Autistic Children Uni-versity of Denver Denver Colo USA edition 1991

[24] M Prior and W Hoffmann ldquoBrief report neuropsychologicaltesting of autistic children through an exploration with frontallobe testsrdquo Journal of Autism and Developmental Disorders vol20 no 4 pp 581ndash590 1990

[25] E Schopler R J Reichler and B R Renner The ChildhoodAutism Rating Scale (CARS) Western Psychological ServicesLos Angeles Calif USA 1988

[26] G H Roid and L J Miller Leiter International PerformanceScale Revised Stoelting Wood Dale Ill USA 1997

[27] S Jacques The Roles of Labelling and Abstraction in the Devel-opment of Cognitive Flexibility University of Toronto TorontoCanada 2001

[28] H J C Berger F H T M Aerts K P M van SpaendonckA R Cools and J Teunisse ldquoCentral coherence and cognitiveshifting in relation to social improvement in high-functioningyoung adults with autismrdquo Journal of Clinical and ExperimentalNeuropsychology vol 25 no 4 pp 502ndash511 2003

[29] H J C Berger K P M van Spaendonck M W I MHorstink E L Buytenhuijs P W J M Lammers and A RCools ldquoCognitive shifting as a predictor of progress in socialunderstanding in high-functioning adolescents with autisma prospective studyrdquo Journal of Autism and DevelopmentalDisorders vol 23 no 2 pp 341ndash359 1993

[30] F Happe and U Frith ldquoThe weak coherence account detail-focused cognitive style in autism spectrum disordersrdquo Journalof Autism and Developmental Disorders vol 36 no 1 pp 5ndash252006

[31] B R Lopez A J Lincoln S Ozonoff and Z Lai ldquoExaminingthe relationship between executive functions and restrictedrepetitive symptoms ofAutisticDisorderrdquo Journal of Autism andDevelopmental Disorders vol 35 no 4 pp 445ndash460 2005

[32] S S Sparrow D A Balla andD V CicchettiVineland AdaptiveBehavior Scales Interview Edition Survey form Manual Amer-ican Guidance Service Circle Pines Minn USA 1984

[33] T E Scruggs and M A Mastropieri ldquoSummarizing single-subject research Issues and applicationsrdquo Behavior Modifica-tion vol 22 no 3 pp 221ndash242 1998

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

The Scientific World Journal 5

Phases

Week

Session

Baseline Training Follow-up

1 2 3 4 5

1 2 3 4 5 6 7 8 9 10 11 12

Figure 1 Overall study design for a hypothetical participant with a 5-session baseline phase 6-session training phase and 2-week follow-upsession

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11

0102030405060708090

100

0 1 2 3 4 5 6 7 8 9 10 11 12

Baseline Training Follow-upPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cyPe

rcen

tage

accu

racy

Perc

enta

ge ac

cura

cy

Child

3

Session number

Child

1Ch

ild 2

Child

4

Figure 2 Percentage accuracy on the VR test demonstrated by each child across all phases of the study

improvements after the first training session which weremaintained throughout the training phase and at follow-up Child 1 showed significant improvements after the thirdtraining session and maintained this at follow-up

The PercentageNonoverlappingData statistic was used toanalyze the data [33] Child 2 Child 3 and Child 4 all demon-strated 100nonoverlapping data Child 1rsquos performance datashowed 60 nonoverlapping data

Overall the results demonstrate improvements in con-textual processing ability from baseline to treatment for

each child with average increases from 15 (Child 2) to46 (Child 4) All children maintained a high level ofperformance at the two-week follow-up assessment

32 FIST-m Percentage accuracy scores were calculated forpretraining and posttraining administrations of the FIST-m Table 2 shows these data for Selection 1 and Selection 2separately

All four children displayed ceiling or close to ceilingscores on Selection 1 (abstraction) scores at pretraining

6 The Scientific World Journal

Table 2 Percentage accuracy scores for Selections 1 and 2 of theFIST-m at pretraining posttraining and follow-up for each child

Selection 1 Selection 2Child 1

Pretraining 92 55Posttraining 92 73Follow-up 75 100

Child 2Pretraining 100 50Posttraining 100 100Follow-up 92 100

Child 3Pretraining 100 83Posttraining 100 92Follow-up 100 92

Child 4Pretraining 100 33Posttraining 92 100Follow-up 100 100

Table 3 Scaled scores for correct responses and error responsesof the attention sustained subtest at pretraining posttraining andfollow-up for each child

Scaled scores ofcorrect responses

Scaled scores of errorresponses

Child 1Pretraining 1 10Posttraining 1 10Follow-up 1 10

Child 2Pretraining 1 12Posttraining 1 12Follow-up 1 12

Child 3Pretraining 10 8Posttraining 16 7Follow-up 16 7

Child 4Pretraining 1 3Posttraining 1 3Follow-up 1 3

These high scores were generally maintained at posttrainingand at follow-up

All children showed improvements on Selection 2 (cog-nitive flexibility) Child 1 made progressive improvements onSelection 2 from 55 at baseline to 73 at posttraining and100 at follow-up Child 2 doubled her baseline score of 50and maintained 100 at follow-up Child 3 scored 83 atbaseline He showed improvement to 92 at posttraining andmaintained this at follow-up Child 4 more than tripled his

baseline score from 33 to 100 and maintained the highscore at follow-up

33 Control Test Attention Sustained Table 3 shows thescaled scores for the childrenrsquos performance on the AttentionSustained subtest Scaled scores for Correct Responses andError Responses are shown separately

Child 1 Child 2 and Child 4 showed no changes inperformance on the scaled scores for both correct responsesand error responses Child 3 showed an increase of 6 pointson scaled correct responses from baseline to posttrainingand a decrease in 1 point in scaled error responses atposttraining

34 Final Feedback Questionnaire The parents of Child 1and Child 4 reported no changes on any of the items onthe Final Feedback Questionnaire Both Child 2rsquos motherand Child 3rsquos mother reported changes in the category ofappropriate language and communication in social contextsChild 2rsquosmother reported thatChild 2 ldquoseems to initiate socialinteraction with more appropriate language (eg Hi Whatrsquosyour name I like your hair style rather than what colouris your nail polish)rdquo Child 3rsquos mother reported that Child3 ldquocan answer question(s) appropriatelyrdquo and that since thebeginning of the study Child 3 ldquois more flexible and like(s) totry something newrdquo

4 Discussion

Overall three of the four children showed 100 nonover-lapping data on the VR-test The fourth child showed 60nonoverlapping data No changes were found on the controltest for three children while the fourth child showed sig-nificant improvements on this test Finally parents of twochildren reported changes in the presence of appropriatesocial behaviours

The primary objective of the study was to explore if thenovel VR-CR intervention could improve contextual process-ing in children with autism According to the standards setby Logan and colleagues [34] the highest level of evidencefor single-subject research designs (Level 1) can be achievedby ldquoconcurrent or non-concurrent multiple baseline designs(MBD) with clear-cut results generalizability if the MBDdesign consists of a minimum of three subjects behavioursor settingsrdquo In determining ldquoclear-cut resultsrdquo Scruggs andMastropieri [33] assert that a treatment outcome with over90 nonoverlapping data can be considered a ldquohighly effec-tive treatmentrdquo The results of this study demonstrate clear100 nonoverlapping data for three children with threedifferent nonconcurrent baseline lengths Thus the currentstudy fulfills the criteria for a Level 1 single-subject designstudy

Changes in abstraction and cognitive flexibility were alsomeasured to verify improvements in contextual processingAs expected initial performance on Selection 2 (cognitiveflexibility) was low Three children performed at chance(50) or below After the training phase all children dis-played substantial improvements andmaintained a high level

The Scientific World Journal 7

of performance at follow-up Both Child 2 and Child 4showed notable improvements on Selection 2 on the FIST-m doubling and tripling their scores respectively Child 3also showed improvements however his performance onSelection 2 at baseline was initially high Although there is nonormative data available for comparison studies performedon the original FIST showed that during a crucial phase ofcognitive flexibility development in typical children (between4 and 5 years) 5-year-olds performed almost 18 better than4-year-olds on Selection 2 [27] The difference reflects therapid development in cognitive flexibility within that one-year span Three children Child 1 Child 2 and Child 4matched or exceeded this spike in development within onlytwo weeks Thus this is emerging evidence that cognitiveflexibility can be improved in this subgroup of children

Although impairments in abstraction have been demon-strated consistently in the literature [16] the children in thecurrent study demonstrated unexpectedly high abstractionperformance (FIST-m Selection 1) at baseline Findings fromRopar and Peebles [16] indicate that autistic children havedifficulty accessing ldquohigh-levelrdquo abstract categories such assports or games Although the current VR-CR interventionrequired the children to abstract qualities above the salientor perceptual the demands of the current task were demon-strated to be within the childrenrsquos range of capabilities

The control test was used to verify that the VR-CRintervention could specifically target contextual processingof objects rather than overall cognitive functioning Onthe Attention Sustained Subtest three of the four childrenshowed no changes over the course of the study supportingthe specificity of the intervention However one child Child3 showed significant improvements on one component of thesubtest from pretraining to posttraining Child 3 improved by6 points on the correct responses score which resulted in a10ndash13 increase in his ranked percentile This change is sig-nificant and may have been influenced by a couple of factorsincluding maturational effects and the level of rapport withinthe child-instructor therapeutic relationship Overall basedon the results of the control test the VR-CR interventiondemonstrated specificity to contextual processing for three offour of the children

The final objective of the pilot study was to explorechanges in context-related behaviours that may haveoccurred during the course of the study The commentsobtained by parents on the Final Feedback Questionnairewere mixed Child 1 and Child 4rsquos parents reported nobehavioural changes that occurred between the start andcompletion of the study Child 2 and Child 3rsquos mothersreported changes in the category of appropriate languageand communication in social contexts However it is notedthat both Child 2 and Child 3 were actively engaged inweekly social activities such as swimming and horsebackriding while Child 1 and Child 4 were not Thus the positivebehavioural changes reported are likely due to factors outsideof the intervention itself

The major limitation of this study is the high baselinelevels achieved by all the children prior to intervention Thissuggests that the participating children were already showinghigh levels of contextual processing ability before the study

particularly in abstraction Assessing performance on theVR-test prior to enrolment would help to exclude childrenfrom the study who are already performing at high levels onthe task In addition determining language levels or verbalintelligence at baseline may help to clarify those childrenwho would benefit maximally from this type of cognitiveintervention

A second limitation of this study is the lack of multipleindependent assessors The degree of therapeutic effective-ness and test performance was likely influenced by the devel-opment of rapport between the researcher and each child Tominimize this effect of researcher bias in the current studythere were no subjective rating scales involved In additionscoring of the VR test was verified through computer records

5 Conclusion

This pilot study evaluated a novel intervention that com-bined virtual reality technology with traditional cognitiverehabilitation methods to address impairments in contextualprocessing of objects in children with autism All childrenwho participated in the VR-CR program demonstratedstatistical improvements in overall contextual processingability and improvements in cognitive flexibility How-ever a larger-scale study will need to evaluate the mag-nitude of change resulting from the intervention and thedegree that cognitive improvements generalize to behaviourchanges

Overall this pilot study provides initial supporting evi-dence for the efficacy of this particular VR-CR interventionContinued research may further generate support for thebroader use of the VR-CR approach with children withautismThe hope is that comprehensive interventions will bedeveloped to target both cognitive and behavioural levels ofimpairment in children with autismThis may lead to greateroverall improvements in the daily functioning and quality oflife of these children

Conflict of Interests

The authors declare no conflict of interests

Acknowledgments

This study has been supported by the following Grantsawarded to the first author Canadian Institutes in HealthResearch 2008-09Masterrsquos Award Hilda andWilliam Court-neyClayton Paediatric ResearchAward 2008-09 andOntarioCouncil of Graduate Studies Autism Scholarrsquos Award 2009-10 This paper was prepared from a Masters thesis by thefirst author The authors give thanks to Bloorview KidsRehab Autism Ontario and the Geneva Centre for helpwith recruitment Special thanks to Fan (Mimi) Zhang fortechnical support and Luigi Girolametto and Tom Chaufor their mentorship and valuable feedback throughout theproject

8 The Scientific World Journal

References

[1] American Psychology Association Diagnostic and StatisticalManual ofMental Disorders DSM-IV-TR American PsychiatricAssociation Washington DC USA 4th edition 2000

[2] C Rice ldquoPrevalence of autism spectrum disordersmdashautism anddevelopmental disabilities monitoring network United States2006rdquo Morbidity and Mortality Weekly Report vol 58 no SS-10 pp 1ndash20 2009

[3] P A Filipek P J Accardo G T Baranek et al ldquoThe screeningand diagnosis of autistic spectrum disordersrdquo Journal of Autismand Developmental Disorders vol 29 pp 439ndash484 1999

[4] A Rizzo and J Buckwater ldquoVirtual reality and cognitiveassessment and rehabilitation the state of the artrdquo in VirtualReality in Neuro-Psycho-Physiology G Riva Ed vol 44 pp123ndash146 IOS Press Washington DC USA 1997

[5] R W Butler ldquoCognitive rehabilitationrdquo in Pediatric Neuropsy-chological Intervention S J Hunter and J Donders EdsCambridge University Press New York NY USA 2007

[6] A A RizzoM Schultheis K A Kerns andCMateer ldquoAnalysisof assets for virtual reality applications in neuropsychologyrdquoNeuropsychological Rehabilitation vol 14 no 1 pp 207ndash2392004

[7] G Burdea and P CoiffetVirtual Reality Technology JohnWileyamp Sons Hoboken NJ USA 2003

[8] BGWitmer C J Jerome andM J Singer ldquoThe factor structureof the presence questionnairerdquo Presence vol 14 no 3 pp 298ndash312 2005

[9] F D Rose E A Attree B M Brooks D M Parslow P RPenn and N Ambihaipahan ldquoTraining in virtual environ-ments transfer to real world tasks and equivalence to real tasktrainingrdquo Ergonomics vol 43 no 4 pp 494ndash511 2000

[10] D Strickland L M Marcus G B Mesibov and K HoganldquoBrief report two case studies using virtual reality as a learningtool for autistic childrenrdquo Journal of Autism and DevelopmentalDisorders vol 26 no 6 pp 651ndash659 1996

[11] P Mitchell S Parsons and A Leonard ldquoUsing virtual environ-ments for teaching social understanding to 6 adolescents withautistic spectrum disordersrdquo Journal of Autism and Develop-mental Disorders vol 37 no 3 pp 589ndash600 2007

[12] T Self R R Scudder G Weheba and D Crumrine ldquoA virtualapproach to teaching safety skills to children with autismspectrum disorderrdquo Topics in Language Disorders vol 27 no 3pp 242ndash253 2007

[13] M Wang and D Reid ldquoThe virtual reality-cognitive rehabili-tation (VR-CR) approach for children with autismrdquo Journal ofCyberTherapy and Rehabilitation vol 2 no 2 pp 95ndash102 2009

[14] M Bar ldquoVisual objects in contextrdquo Nature Reviews Neuro-science vol 5 no 8 pp 617ndash629 2004

[15] A Blaye and F Bonthoux ldquoThematic and taxonomic relationsin preschoolers the development of flexibility in categorizationchoicesrdquoBritish Journal of Developmental Psychology vol 19 no3 pp 395ndash412 2001

[16] D Ropar and D Peebles ldquoSorting preference in children withautism the dominance of concrete featuresrdquo Journal of Autismand Developmental Disorders vol 37 no 2 pp 270ndash280 2007

[17] L G Klinger and G Dawson ldquoPrototype formation in autismrdquoDevelopment and Psychopathology vol 13 no 1 pp 111ndash1242001

[18] T Jolliffe and S Baron-Cohen ldquoA test of central coherencetheory can adults with high-functioning autism or Asperger

syndrome integrate objects in contextrdquo Visual Cognition vol8 no 1 pp 67ndash101 2001

[19] E A Berg ldquoA simple objective technique formeasuring flexibil-ity in thinkingrdquo Journal of General Psychology vol 39 pp 15ndash221948

[20] D A Grant and E Berg ldquoA behavioral analysis of degree ofreinforcement and ease of shifting to new responses in aWeigl-type card-sorting problemrdquo Journal of Experimental Psychologyvol 38 no 4 pp 404ndash411 1948

[21] N Kaland L Smith and E L Mortensen ldquoBrief reportcognitive flexibility and focused attention in children and ado-lescents with Asperger syndrome or high-functioning autism asmeasured on the computerized version of the Wisconsin cardsorting testrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1161ndash1165 2008

[22] N J Minshew J Meyer and G Goldstein ldquoAbstract reasoningin autism a dissociation between concept formation andconcept identificationrdquo Neuropsychology vol 16 no 3 pp 327ndash334 2002

[23] S Ozonoff Specifying the Nature of the Primary Deficit inAutism an Investigation of Neuropsychological and Social-Cognitive Abilities in High-Functioning Autistic Children Uni-versity of Denver Denver Colo USA edition 1991

[24] M Prior and W Hoffmann ldquoBrief report neuropsychologicaltesting of autistic children through an exploration with frontallobe testsrdquo Journal of Autism and Developmental Disorders vol20 no 4 pp 581ndash590 1990

[25] E Schopler R J Reichler and B R Renner The ChildhoodAutism Rating Scale (CARS) Western Psychological ServicesLos Angeles Calif USA 1988

[26] G H Roid and L J Miller Leiter International PerformanceScale Revised Stoelting Wood Dale Ill USA 1997

[27] S Jacques The Roles of Labelling and Abstraction in the Devel-opment of Cognitive Flexibility University of Toronto TorontoCanada 2001

[28] H J C Berger F H T M Aerts K P M van SpaendonckA R Cools and J Teunisse ldquoCentral coherence and cognitiveshifting in relation to social improvement in high-functioningyoung adults with autismrdquo Journal of Clinical and ExperimentalNeuropsychology vol 25 no 4 pp 502ndash511 2003

[29] H J C Berger K P M van Spaendonck M W I MHorstink E L Buytenhuijs P W J M Lammers and A RCools ldquoCognitive shifting as a predictor of progress in socialunderstanding in high-functioning adolescents with autisma prospective studyrdquo Journal of Autism and DevelopmentalDisorders vol 23 no 2 pp 341ndash359 1993

[30] F Happe and U Frith ldquoThe weak coherence account detail-focused cognitive style in autism spectrum disordersrdquo Journalof Autism and Developmental Disorders vol 36 no 1 pp 5ndash252006

[31] B R Lopez A J Lincoln S Ozonoff and Z Lai ldquoExaminingthe relationship between executive functions and restrictedrepetitive symptoms ofAutisticDisorderrdquo Journal of Autism andDevelopmental Disorders vol 35 no 4 pp 445ndash460 2005

[32] S S Sparrow D A Balla andD V CicchettiVineland AdaptiveBehavior Scales Interview Edition Survey form Manual Amer-ican Guidance Service Circle Pines Minn USA 1984

[33] T E Scruggs and M A Mastropieri ldquoSummarizing single-subject research Issues and applicationsrdquo Behavior Modifica-tion vol 22 no 3 pp 221ndash242 1998

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

6 The Scientific World Journal

Table 2 Percentage accuracy scores for Selections 1 and 2 of theFIST-m at pretraining posttraining and follow-up for each child

Selection 1 Selection 2Child 1

Pretraining 92 55Posttraining 92 73Follow-up 75 100

Child 2Pretraining 100 50Posttraining 100 100Follow-up 92 100

Child 3Pretraining 100 83Posttraining 100 92Follow-up 100 92

Child 4Pretraining 100 33Posttraining 92 100Follow-up 100 100

Table 3 Scaled scores for correct responses and error responsesof the attention sustained subtest at pretraining posttraining andfollow-up for each child

Scaled scores ofcorrect responses

Scaled scores of errorresponses

Child 1Pretraining 1 10Posttraining 1 10Follow-up 1 10

Child 2Pretraining 1 12Posttraining 1 12Follow-up 1 12

Child 3Pretraining 10 8Posttraining 16 7Follow-up 16 7

Child 4Pretraining 1 3Posttraining 1 3Follow-up 1 3

These high scores were generally maintained at posttrainingand at follow-up

All children showed improvements on Selection 2 (cog-nitive flexibility) Child 1 made progressive improvements onSelection 2 from 55 at baseline to 73 at posttraining and100 at follow-up Child 2 doubled her baseline score of 50and maintained 100 at follow-up Child 3 scored 83 atbaseline He showed improvement to 92 at posttraining andmaintained this at follow-up Child 4 more than tripled his

baseline score from 33 to 100 and maintained the highscore at follow-up

33 Control Test Attention Sustained Table 3 shows thescaled scores for the childrenrsquos performance on the AttentionSustained subtest Scaled scores for Correct Responses andError Responses are shown separately

Child 1 Child 2 and Child 4 showed no changes inperformance on the scaled scores for both correct responsesand error responses Child 3 showed an increase of 6 pointson scaled correct responses from baseline to posttrainingand a decrease in 1 point in scaled error responses atposttraining

34 Final Feedback Questionnaire The parents of Child 1and Child 4 reported no changes on any of the items onthe Final Feedback Questionnaire Both Child 2rsquos motherand Child 3rsquos mother reported changes in the category ofappropriate language and communication in social contextsChild 2rsquosmother reported thatChild 2 ldquoseems to initiate socialinteraction with more appropriate language (eg Hi Whatrsquosyour name I like your hair style rather than what colouris your nail polish)rdquo Child 3rsquos mother reported that Child3 ldquocan answer question(s) appropriatelyrdquo and that since thebeginning of the study Child 3 ldquois more flexible and like(s) totry something newrdquo

4 Discussion

Overall three of the four children showed 100 nonover-lapping data on the VR-test The fourth child showed 60nonoverlapping data No changes were found on the controltest for three children while the fourth child showed sig-nificant improvements on this test Finally parents of twochildren reported changes in the presence of appropriatesocial behaviours

The primary objective of the study was to explore if thenovel VR-CR intervention could improve contextual process-ing in children with autism According to the standards setby Logan and colleagues [34] the highest level of evidencefor single-subject research designs (Level 1) can be achievedby ldquoconcurrent or non-concurrent multiple baseline designs(MBD) with clear-cut results generalizability if the MBDdesign consists of a minimum of three subjects behavioursor settingsrdquo In determining ldquoclear-cut resultsrdquo Scruggs andMastropieri [33] assert that a treatment outcome with over90 nonoverlapping data can be considered a ldquohighly effec-tive treatmentrdquo The results of this study demonstrate clear100 nonoverlapping data for three children with threedifferent nonconcurrent baseline lengths Thus the currentstudy fulfills the criteria for a Level 1 single-subject designstudy

Changes in abstraction and cognitive flexibility were alsomeasured to verify improvements in contextual processingAs expected initial performance on Selection 2 (cognitiveflexibility) was low Three children performed at chance(50) or below After the training phase all children dis-played substantial improvements andmaintained a high level

The Scientific World Journal 7

of performance at follow-up Both Child 2 and Child 4showed notable improvements on Selection 2 on the FIST-m doubling and tripling their scores respectively Child 3also showed improvements however his performance onSelection 2 at baseline was initially high Although there is nonormative data available for comparison studies performedon the original FIST showed that during a crucial phase ofcognitive flexibility development in typical children (between4 and 5 years) 5-year-olds performed almost 18 better than4-year-olds on Selection 2 [27] The difference reflects therapid development in cognitive flexibility within that one-year span Three children Child 1 Child 2 and Child 4matched or exceeded this spike in development within onlytwo weeks Thus this is emerging evidence that cognitiveflexibility can be improved in this subgroup of children

Although impairments in abstraction have been demon-strated consistently in the literature [16] the children in thecurrent study demonstrated unexpectedly high abstractionperformance (FIST-m Selection 1) at baseline Findings fromRopar and Peebles [16] indicate that autistic children havedifficulty accessing ldquohigh-levelrdquo abstract categories such assports or games Although the current VR-CR interventionrequired the children to abstract qualities above the salientor perceptual the demands of the current task were demon-strated to be within the childrenrsquos range of capabilities

The control test was used to verify that the VR-CRintervention could specifically target contextual processingof objects rather than overall cognitive functioning Onthe Attention Sustained Subtest three of the four childrenshowed no changes over the course of the study supportingthe specificity of the intervention However one child Child3 showed significant improvements on one component of thesubtest from pretraining to posttraining Child 3 improved by6 points on the correct responses score which resulted in a10ndash13 increase in his ranked percentile This change is sig-nificant and may have been influenced by a couple of factorsincluding maturational effects and the level of rapport withinthe child-instructor therapeutic relationship Overall basedon the results of the control test the VR-CR interventiondemonstrated specificity to contextual processing for three offour of the children

The final objective of the pilot study was to explorechanges in context-related behaviours that may haveoccurred during the course of the study The commentsobtained by parents on the Final Feedback Questionnairewere mixed Child 1 and Child 4rsquos parents reported nobehavioural changes that occurred between the start andcompletion of the study Child 2 and Child 3rsquos mothersreported changes in the category of appropriate languageand communication in social contexts However it is notedthat both Child 2 and Child 3 were actively engaged inweekly social activities such as swimming and horsebackriding while Child 1 and Child 4 were not Thus the positivebehavioural changes reported are likely due to factors outsideof the intervention itself

The major limitation of this study is the high baselinelevels achieved by all the children prior to intervention Thissuggests that the participating children were already showinghigh levels of contextual processing ability before the study

particularly in abstraction Assessing performance on theVR-test prior to enrolment would help to exclude childrenfrom the study who are already performing at high levels onthe task In addition determining language levels or verbalintelligence at baseline may help to clarify those childrenwho would benefit maximally from this type of cognitiveintervention

A second limitation of this study is the lack of multipleindependent assessors The degree of therapeutic effective-ness and test performance was likely influenced by the devel-opment of rapport between the researcher and each child Tominimize this effect of researcher bias in the current studythere were no subjective rating scales involved In additionscoring of the VR test was verified through computer records

5 Conclusion

This pilot study evaluated a novel intervention that com-bined virtual reality technology with traditional cognitiverehabilitation methods to address impairments in contextualprocessing of objects in children with autism All childrenwho participated in the VR-CR program demonstratedstatistical improvements in overall contextual processingability and improvements in cognitive flexibility How-ever a larger-scale study will need to evaluate the mag-nitude of change resulting from the intervention and thedegree that cognitive improvements generalize to behaviourchanges

Overall this pilot study provides initial supporting evi-dence for the efficacy of this particular VR-CR interventionContinued research may further generate support for thebroader use of the VR-CR approach with children withautismThe hope is that comprehensive interventions will bedeveloped to target both cognitive and behavioural levels ofimpairment in children with autismThis may lead to greateroverall improvements in the daily functioning and quality oflife of these children

Conflict of Interests

The authors declare no conflict of interests

Acknowledgments

This study has been supported by the following Grantsawarded to the first author Canadian Institutes in HealthResearch 2008-09Masterrsquos Award Hilda andWilliam Court-neyClayton Paediatric ResearchAward 2008-09 andOntarioCouncil of Graduate Studies Autism Scholarrsquos Award 2009-10 This paper was prepared from a Masters thesis by thefirst author The authors give thanks to Bloorview KidsRehab Autism Ontario and the Geneva Centre for helpwith recruitment Special thanks to Fan (Mimi) Zhang fortechnical support and Luigi Girolametto and Tom Chaufor their mentorship and valuable feedback throughout theproject

8 The Scientific World Journal

References

[1] American Psychology Association Diagnostic and StatisticalManual ofMental Disorders DSM-IV-TR American PsychiatricAssociation Washington DC USA 4th edition 2000

[2] C Rice ldquoPrevalence of autism spectrum disordersmdashautism anddevelopmental disabilities monitoring network United States2006rdquo Morbidity and Mortality Weekly Report vol 58 no SS-10 pp 1ndash20 2009

[3] P A Filipek P J Accardo G T Baranek et al ldquoThe screeningand diagnosis of autistic spectrum disordersrdquo Journal of Autismand Developmental Disorders vol 29 pp 439ndash484 1999

[4] A Rizzo and J Buckwater ldquoVirtual reality and cognitiveassessment and rehabilitation the state of the artrdquo in VirtualReality in Neuro-Psycho-Physiology G Riva Ed vol 44 pp123ndash146 IOS Press Washington DC USA 1997

[5] R W Butler ldquoCognitive rehabilitationrdquo in Pediatric Neuropsy-chological Intervention S J Hunter and J Donders EdsCambridge University Press New York NY USA 2007

[6] A A RizzoM Schultheis K A Kerns andCMateer ldquoAnalysisof assets for virtual reality applications in neuropsychologyrdquoNeuropsychological Rehabilitation vol 14 no 1 pp 207ndash2392004

[7] G Burdea and P CoiffetVirtual Reality Technology JohnWileyamp Sons Hoboken NJ USA 2003

[8] BGWitmer C J Jerome andM J Singer ldquoThe factor structureof the presence questionnairerdquo Presence vol 14 no 3 pp 298ndash312 2005

[9] F D Rose E A Attree B M Brooks D M Parslow P RPenn and N Ambihaipahan ldquoTraining in virtual environ-ments transfer to real world tasks and equivalence to real tasktrainingrdquo Ergonomics vol 43 no 4 pp 494ndash511 2000

[10] D Strickland L M Marcus G B Mesibov and K HoganldquoBrief report two case studies using virtual reality as a learningtool for autistic childrenrdquo Journal of Autism and DevelopmentalDisorders vol 26 no 6 pp 651ndash659 1996

[11] P Mitchell S Parsons and A Leonard ldquoUsing virtual environ-ments for teaching social understanding to 6 adolescents withautistic spectrum disordersrdquo Journal of Autism and Develop-mental Disorders vol 37 no 3 pp 589ndash600 2007

[12] T Self R R Scudder G Weheba and D Crumrine ldquoA virtualapproach to teaching safety skills to children with autismspectrum disorderrdquo Topics in Language Disorders vol 27 no 3pp 242ndash253 2007

[13] M Wang and D Reid ldquoThe virtual reality-cognitive rehabili-tation (VR-CR) approach for children with autismrdquo Journal ofCyberTherapy and Rehabilitation vol 2 no 2 pp 95ndash102 2009

[14] M Bar ldquoVisual objects in contextrdquo Nature Reviews Neuro-science vol 5 no 8 pp 617ndash629 2004

[15] A Blaye and F Bonthoux ldquoThematic and taxonomic relationsin preschoolers the development of flexibility in categorizationchoicesrdquoBritish Journal of Developmental Psychology vol 19 no3 pp 395ndash412 2001

[16] D Ropar and D Peebles ldquoSorting preference in children withautism the dominance of concrete featuresrdquo Journal of Autismand Developmental Disorders vol 37 no 2 pp 270ndash280 2007

[17] L G Klinger and G Dawson ldquoPrototype formation in autismrdquoDevelopment and Psychopathology vol 13 no 1 pp 111ndash1242001

[18] T Jolliffe and S Baron-Cohen ldquoA test of central coherencetheory can adults with high-functioning autism or Asperger

syndrome integrate objects in contextrdquo Visual Cognition vol8 no 1 pp 67ndash101 2001

[19] E A Berg ldquoA simple objective technique formeasuring flexibil-ity in thinkingrdquo Journal of General Psychology vol 39 pp 15ndash221948

[20] D A Grant and E Berg ldquoA behavioral analysis of degree ofreinforcement and ease of shifting to new responses in aWeigl-type card-sorting problemrdquo Journal of Experimental Psychologyvol 38 no 4 pp 404ndash411 1948

[21] N Kaland L Smith and E L Mortensen ldquoBrief reportcognitive flexibility and focused attention in children and ado-lescents with Asperger syndrome or high-functioning autism asmeasured on the computerized version of the Wisconsin cardsorting testrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1161ndash1165 2008

[22] N J Minshew J Meyer and G Goldstein ldquoAbstract reasoningin autism a dissociation between concept formation andconcept identificationrdquo Neuropsychology vol 16 no 3 pp 327ndash334 2002

[23] S Ozonoff Specifying the Nature of the Primary Deficit inAutism an Investigation of Neuropsychological and Social-Cognitive Abilities in High-Functioning Autistic Children Uni-versity of Denver Denver Colo USA edition 1991

[24] M Prior and W Hoffmann ldquoBrief report neuropsychologicaltesting of autistic children through an exploration with frontallobe testsrdquo Journal of Autism and Developmental Disorders vol20 no 4 pp 581ndash590 1990

[25] E Schopler R J Reichler and B R Renner The ChildhoodAutism Rating Scale (CARS) Western Psychological ServicesLos Angeles Calif USA 1988

[26] G H Roid and L J Miller Leiter International PerformanceScale Revised Stoelting Wood Dale Ill USA 1997

[27] S Jacques The Roles of Labelling and Abstraction in the Devel-opment of Cognitive Flexibility University of Toronto TorontoCanada 2001

[28] H J C Berger F H T M Aerts K P M van SpaendonckA R Cools and J Teunisse ldquoCentral coherence and cognitiveshifting in relation to social improvement in high-functioningyoung adults with autismrdquo Journal of Clinical and ExperimentalNeuropsychology vol 25 no 4 pp 502ndash511 2003

[29] H J C Berger K P M van Spaendonck M W I MHorstink E L Buytenhuijs P W J M Lammers and A RCools ldquoCognitive shifting as a predictor of progress in socialunderstanding in high-functioning adolescents with autisma prospective studyrdquo Journal of Autism and DevelopmentalDisorders vol 23 no 2 pp 341ndash359 1993

[30] F Happe and U Frith ldquoThe weak coherence account detail-focused cognitive style in autism spectrum disordersrdquo Journalof Autism and Developmental Disorders vol 36 no 1 pp 5ndash252006

[31] B R Lopez A J Lincoln S Ozonoff and Z Lai ldquoExaminingthe relationship between executive functions and restrictedrepetitive symptoms ofAutisticDisorderrdquo Journal of Autism andDevelopmental Disorders vol 35 no 4 pp 445ndash460 2005

[32] S S Sparrow D A Balla andD V CicchettiVineland AdaptiveBehavior Scales Interview Edition Survey form Manual Amer-ican Guidance Service Circle Pines Minn USA 1984

[33] T E Scruggs and M A Mastropieri ldquoSummarizing single-subject research Issues and applicationsrdquo Behavior Modifica-tion vol 22 no 3 pp 221ndash242 1998

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

The Scientific World Journal 7

of performance at follow-up Both Child 2 and Child 4showed notable improvements on Selection 2 on the FIST-m doubling and tripling their scores respectively Child 3also showed improvements however his performance onSelection 2 at baseline was initially high Although there is nonormative data available for comparison studies performedon the original FIST showed that during a crucial phase ofcognitive flexibility development in typical children (between4 and 5 years) 5-year-olds performed almost 18 better than4-year-olds on Selection 2 [27] The difference reflects therapid development in cognitive flexibility within that one-year span Three children Child 1 Child 2 and Child 4matched or exceeded this spike in development within onlytwo weeks Thus this is emerging evidence that cognitiveflexibility can be improved in this subgroup of children

Although impairments in abstraction have been demon-strated consistently in the literature [16] the children in thecurrent study demonstrated unexpectedly high abstractionperformance (FIST-m Selection 1) at baseline Findings fromRopar and Peebles [16] indicate that autistic children havedifficulty accessing ldquohigh-levelrdquo abstract categories such assports or games Although the current VR-CR interventionrequired the children to abstract qualities above the salientor perceptual the demands of the current task were demon-strated to be within the childrenrsquos range of capabilities

The control test was used to verify that the VR-CRintervention could specifically target contextual processingof objects rather than overall cognitive functioning Onthe Attention Sustained Subtest three of the four childrenshowed no changes over the course of the study supportingthe specificity of the intervention However one child Child3 showed significant improvements on one component of thesubtest from pretraining to posttraining Child 3 improved by6 points on the correct responses score which resulted in a10ndash13 increase in his ranked percentile This change is sig-nificant and may have been influenced by a couple of factorsincluding maturational effects and the level of rapport withinthe child-instructor therapeutic relationship Overall basedon the results of the control test the VR-CR interventiondemonstrated specificity to contextual processing for three offour of the children

The final objective of the pilot study was to explorechanges in context-related behaviours that may haveoccurred during the course of the study The commentsobtained by parents on the Final Feedback Questionnairewere mixed Child 1 and Child 4rsquos parents reported nobehavioural changes that occurred between the start andcompletion of the study Child 2 and Child 3rsquos mothersreported changes in the category of appropriate languageand communication in social contexts However it is notedthat both Child 2 and Child 3 were actively engaged inweekly social activities such as swimming and horsebackriding while Child 1 and Child 4 were not Thus the positivebehavioural changes reported are likely due to factors outsideof the intervention itself

The major limitation of this study is the high baselinelevels achieved by all the children prior to intervention Thissuggests that the participating children were already showinghigh levels of contextual processing ability before the study

particularly in abstraction Assessing performance on theVR-test prior to enrolment would help to exclude childrenfrom the study who are already performing at high levels onthe task In addition determining language levels or verbalintelligence at baseline may help to clarify those childrenwho would benefit maximally from this type of cognitiveintervention

A second limitation of this study is the lack of multipleindependent assessors The degree of therapeutic effective-ness and test performance was likely influenced by the devel-opment of rapport between the researcher and each child Tominimize this effect of researcher bias in the current studythere were no subjective rating scales involved In additionscoring of the VR test was verified through computer records

5 Conclusion

This pilot study evaluated a novel intervention that com-bined virtual reality technology with traditional cognitiverehabilitation methods to address impairments in contextualprocessing of objects in children with autism All childrenwho participated in the VR-CR program demonstratedstatistical improvements in overall contextual processingability and improvements in cognitive flexibility How-ever a larger-scale study will need to evaluate the mag-nitude of change resulting from the intervention and thedegree that cognitive improvements generalize to behaviourchanges

Overall this pilot study provides initial supporting evi-dence for the efficacy of this particular VR-CR interventionContinued research may further generate support for thebroader use of the VR-CR approach with children withautismThe hope is that comprehensive interventions will bedeveloped to target both cognitive and behavioural levels ofimpairment in children with autismThis may lead to greateroverall improvements in the daily functioning and quality oflife of these children

Conflict of Interests

The authors declare no conflict of interests

Acknowledgments

This study has been supported by the following Grantsawarded to the first author Canadian Institutes in HealthResearch 2008-09Masterrsquos Award Hilda andWilliam Court-neyClayton Paediatric ResearchAward 2008-09 andOntarioCouncil of Graduate Studies Autism Scholarrsquos Award 2009-10 This paper was prepared from a Masters thesis by thefirst author The authors give thanks to Bloorview KidsRehab Autism Ontario and the Geneva Centre for helpwith recruitment Special thanks to Fan (Mimi) Zhang fortechnical support and Luigi Girolametto and Tom Chaufor their mentorship and valuable feedback throughout theproject

8 The Scientific World Journal

References

[1] American Psychology Association Diagnostic and StatisticalManual ofMental Disorders DSM-IV-TR American PsychiatricAssociation Washington DC USA 4th edition 2000

[2] C Rice ldquoPrevalence of autism spectrum disordersmdashautism anddevelopmental disabilities monitoring network United States2006rdquo Morbidity and Mortality Weekly Report vol 58 no SS-10 pp 1ndash20 2009

[3] P A Filipek P J Accardo G T Baranek et al ldquoThe screeningand diagnosis of autistic spectrum disordersrdquo Journal of Autismand Developmental Disorders vol 29 pp 439ndash484 1999

[4] A Rizzo and J Buckwater ldquoVirtual reality and cognitiveassessment and rehabilitation the state of the artrdquo in VirtualReality in Neuro-Psycho-Physiology G Riva Ed vol 44 pp123ndash146 IOS Press Washington DC USA 1997

[5] R W Butler ldquoCognitive rehabilitationrdquo in Pediatric Neuropsy-chological Intervention S J Hunter and J Donders EdsCambridge University Press New York NY USA 2007

[6] A A RizzoM Schultheis K A Kerns andCMateer ldquoAnalysisof assets for virtual reality applications in neuropsychologyrdquoNeuropsychological Rehabilitation vol 14 no 1 pp 207ndash2392004

[7] G Burdea and P CoiffetVirtual Reality Technology JohnWileyamp Sons Hoboken NJ USA 2003

[8] BGWitmer C J Jerome andM J Singer ldquoThe factor structureof the presence questionnairerdquo Presence vol 14 no 3 pp 298ndash312 2005

[9] F D Rose E A Attree B M Brooks D M Parslow P RPenn and N Ambihaipahan ldquoTraining in virtual environ-ments transfer to real world tasks and equivalence to real tasktrainingrdquo Ergonomics vol 43 no 4 pp 494ndash511 2000

[10] D Strickland L M Marcus G B Mesibov and K HoganldquoBrief report two case studies using virtual reality as a learningtool for autistic childrenrdquo Journal of Autism and DevelopmentalDisorders vol 26 no 6 pp 651ndash659 1996

[11] P Mitchell S Parsons and A Leonard ldquoUsing virtual environ-ments for teaching social understanding to 6 adolescents withautistic spectrum disordersrdquo Journal of Autism and Develop-mental Disorders vol 37 no 3 pp 589ndash600 2007

[12] T Self R R Scudder G Weheba and D Crumrine ldquoA virtualapproach to teaching safety skills to children with autismspectrum disorderrdquo Topics in Language Disorders vol 27 no 3pp 242ndash253 2007

[13] M Wang and D Reid ldquoThe virtual reality-cognitive rehabili-tation (VR-CR) approach for children with autismrdquo Journal ofCyberTherapy and Rehabilitation vol 2 no 2 pp 95ndash102 2009

[14] M Bar ldquoVisual objects in contextrdquo Nature Reviews Neuro-science vol 5 no 8 pp 617ndash629 2004

[15] A Blaye and F Bonthoux ldquoThematic and taxonomic relationsin preschoolers the development of flexibility in categorizationchoicesrdquoBritish Journal of Developmental Psychology vol 19 no3 pp 395ndash412 2001

[16] D Ropar and D Peebles ldquoSorting preference in children withautism the dominance of concrete featuresrdquo Journal of Autismand Developmental Disorders vol 37 no 2 pp 270ndash280 2007

[17] L G Klinger and G Dawson ldquoPrototype formation in autismrdquoDevelopment and Psychopathology vol 13 no 1 pp 111ndash1242001

[18] T Jolliffe and S Baron-Cohen ldquoA test of central coherencetheory can adults with high-functioning autism or Asperger

syndrome integrate objects in contextrdquo Visual Cognition vol8 no 1 pp 67ndash101 2001

[19] E A Berg ldquoA simple objective technique formeasuring flexibil-ity in thinkingrdquo Journal of General Psychology vol 39 pp 15ndash221948

[20] D A Grant and E Berg ldquoA behavioral analysis of degree ofreinforcement and ease of shifting to new responses in aWeigl-type card-sorting problemrdquo Journal of Experimental Psychologyvol 38 no 4 pp 404ndash411 1948

[21] N Kaland L Smith and E L Mortensen ldquoBrief reportcognitive flexibility and focused attention in children and ado-lescents with Asperger syndrome or high-functioning autism asmeasured on the computerized version of the Wisconsin cardsorting testrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1161ndash1165 2008

[22] N J Minshew J Meyer and G Goldstein ldquoAbstract reasoningin autism a dissociation between concept formation andconcept identificationrdquo Neuropsychology vol 16 no 3 pp 327ndash334 2002

[23] S Ozonoff Specifying the Nature of the Primary Deficit inAutism an Investigation of Neuropsychological and Social-Cognitive Abilities in High-Functioning Autistic Children Uni-versity of Denver Denver Colo USA edition 1991

[24] M Prior and W Hoffmann ldquoBrief report neuropsychologicaltesting of autistic children through an exploration with frontallobe testsrdquo Journal of Autism and Developmental Disorders vol20 no 4 pp 581ndash590 1990

[25] E Schopler R J Reichler and B R Renner The ChildhoodAutism Rating Scale (CARS) Western Psychological ServicesLos Angeles Calif USA 1988

[26] G H Roid and L J Miller Leiter International PerformanceScale Revised Stoelting Wood Dale Ill USA 1997

[27] S Jacques The Roles of Labelling and Abstraction in the Devel-opment of Cognitive Flexibility University of Toronto TorontoCanada 2001

[28] H J C Berger F H T M Aerts K P M van SpaendonckA R Cools and J Teunisse ldquoCentral coherence and cognitiveshifting in relation to social improvement in high-functioningyoung adults with autismrdquo Journal of Clinical and ExperimentalNeuropsychology vol 25 no 4 pp 502ndash511 2003

[29] H J C Berger K P M van Spaendonck M W I MHorstink E L Buytenhuijs P W J M Lammers and A RCools ldquoCognitive shifting as a predictor of progress in socialunderstanding in high-functioning adolescents with autisma prospective studyrdquo Journal of Autism and DevelopmentalDisorders vol 23 no 2 pp 341ndash359 1993

[30] F Happe and U Frith ldquoThe weak coherence account detail-focused cognitive style in autism spectrum disordersrdquo Journalof Autism and Developmental Disorders vol 36 no 1 pp 5ndash252006

[31] B R Lopez A J Lincoln S Ozonoff and Z Lai ldquoExaminingthe relationship between executive functions and restrictedrepetitive symptoms ofAutisticDisorderrdquo Journal of Autism andDevelopmental Disorders vol 35 no 4 pp 445ndash460 2005

[32] S S Sparrow D A Balla andD V CicchettiVineland AdaptiveBehavior Scales Interview Edition Survey form Manual Amer-ican Guidance Service Circle Pines Minn USA 1984

[33] T E Scruggs and M A Mastropieri ldquoSummarizing single-subject research Issues and applicationsrdquo Behavior Modifica-tion vol 22 no 3 pp 221ndash242 1998

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

8 The Scientific World Journal

References

[1] American Psychology Association Diagnostic and StatisticalManual ofMental Disorders DSM-IV-TR American PsychiatricAssociation Washington DC USA 4th edition 2000

[2] C Rice ldquoPrevalence of autism spectrum disordersmdashautism anddevelopmental disabilities monitoring network United States2006rdquo Morbidity and Mortality Weekly Report vol 58 no SS-10 pp 1ndash20 2009

[3] P A Filipek P J Accardo G T Baranek et al ldquoThe screeningand diagnosis of autistic spectrum disordersrdquo Journal of Autismand Developmental Disorders vol 29 pp 439ndash484 1999

[4] A Rizzo and J Buckwater ldquoVirtual reality and cognitiveassessment and rehabilitation the state of the artrdquo in VirtualReality in Neuro-Psycho-Physiology G Riva Ed vol 44 pp123ndash146 IOS Press Washington DC USA 1997

[5] R W Butler ldquoCognitive rehabilitationrdquo in Pediatric Neuropsy-chological Intervention S J Hunter and J Donders EdsCambridge University Press New York NY USA 2007

[6] A A RizzoM Schultheis K A Kerns andCMateer ldquoAnalysisof assets for virtual reality applications in neuropsychologyrdquoNeuropsychological Rehabilitation vol 14 no 1 pp 207ndash2392004

[7] G Burdea and P CoiffetVirtual Reality Technology JohnWileyamp Sons Hoboken NJ USA 2003

[8] BGWitmer C J Jerome andM J Singer ldquoThe factor structureof the presence questionnairerdquo Presence vol 14 no 3 pp 298ndash312 2005

[9] F D Rose E A Attree B M Brooks D M Parslow P RPenn and N Ambihaipahan ldquoTraining in virtual environ-ments transfer to real world tasks and equivalence to real tasktrainingrdquo Ergonomics vol 43 no 4 pp 494ndash511 2000

[10] D Strickland L M Marcus G B Mesibov and K HoganldquoBrief report two case studies using virtual reality as a learningtool for autistic childrenrdquo Journal of Autism and DevelopmentalDisorders vol 26 no 6 pp 651ndash659 1996

[11] P Mitchell S Parsons and A Leonard ldquoUsing virtual environ-ments for teaching social understanding to 6 adolescents withautistic spectrum disordersrdquo Journal of Autism and Develop-mental Disorders vol 37 no 3 pp 589ndash600 2007

[12] T Self R R Scudder G Weheba and D Crumrine ldquoA virtualapproach to teaching safety skills to children with autismspectrum disorderrdquo Topics in Language Disorders vol 27 no 3pp 242ndash253 2007

[13] M Wang and D Reid ldquoThe virtual reality-cognitive rehabili-tation (VR-CR) approach for children with autismrdquo Journal ofCyberTherapy and Rehabilitation vol 2 no 2 pp 95ndash102 2009

[14] M Bar ldquoVisual objects in contextrdquo Nature Reviews Neuro-science vol 5 no 8 pp 617ndash629 2004

[15] A Blaye and F Bonthoux ldquoThematic and taxonomic relationsin preschoolers the development of flexibility in categorizationchoicesrdquoBritish Journal of Developmental Psychology vol 19 no3 pp 395ndash412 2001

[16] D Ropar and D Peebles ldquoSorting preference in children withautism the dominance of concrete featuresrdquo Journal of Autismand Developmental Disorders vol 37 no 2 pp 270ndash280 2007

[17] L G Klinger and G Dawson ldquoPrototype formation in autismrdquoDevelopment and Psychopathology vol 13 no 1 pp 111ndash1242001

[18] T Jolliffe and S Baron-Cohen ldquoA test of central coherencetheory can adults with high-functioning autism or Asperger

syndrome integrate objects in contextrdquo Visual Cognition vol8 no 1 pp 67ndash101 2001

[19] E A Berg ldquoA simple objective technique formeasuring flexibil-ity in thinkingrdquo Journal of General Psychology vol 39 pp 15ndash221948

[20] D A Grant and E Berg ldquoA behavioral analysis of degree ofreinforcement and ease of shifting to new responses in aWeigl-type card-sorting problemrdquo Journal of Experimental Psychologyvol 38 no 4 pp 404ndash411 1948

[21] N Kaland L Smith and E L Mortensen ldquoBrief reportcognitive flexibility and focused attention in children and ado-lescents with Asperger syndrome or high-functioning autism asmeasured on the computerized version of the Wisconsin cardsorting testrdquo Journal of Autism and Developmental Disordersvol 38 no 6 pp 1161ndash1165 2008

[22] N J Minshew J Meyer and G Goldstein ldquoAbstract reasoningin autism a dissociation between concept formation andconcept identificationrdquo Neuropsychology vol 16 no 3 pp 327ndash334 2002

[23] S Ozonoff Specifying the Nature of the Primary Deficit inAutism an Investigation of Neuropsychological and Social-Cognitive Abilities in High-Functioning Autistic Children Uni-versity of Denver Denver Colo USA edition 1991

[24] M Prior and W Hoffmann ldquoBrief report neuropsychologicaltesting of autistic children through an exploration with frontallobe testsrdquo Journal of Autism and Developmental Disorders vol20 no 4 pp 581ndash590 1990

[25] E Schopler R J Reichler and B R Renner The ChildhoodAutism Rating Scale (CARS) Western Psychological ServicesLos Angeles Calif USA 1988

[26] G H Roid and L J Miller Leiter International PerformanceScale Revised Stoelting Wood Dale Ill USA 1997

[27] S Jacques The Roles of Labelling and Abstraction in the Devel-opment of Cognitive Flexibility University of Toronto TorontoCanada 2001

[28] H J C Berger F H T M Aerts K P M van SpaendonckA R Cools and J Teunisse ldquoCentral coherence and cognitiveshifting in relation to social improvement in high-functioningyoung adults with autismrdquo Journal of Clinical and ExperimentalNeuropsychology vol 25 no 4 pp 502ndash511 2003

[29] H J C Berger K P M van Spaendonck M W I MHorstink E L Buytenhuijs P W J M Lammers and A RCools ldquoCognitive shifting as a predictor of progress in socialunderstanding in high-functioning adolescents with autisma prospective studyrdquo Journal of Autism and DevelopmentalDisorders vol 23 no 2 pp 341ndash359 1993

[30] F Happe and U Frith ldquoThe weak coherence account detail-focused cognitive style in autism spectrum disordersrdquo Journalof Autism and Developmental Disorders vol 36 no 1 pp 5ndash252006

[31] B R Lopez A J Lincoln S Ozonoff and Z Lai ldquoExaminingthe relationship between executive functions and restrictedrepetitive symptoms ofAutisticDisorderrdquo Journal of Autism andDevelopmental Disorders vol 35 no 4 pp 445ndash460 2005

[32] S S Sparrow D A Balla andD V CicchettiVineland AdaptiveBehavior Scales Interview Edition Survey form Manual Amer-ican Guidance Service Circle Pines Minn USA 1984

[33] T E Scruggs and M A Mastropieri ldquoSummarizing single-subject research Issues and applicationsrdquo Behavior Modifica-tion vol 22 no 3 pp 221ndash242 1998

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

The Scientific World Journal 9

[34] L R Logan R R Hickman S R Harris and C B HerizaldquoSingle-subject research design recommendations for levels ofevidence and quality ratingrdquoDevelopmentalMedicine and ChildNeurology vol 50 no 2 pp 99ndash103 2008

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

MEDIATORSINFLAMMATION

of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Behavioural Neurology

EndocrinologyInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Disease Markers

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

BioMed Research International

OncologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Oxidative Medicine and Cellular Longevity

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Computational and Mathematical Methods in Medicine

OphthalmologyJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Diabetes ResearchJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Research and TreatmentAIDS

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Gastroenterology Research and Practice

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom