Debbie RandDepartment of OccupationalTherapyUniversity of HaifaHaifa Israel

Rachel KizonyDepartment of OccupationalTherapyUniversity of HaifaHaifa Israelamp School of Occupational TherapyHadassah-Hebrew UniversityJerusalem Israel

Uri FeintuchSchool of Occupational TherapyHadassah-Hebrew UniversityJerusalem Israelamp Caesarea-Rothschild Institute forInterdisciplinary Applications ofComputer ScienceUniversity of HaifaHaifa Israel

Noomi KatzSchool of Occupational TherapyHadassah-Hebrew UniversityJerusalem Israel

Naomi JosmanDepartment of OccupationalTherapyUniversity of HaifaHaifa Israel

Albert ldquoSkiprdquo RizzoIntegrated Media SystemsUniversity of Southern CaliforniaLos Angeles California

Patrice L (Tamar) WeissDepartment of OccupationalTherapyUniversity of HaifaHaifa Israeltamarresearchhaifaacil

Presence Vol 14 No 2 April 2005 147ndash160

copy 2005 by the Massachusetts Institute of Technology

Comparison of Two VR Platformsfor Rehabilitation Video Captureversus HMD

Abstract

In recent years clinical studies have begun to demonstrate the effectiveness of VRas an intervention tool for a variety of neurological conditions There remain how-ever a number of important issues that must be addressed in order to determinehow widely VR-based intervention should be applied and the user and platformcharacteristics that may be important when using VR in clinical settings One of theunresolved issues that must be addressed is the suitability of particular VR platformsin relation to the therapeutic goals one wishes to achieve Studying and identifyingthe characteristics of each platform may assist the therapist in choosing a suitableVR platform for the patientrsquos needs The purpose of this paper is to describe theresults of a study of healthy participants (N 89) using 2 different VR platforms incombination with 1 of the 2 virtual environments that was designed to comparethe sense of presence incidence of side effects perceived exertion and perfor-mance The data demonstrate significant differences in some of the key characteris-tics of both VR platforms and environments as they affect participantsrsquo sense ofpresence performance side effects and exertion We conclude that when seekinga suitable VR therapeutic application the userrsquos characteristics together with at-tributes of the VR platform must be taken into consideration since both appear tohave an impact on key outcome measures

1 Introduction

Rehabilitation of patients with neurological deficits resulting from strokehead trauma spinal cord injury or degenerative diseases such as multiple scle-rosis is a complex process since they typically suffer from serious motor andcognitive impairments These patients require extensive periods of rehabilita-tion in order to minimize the impairments resulting from their neurologicaldeficits and to improve their functional level in the performance of activities ofdaily living as well as enable them to participate in community life Thus anessential part of the rehabilitation process is remediation of cognitive and mo-tor deficits in order to improve the functional ability of the patient and to en-able him or her to achieve greater independence It is thought that the func-tional relevance of therapeutic intervention is of paramount importance(AOTA 2002) Moreover the improvement in function must go beyond whatis achieved in the clinical setting that is it must be transferred to real-life ac-tivities Unfortunately such objectives are often difficult if not impossible toachieve via conventional therapy

Rand et al 147

In recent years clinical studies have begun to demon-strate the effectiveness of VR as an intervention tool fora variety of neurological conditions (eg Jack et al2001 Schultheis amp Rizzo 2001 Weiss Naveh amp Katz2003) There remains however a number of importantissues that must be addressed in order to determine howwidely VR-based intervention should be applied andthe ways in which specific patient populations can bene-fit from its unique attributes One of the unresolvedissues that must be addressed is the suitability of partic-ular VR platforms in relation to the therapeutic goalsone wishes to achieve Given the growing options avail-able on todayrsquos market including numerous flat-screensystems a variety of head-mounted-display models(HMDs) and projection-based platforms it has becomeimportant for clinicians who are considering the use ofVR to identify both the assets and limitations of eachapproach Some platform comparisons are relativelystraightforward (eg size cost ease of use) and dependprimarily on the availability of space budget and tech-nical support However other comparisons are morecomplex and involve the characteristics of the systemsand interactions between these characteristics and theirinfluence on the user

To date information regarding the characteristics ofthe VR system attributes that may be of major therapeu-tic importance is scarce and there are a number of openquestions Does the donning of external devices such asgoggles gloves or helmets disturb patient performancedue to weight discomfort or isolation What effectdoes the way in which the patient is represented in avirtual environment (eg as an avatar in ldquofirstrdquo personin ldquothirdrdquo person) have on the sense of presence and onperformance What effect does the way in which thepatient navigates or manipulates objects within a virtualenvironment (eg via a tracker joystick natural limbmovements) have on the sense of presence and on per-formance With what frequency or severity does a par-ticular VR platform andor environment lead to cyber-sickness-like symptoms Are certain segments of thepopulation (eg men versus women elderly versusyoung specific pathologies) more susceptible to suchsymptoms

The answers to these and other questions could guide

clinical decision making as to which platform might bebest suited for specific assessment and rehabilitationtasks with different patient populations The purpose ofthis study was to compare the sense of presence level ofperformance side effects and perceived exertion ofhealthy participants who played virtual games orscanned a virtual office displayed via two different VRplatforms The effect of age and gender on these vari-ables was also examined This comparison with healthyparticipants was designed to provide important informa-tion about the relative assets and limitations of the twoVR platforms so that the clinician may justify their usefor therapeutic purposes

2 Method

21 Participants

The study sample consisted of 89 healthy volun-teer participants Seventy-three participants were aged16 to 35 years and 16 participants were aged 60 to 75years As illustrated in Table 1 the 16 to 35 year oldparticipants were divided into two primary groups onegroup (n 40) experienced three virtual games on twodifferent VR platforms (Gesture Xtreme [GX]-monitorand GX-Head Mounted Display [HMD] and the othergroup (n 33) experienced the Virtual Office via a GX-monitor versus an HMD platform The 16 older partici-pants all experienced the Virtual Office on the two dif-ferent platforms

22 VR Platforms

Two VR platforms that differed considerably intheir features were used in this study along with ahybrid platform that combined features from themboth

VividGrouprsquos Gesture Xtreme (GX) VR (wwwvividgroupcom) is a projected video-capture VR plat-form originally developed for entertainment purposesthat has been adapted for use in rehabilitation (KizonyKatz amp Weiss 2003) Participants stand or sit in a de-marcated area viewing a large monitor that displays agame or functional tasks such as touching virtual balls

148 PRESENCE VOLUME 14 NUMBER 2

In recent years clinical studies have begun to demon-strate the effectiveness of VR as an intervention tool fora variety of neurological conditions (eg Jack et al2001 Schultheis amp Rizzo 2001 Weiss Naveh amp Katz2003) There remains however a number of importantissues that must be addressed in order to determine howwidely VR-based intervention should be applied andthe ways in which specific patient populations can bene-fit from its unique attributes One of the unresolvedissues that must be addressed is the suitability of partic-ular VR platforms in relation to the therapeutic goalsone wishes to achieve Given the growing options avail-able on todayrsquos market including numerous flat-screensystems a variety of head-mounted-display models(HMDs) and projection-based platforms it has becomeimportant for clinicians who are considering the use ofVR to identify both the assets and limitations of eachapproach Some platform comparisons are relativelystraightforward (eg size cost ease of use) and dependprimarily on the availability of space budget and tech-nical support However other comparisons are morecomplex and involve the characteristics of the systemsand interactions between these characteristics and theirinfluence on the user

To date information regarding the characteristics ofthe VR system attributes that may be of major therapeu-tic importance is scarce and there are a number of openquestions Does the donning of external devices such asgoggles gloves or helmets disturb patient performancedue to weight discomfort or isolation What effectdoes the way in which the patient is represented in avirtual environment (eg as an avatar in ldquofirstrdquo personin ldquothirdrdquo person) have on the sense of presence and onperformance What effect does the way in which thepatient navigates or manipulates objects within a virtualenvironment (eg via a tracker joystick natural limbmovements) have on the sense of presence and on per-formance With what frequency or severity does a par-ticular VR platform andor environment lead to cyber-sickness-like symptoms Are certain segments of thepopulation (eg men versus women elderly versusyoung specific pathologies) more susceptible to suchsymptoms

The answers to these and other questions could guide

clinical decision making as to which platform might bebest suited for specific assessment and rehabilitationtasks with different patient populations The purpose ofthis study was to compare the sense of presence level ofperformance side effects and perceived exertion ofhealthy participants who played virtual games orscanned a virtual office displayed via two different VRplatforms The effect of age and gender on these vari-ables was also examined This comparison with healthyparticipants was designed to provide important informa-tion about the relative assets and limitations of the twoVR platforms so that the clinician may justify their usefor therapeutic purposes

2 Method

21 Participants

The study sample consisted of 89 healthy volun-teer participants Seventy-three participants were aged16 to 35 years and 16 participants were aged 60 to 75years As illustrated in Table 1 the 16 to 35 year oldparticipants were divided into two primary groups onegroup (n 40) experienced three virtual games on twodifferent VR platforms (Gesture Xtreme [GX]-monitorand GX-Head Mounted Display [HMD] and the othergroup (n 33) experienced the Virtual Office via a GX-monitor versus an HMD platform The 16 older partici-pants all experienced the Virtual Office on the two dif-ferent platforms

22 VR Platforms

Two VR platforms that differed considerably intheir features were used in this study along with ahybrid platform that combined features from themboth

VividGrouprsquos Gesture Xtreme (GX) VR (wwwvividgroupcom) is a projected video-capture VR plat-form originally developed for entertainment purposesthat has been adapted for use in rehabilitation (KizonyKatz amp Weiss 2003) Participants stand or sit in a de-marcated area viewing a large monitor that displays agame or functional tasks such as touching virtual balls

148 PRESENCE VOLUME 14 NUMBER 2

A single-camera vision-based tracking system capturesand converts the userrsquos movements for processing theuserrsquos live on-screen video image corresponds in realtime to his movements The users can interact withgraphical objects as depicted in this environment Theorigins of this work can be traced back to Kruegerrsquosseminal ldquoVideoplacerdquo application in the early 1970swhere it was observed that humans felt compelled tointeract with graphic objects displayed in this format

In the 1990s VividGroup designed and marketed aseries of single-camera vision-based applications aslocation-based arcade entertainment systems Known asthe Gesture Xtreme VR System it uses a blue backdropand a chroma key to separate the userrsquos image from thebackground This system has now come to be embracedby rehabilitation specialists as a research and clinical toolfor the treatment of motor and cognitive impairments(eg Kizony et al 2003 Kizony Raz Katz Weingar-den amp Weiss 2003 Sveistrup et al 2003 Weiss et al2003) Its advantages include the fact that patients seethemselves rather than being represented as an avatarThey do not have to wear special apparatus such as anHMD which encourages the use of active movementand reduces their chances of experiencing side effectsMoreover the user is not isolated from the real worldand the therapist is able to readily intervene during thesession in order to support and guide the patientrsquosmovements Despite these many advantages GX VR isstill not widely used in rehabilitation facilities One ofthe major reasons for this limited usage is the cost of thesystem which keeps it out of the price range of manyclinical settings

Head Mounted Display Platform (Fifth DimensionTechnologies www5dtcom) HMD 800 unit is alightweight (594 g) head-mounted display The user isisolated from the real world while his viewpoint is ldquofirstpersonrdquo he or she manipulates objects within the virtualenvironment via Intersensersquos (wwwisensecom) Inter-Trax2 a 3-degree-of-freedom inertial orientationtracker used to track pitch roll and yaw movements ofthe head The user navigates via two hand-held switchesthat enable him to move to the left or to the rightwithin the virtual environment

A combination of the two VR platforms that integratedfeatures from the two platforms described above Thishybrid approach used the 5DT HMD to display the GXVR games scenarios to users The goal of combiningfeatures from the two platforms was to strengthen theattributes of each (the natural movement from the GX-monitor with the isolation created by the HMD) Thefeatures of the two different VR platforms as well as thehybrid platform are listed in Table 2

23 Virtual Environments

Three of the virtual environments (referred to hereas ldquogamesrdquo) are normally used with the GX-monitor VRplatform and have been described in detail elsewhere(Kizony et al 2003) Briefly they include

1 Birds amp Ballsmdashwherein the user sees himselfstanding in a pastoral setting where balls of differentcolors emerge from peripheral locations and fly to-ward the user Depending on the intensity of contact

Table 1 Distribution of Participants in the Study

Virtualenvironment

Virtual games(N 40) Virtual office (N 49)

Age group Young Young Elderly

Mean (SD) age 245 44 221 30 666 40

Gender Male Female Male Female Male FemaleN 24 16 13 20 11 5

Rand et al 149

Table 2 Characteristics of the VR Platforms as Tested

Feature GX-monitor HMD GX-HMD

Description User faces large monitor(34) while videocamera recordsmovements in 2Dworld

Wearing an HMD usernavigates in a 3D virtualworld

Wearing an HMD users see imagesof their own movements in 2Dworld that the video cameracaptures

Virtualenvironments

Three games Birds ampBalls SoccerSnowboard OR theldquoVirtual Officerdquo

ldquoVirtual Officerdquo Three games Birds amp Balls SoccerSnowboard

Encumbrance No external attachmentsto user

User wears HMD navigateswith hand-held switches

User wears an HMD

Point of view User sees himself as amirror image (not anavatar)

Userrsquos viewpoint is ldquofirstpersonrdquo

User sees himself as a mirror imagewearing an HMD

Interaction withenvironment

User can use any part ofbody to interact withthe environment butmainly his hands

User can look around (full360deg) while navigatingwith two hand-heldswitches

User can use any part of body tointeract with the environmentbut mainly his hands

Body movement inspace

User can move freelymdashforward back leftright

Standing or sitting user cantake a step or two mustbe careful not to trip overHMD wires

User can move around must becareful not to trip over HMDwires

Record ofmovement

Whole-body movement isrecorded on videotape

Only head movement istracked and recorded

Whole-body movement is recordedon videotape

Isolation Moderate Considerable ConsiderableFeedback Visual auditory Visual Visual auditorySide effects Anticipated to be minimal Anticipated to be minimal

to moderate due to theHMD

Anticipated to be moderate due toincongruence betweenenvironmental cues and headmovement

Treatment goals To train both motor andcognitive aspects whileengaged in a purposefulactivity

Primarily to train cognitiveaspects (eg visualscanning or visualmemory)

To train both motor and cognitiveaspects while engaged in apurposeful activity and whileisolated from real world

VR leading toparticipation

Improved performancefor daily tasks thatrequire handmovement balanceand visual search

Improved performance fordaily tasks that requirevisual search memoryetc

Improved performance for dailytasks that require handmovement balance and visualsearch

150 PRESENCE VOLUME 14 NUMBER 2

by any part of the userrsquos body the balls will eitherldquoburstrdquo or ldquotransformrdquo into doves and fly away Per-formance was rated by the mean response time (RT)of touching the balls

2 Soccermdashwherein the user sees a video reflectionhimself as the goalkeeper in a soccer game Soccerballs are shot at him from different locations and histask is to hit them with different parts of his video-represented body in order to prevent them from en-tering the goal area Performance was rated by thepercent success of repelling the balls

3 Snowboardmdashwherein the user sees a video-represented back view of himself mounted on a snow-board As he skis downhill he needs to avoid obstaclesby leaning from side to side or by moving his wholebody Performance was rated by percent success ofobstacle avoidance

For all these games the 3rd minute (out of a total of4 minutes) of each VR experience was analyzed since itshould reflect the participantrsquos best performance that isafter participants had practiced but prior to the onset offatigue

The three games were viewed by the participants us-ing the GX video-projected VR platform via an enlargedmonitor (referred to as GX-monitor) and also with theHMD (referred to as GX-HMD) In this case an HMD

is worn so that the displays of the three virtual games(Birds amp Balls Soccer Snowboard) are viewed via thisdisplay rather than on the normal enlarged monitor Inboth cases interaction with the VE is the same the par-ticipants use different parts of the body to interact withthe graphic stimuli

A fourth environment is the ldquoVirtual Officerdquo whichwas developed for use with an HMD by Rizzo Schul-thesis amp Rothbaum (2002b) at the University ofSouthern Californiarsquos Integrated Media Systems Center(httpimscuscedu) The task of the participants inthe Virtual Office as used in this study was to visuallyscan the office environment for at least 2 min and nolonger than 5 min In the office there are 16 differentitems 8 of them are uncommon to a typical office (ega dog a fire hydrant) whereas the other items are typi-cal of an office (eg a clock a clipboard) Performancewas measured as the number of objects (out of the 16-item list) that participants were able to name after theyfinished scanning the office The number of backgrounditems (eg door window desk) that were named wasalso recorded

The Virtual Office was used in its traditional way withthe HMD but was also converted by us to be used viathe GX-monitor platform In this case participantsstand in front of the GX monitor and visually scan theVirtual Office (see Figure 2) Movement within the of-

Figure 1 Participant using the GX video-capture system for a

virtual game

Figure 2 Participant using the Virtual Office via the GX video-

capture system

Rand et al 151

by any part of the userrsquos body the balls will eitherldquoburstrdquo or ldquotransformrdquo into doves and fly away Per-formance was rated by the mean response time (RT)of touching the balls

2 Soccermdashwherein the user sees a video reflectionhimself as the goalkeeper in a soccer game Soccerballs are shot at him from different locations and histask is to hit them with different parts of his video-represented body in order to prevent them from en-tering the goal area Performance was rated by thepercent success of repelling the balls

3 Snowboardmdashwherein the user sees a video-represented back view of himself mounted on a snow-board As he skis downhill he needs to avoid obstaclesby leaning from side to side or by moving his wholebody Performance was rated by percent success ofobstacle avoidance

For all these games the 3rd minute (out of a total of4 minutes) of each VR experience was analyzed since itshould reflect the participantrsquos best performance that isafter participants had practiced but prior to the onset offatigue

The three games were viewed by the participants us-ing the GX video-projected VR platform via an enlargedmonitor (referred to as GX-monitor) and also with theHMD (referred to as GX-HMD) In this case an HMD

is worn so that the displays of the three virtual games(Birds amp Balls Soccer Snowboard) are viewed via thisdisplay rather than on the normal enlarged monitor Inboth cases interaction with the VE is the same the par-ticipants use different parts of the body to interact withthe graphic stimuli

A fourth environment is the ldquoVirtual Officerdquo whichwas developed for use with an HMD by Rizzo Schul-thesis amp Rothbaum (2002b) at the University ofSouthern Californiarsquos Integrated Media Systems Center(httpimscuscedu) The task of the participants inthe Virtual Office as used in this study was to visuallyscan the office environment for at least 2 min and nolonger than 5 min In the office there are 16 differentitems 8 of them are uncommon to a typical office (ega dog a fire hydrant) whereas the other items are typi-cal of an office (eg a clock a clipboard) Performancewas measured as the number of objects (out of the 16-item list) that participants were able to name after theyfinished scanning the office The number of backgrounditems (eg door window desk) that were named wasalso recorded

The Virtual Office was used in its traditional way withthe HMD but was also converted by us to be used viathe GX-monitor platform In this case participantsstand in front of the GX monitor and visually scan theVirtual Office (see Figure 2) Movement within the of-

Figure 1 Participant using the GX video-capture system for a

virtual game

Figure 2 Participant using the Virtual Office via the GX video-

capture system

Rand et al 151

fice is accomplished by ldquotouchingrdquo one of two virtualarrows displayed on the screen Touching the left arrowcauses the environment to move to the left thus en-abling the participant to see what is on the left side ofthe office while touching the right arrow reveals theright side of the office The image of the office usedwith the Vivid platform is a 360deg photo-surround ofRizzo Bowerly et alrsquos (2002a) Virtual Office 16 dif-ferent but conceptually similar (eg a ldquocatrdquo instead ofa ldquodogrdquo) objects replace the 16 objects that populatethe original HMD-version of the office in the counter-balanced condition

24 Outcome Measures

A Presence Questionnaire (PQ) (translated fromWitmer amp Singer 1998) was used to assess presence Itis composed of 19 questions in which participants use a7-point scale to rate various experiences within the VEthe maximum total score is 133 points The items as-sessed different aspects of presence involvementcon-trol natural interface quality and resolution

A Scenario Feedback Questionnaire (SFQ) (basedin part on a translated version of Witmer and Sing-errsquos [1998] Presence Questionnaire) was adminis-tered after every environment The 6 items assessedthe participantrsquos (1) feeling of enjoyment (2) senseof being in the environment (3) success (4) control(5) perception of the environment as being realisticand (6) whether the feedback from the computer wasunderstandable Responses to all questions were ratedon a scale of 1ndash5 These questions were combined togive a global response to the experience for a maxi-mum score of 30 This 6-item questionnaire was for-mulated as an abbreviated alternative to the longerPresence Questionnaire The participants were alsoasked whether they felt any discomfort during theexperience

A Simulator Sickness Questionnaire (SSQ) (KennedyLane Berbaum amp Lilienthal 1993) was completed byany participants who reported discomfort (Question 7as indicated above) and at a later stage by all of theparticipants using a 4-point rating scale on a 16-item

symptom checklist for each environment The totalscore was calculated by adding the scores of the threesubscales Nausea Oculomotor and Disorientation andmultiplying the total by 374 as suggested by Kennedyet al The score ranged between 0ndash2244 points

Borgrsquos Scale of Perceived Exertion (Borg 1990) wasused to assess how much physical effort the participantsperceived that they expended during each VR experi-ence This is a 20-point scale that participants ratedfrom 6 (no exertion at all) to 20 (maximal exertion)

25 Procedure

Participants signed an informed consent and werethen assigned to one of two groups One group experi-enced the three virtual games (4 minutes each) usingboth platforms (GX-HMD amp GX-monitor) and theother group experienced the Virtual Office using bothplatforms (HMD amp GX-monitor) (up to 2 minuteseach) the order of platform usage was counterbalancedAfter experiencing each environment participants filledout the SFQ Due to technical problems during the firsthalf of the study only participants who reported feelingcybersickness-like side effects (eg dizziness or nausea)when asked about discomfort were requested to com-plete the SSQ The remaining 31 participants were re-quested to complete the SSQ following their use of eachplatform regardless of their response to the question(Note however that all participants were aware of po-tential VR-related side effects prior to the study sincethey were cautioned about this possibility when theyread and signed the informed-consent form) Partici-pants were also requested to rate their perceived exer-tion using the Borg scale After completing all environ-ments for a given VR platform participants completedthe full Presence Questionnaire

26 Data Analysis

A mixed design within and between subjectsANOVA was used in order to examine the effect of thetype of the VR platform (ie HMD GX-Monitor) andthe user characteristics (ie gender age) as well as the

152 PRESENCE VOLUME 14 NUMBER 2

interaction between these variables on the sense of pres-ence performance and perceived exertion Due to thesmall sample size of the participants who completed theSSQ we used a t-test rather than an ANOVA to explorethe differences between the platforms in the virtualgames and the nonparametric Mann-Whitney test forthe Virtual Office

3 Results

As a first step for each analysis we examinedwhether the order of experiencing the VR platformsinfluenced the results There were no significant differ-ences due to the order in which the VR platforms wereexperienced by participants for any of the outcome mea-sures The results for the virtual games will be presentedfirst followed by the results for the Virtual Office

31 The Three Gesture Xtreme Games

311 Presence and Scenario Feedback Ques-tionnaires Presence Questionnaire (total score)mdashAsignificant main effect was found for the type of plat-form (F(1 37) 376 p 0001) while no significantmain effect was found for gender or for the interactionbetween gender and type of platform The participantsrsquosense of presence was higher when using the GX-monitorplatform than when using the GX-HMD When the PQwas subdivided into its four subscales a significant dif-ference was found for the involvementcontrol subscalebetween the GX-monitor platform and the GX-HMD(t(38) 36 p 001) and for the qualityinterfacesubscale for the whole population (t(38) 55 p

0001) The results of the Presence Questionnaire andother outcome measures for the virtual games are listedin Table 3

Scenario Feedback QuestionnairemdashFor Birds amp Balls andSoccer no significant main effects were found either fortype of platform or for gender nor was the interactionsignificant That is the participants felt similar enjoymentsuccess control and so forth when playing the gameswith a monitor or while wearing an HMD The total SFQ

score for Birds amp Balls was somewhat lower than thosegiven for Soccer For the Snowboard game a significantmain effect was found for gender only (F(1 37) 108p 01) T-tests between males and females within eachplatform revealed that for both platforms scores for themale group were significantly higher than scores for thefemale group (t 35 p 01 for the GX-monitor) and(t 21 p 05 for the GX-HMD)

312 Performance Birds and BallsmdashA signifi-cant main effect was found for the type of the platform(F(1 34) 85 p 01) but not for gender nor wasthe interaction significant The mean response timewhen the participants played Birds amp Balls with the GX-monitor was 37 06 s as compared to 40 07 s forthe GX-HMD

SoccermdashA significant main effect was found only forgender (F(1 35) 257 p 0001) while no significantmain effects were found either for type of platform or forthe interaction between gender and type of platform Thepercent of success in Soccer for male participants was sig-nificantly higher than for female participants when playedwith the GX-monitor (t(36) 44 p 0001) and withthe GX-HMD (t(35) 42 p 0001)

SnowboardmdashNo significant main effect for type ofplatform and gender or the interaction between themwas found In both platforms the participantsrsquo percentof success was high

313 Side Effects Of the 40 participants whoexperienced the virtual games only 22 completed theSSQ for both platforms The differences between thetotal SSQ score for GX-monitor (101 112) was sig-nificantly lower than for the GX-HMD (362 273(t(19) 44 p 0001) These SSQ scores for bothplatforms were very low compared to the maximal SSQof 2244

314 Exertion No significant main effects werefound in the Borg scale either for the type of platformor for gender nor was the interaction significant Thescore ranged from a high of 134 (out of a maximum of20) for Soccer to a low of 83 for Snowboard

Rand et al 153

32 The Virtual Office

321 Presence and Scenario Feedback Ques-tionnaires Presence Questionnaire (total score)mdashAmain effect between participants was found for gender(F(1 45) 76 p 01) and an interaction effect wasfound between age and type of platform (F(1 45)

48 p 05) T-tests were carried out to interpret theinteractions although within the GX-monitor platform

there were no significant differences between the twoage groups when they viewed the Virtual Office via anHMD the older group felt a significantly higher sense ofpresence than did the younger group (t(47) 39 p

0001) Significant differences between males and fe-males were found only for GX-monitor (males 918

151 versus females 825 140) (t(47) 22 p

01) The results of the Presence Questionnaire as well

Table 3 Summary of Results of the Three Virtual Games Played on GX-Monitor and GX-HMD

GX-Monitor GX-HMD

MaleN 24

FemaleN 16

TotalN 40

MaleN 24

FemaleN 16

TotalN 40

PQ total (19ndash133) 991 123 1073 157 1026 142 926 137 973 10 945 124

Involvcont (11ndash77) 595 77 631 95 60 86 576 83 581 65 578 75

Natural (3ndash21) 134 36 158 33 144 36 133 42 162 27 145 4Resolution (2ndash14) 95 23 103 27 10 25 9 25 102 26 94 26Quality (3ndash21) 165 33 18 33 172 33 127 32 126 31 127 31

Scenario FeedbackQuestionnaire (6ndash30)

Birds and Balls 236 38 26 26 246 35 243 35 256 3 248 33Soccer 217 37 224 41 218 38 21 5 216 4 212 45Snowboard 230 40 268 26 245 4 233 51 263 22 246 44

PerformanceBirds and Balls 36 71 39 63 37 06 37 62 43 07 40 07

Soccer 640 10 485 12 578 134 626 120 452 127 401 15

Snowboard 954 65 995 14 97 56 967 6 971 27 97 5SSQ (16ndash2244) 115 124 83 10 101 112 36 282 365 28 362 273

Borg scale perceivedexertion (0ndash20)

Birds and Balls 9 26 91 28 90 26 95 303 98 28 96 3Soccer 13 22 141 22 134 22 131 34 136 28 133 31Snowboard 86 23 81 27 84 25 94 32 9 304 92 31

significantNote differences (t-test or paired t-test)

154 PRESENCE VOLUME 14 NUMBER 2

as other outcome measures for the Virtual Office arelisted in Table 4

When the PQ was divided into its four subscales foreach of the age groups significant differences betweenthe platforms were found for the elderly group for theInterfacequality subscale (t(15) 25 p 05) Dif-ferences in the natural PQ subscale were found betweenthe age groups within the Virtual Office when viewedvia an HMD and GX-monitor whereas the involvementsubscale was found to have a significant difference be-tween the age groups when viewed via the HMD onlyThus the older group had higher scores on the naturalsubscale than did the younger group for both the GX-monitor (t(47) 20 p 05) and for HMD (t(47)

26 p 01) Moreover the older group had a greater

sense of involvementcontrol when using the VirtualOffice with an HMD than did the younger group(t(47) 42 p 0001)

Scenario Feedback QuestionnairemdashA significant maineffect for age was found (F(1 30) 50 p 05) butthere was no significant interaction T-tests were carriedout to analyze the main effect and the results showedthat for the HMD a significant difference was foundbetween the younger and the older participants(t(33) 262 p 05) but no significant differencewas found when using the GX-monitor

322 Performance Visual ScanningmdashA maineffect was found for age (F(1 45) 183 p 0001)and for the type of platform (F(1 45) 117 p

Table 4 Summary of Results of the Virtual OfficemdashGX-Monitor versus HMD

Age group

GX-Monitor HMD

16ndash35 yearsN 33

60ndash75 yearsN 16

16ndash35 yearsN 33

60ndash75 yearsN 16

PQ total (19ndash133) 857 135 898 181 835 127 984 116

Involvementcontrol (11ndash77) 555 78 57 92 524 85 593 84

Natural (3ndash21) 106 44 135 47 109 31 155 42

Resolution (2ndash14) 76 33 906 32 75 31 102 26Interface quality (3ndash21) 12 37 102 35 126 307 133 35

Scenario feedback questionnaire (6ndash30) 238 31 220 36 214 39 251 28

PerformanceVisual scanning 97 28 68 25 82 22 54 15

Naming of background objects 5 22 47 27 55 24 6 203Scan time 36 12 29 113 41 11 41 14

Borg scale of perceived exertion (0ndash20) 912 30 72 14 102 22 75 18

significantNote differences (t-test or paired t-test)

Rand et al 155

001) but no significant interaction effect was foundSignificant differences between age groups were foundfor the number of objects named by the younger partic-ipants when compared to the older participants for boththe GX-monitor (t(47) 34 p 001) and the HMDt(47) 44 p 0001) Participants from both agegroups were able to name significantly more objectswhen using the GX-monitor platform (youngert(32) 29 p 01 and older t(15) 24 p 01)

Naming of Background ObjectsmdashA main effect wasfound only for the platform (F(1 43) 49 p 05)but no significant interaction effect was found Signifi-cant differences were found between the backgroundobjects named after scanning the office via the GX-monitor (48 24 for participants from both agegroups) as compared to the number named using anHMD (56 23 for participants from both agegroups) (t(46) 21 p 05)

Scan TimemdashA main effect was found only for theplatform (F(1 44) 148 p 0001) while no signifi-cant interaction effect was found A significant differ-ence was found between the time participants took toscan the office presented via the GX-monitor (33

12 s for the whole population) compared to the HMD(41 12 s for participants from both age groups)(t(47) 38 p 0001)

323 Side Effects A significant difference (z

26 p 01) was found in the total score of the SSQfor the GX-monitor (17 306) as compared to thetotal score of the SSQ for the HMD (248 282)Here too the scores were very low for both platformsbut it must be noted that only 6 participants filled in theSSQ after both platforms

324 Exertion A significant main effect wasfound for age group (F(1 41) 93 p 0004) butno significant interaction effect was found For both theGX-monitor and the HMD the younger participantsperceived significantly more exertion while using theGX-monitor (t(38) 27 p 001) and for the HMD(t(43) 36 p 001)

4 Discussion

The objective of this study was to compare the senseof presence level of performance side effects and per-ceived exertion of healthy participants for two different VRplatforms when they played three virtual games or whenthey scanned the Virtual Office In addition the relation-ship or effect of age and gender on these variables was ex-amined The results of this study have highlighted a num-ber of significant differences in some of the keycharacteristics of the VR platforms and environments aswell as user characteristics as they affect usersrsquo sense ofpresence performance side effects and exertion

41 Presence

A significant main effect for the sense of presenceduring the three virtual games was found only for typeof platform That is presence was significantly higherfor the GX-monitor whereas for the Virtual Office amain effect between the participantsrsquo sense of presencewas found for gender and an interaction effect wasfound between age and type of platform These findingshighlight the fact that a participantrsquos sense of presence isnot influenced only by attributes of the VR platformrather by features of the virtual environment (eg typeof games extent of functionality) as well as characteris-tics of the individual user (eg age) and the task itself(eg scanning movement) as suggested by Nash Ed-wards Thompson and Barfield (2000) At least twoother studies have compared the sense of presence re-ported by healthy participants when they used differentVR platforms (Mania amp Chalmers 2001 Lo PrioreCasterlnuovo Liccione amp Liccione 2003) Thesestudies did not find differences in the sense of pres-ence between VR platforms since they did not testfor difference in other variables as was the case in thepresent study further comparison of the results arenot possible

In contrast to the PQ that was completed for eachplatform after participants had experienced all environ-ments the Scenario Feedback questionnaire was com-pleted following each virtual environment (ie after

156 PRESENCE VOLUME 14 NUMBER 2

32 The Virtual Office

321 Presence and Scenario Feedback Ques-tionnaires Presence Questionnaire (total score)mdashAmain effect between participants was found for gender(F(1 45) 76 p 01) and an interaction effect wasfound between age and type of platform (F(1 45)

48 p 05) T-tests were carried out to interpret theinteractions although within the GX-monitor platform

there were no significant differences between the twoage groups when they viewed the Virtual Office via anHMD the older group felt a significantly higher sense ofpresence than did the younger group (t(47) 39 p

0001) Significant differences between males and fe-males were found only for GX-monitor (males 918

151 versus females 825 140) (t(47) 22 p

01) The results of the Presence Questionnaire as well

Table 3 Summary of Results of the Three Virtual Games Played on GX-Monitor and GX-HMD

GX-Monitor GX-HMD

MaleN 24

FemaleN 16

TotalN 40

MaleN 24

FemaleN 16

TotalN 40

PQ total (19ndash133) 991 123 1073 157 1026 142 926 137 973 10 945 124

Involvcont (11ndash77) 595 77 631 95 60 86 576 83 581 65 578 75

Natural (3ndash21) 134 36 158 33 144 36 133 42 162 27 145 4Resolution (2ndash14) 95 23 103 27 10 25 9 25 102 26 94 26Quality (3ndash21) 165 33 18 33 172 33 127 32 126 31 127 31

Scenario FeedbackQuestionnaire (6ndash30)

Birds and Balls 236 38 26 26 246 35 243 35 256 3 248 33Soccer 217 37 224 41 218 38 21 5 216 4 212 45Snowboard 230 40 268 26 245 4 233 51 263 22 246 44

PerformanceBirds and Balls 36 71 39 63 37 06 37 62 43 07 40 07

Soccer 640 10 485 12 578 134 626 120 452 127 401 15

Snowboard 954 65 995 14 97 56 967 6 971 27 97 5SSQ (16ndash2244) 115 124 83 10 101 112 36 282 365 28 362 273

Borg scale perceivedexertion (0ndash20)

Birds and Balls 9 26 91 28 90 26 95 303 98 28 96 3Soccer 13 22 141 22 134 22 131 34 136 28 133 31Snowboard 86 23 81 27 84 25 94 32 9 304 92 31

significantNote differences (t-test or paired t-test)

154 PRESENCE VOLUME 14 NUMBER 2

as other outcome measures for the Virtual Office arelisted in Table 4

When the PQ was divided into its four subscales foreach of the age groups significant differences betweenthe platforms were found for the elderly group for theInterfacequality subscale (t(15) 25 p 05) Dif-ferences in the natural PQ subscale were found betweenthe age groups within the Virtual Office when viewedvia an HMD and GX-monitor whereas the involvementsubscale was found to have a significant difference be-tween the age groups when viewed via the HMD onlyThus the older group had higher scores on the naturalsubscale than did the younger group for both the GX-monitor (t(47) 20 p 05) and for HMD (t(47)

26 p 01) Moreover the older group had a greater

sense of involvementcontrol when using the VirtualOffice with an HMD than did the younger group(t(47) 42 p 0001)

Scenario Feedback QuestionnairemdashA significant maineffect for age was found (F(1 30) 50 p 05) butthere was no significant interaction T-tests were carriedout to analyze the main effect and the results showedthat for the HMD a significant difference was foundbetween the younger and the older participants(t(33) 262 p 05) but no significant differencewas found when using the GX-monitor

322 Performance Visual ScanningmdashA maineffect was found for age (F(1 45) 183 p 0001)and for the type of platform (F(1 45) 117 p

Table 4 Summary of Results of the Virtual OfficemdashGX-Monitor versus HMD

Age group

GX-Monitor HMD

16ndash35 yearsN 33

60ndash75 yearsN 16

16ndash35 yearsN 33

60ndash75 yearsN 16

PQ total (19ndash133) 857 135 898 181 835 127 984 116

Involvementcontrol (11ndash77) 555 78 57 92 524 85 593 84

Natural (3ndash21) 106 44 135 47 109 31 155 42

Resolution (2ndash14) 76 33 906 32 75 31 102 26Interface quality (3ndash21) 12 37 102 35 126 307 133 35

Scenario feedback questionnaire (6ndash30) 238 31 220 36 214 39 251 28

PerformanceVisual scanning 97 28 68 25 82 22 54 15

Naming of background objects 5 22 47 27 55 24 6 203Scan time 36 12 29 113 41 11 41 14

Borg scale of perceived exertion (0ndash20) 912 30 72 14 102 22 75 18

significantNote differences (t-test or paired t-test)

Rand et al 155

001) but no significant interaction effect was foundSignificant differences between age groups were foundfor the number of objects named by the younger partic-ipants when compared to the older participants for boththe GX-monitor (t(47) 34 p 001) and the HMDt(47) 44 p 0001) Participants from both agegroups were able to name significantly more objectswhen using the GX-monitor platform (youngert(32) 29 p 01 and older t(15) 24 p 01)

Naming of Background ObjectsmdashA main effect wasfound only for the platform (F(1 43) 49 p 05)but no significant interaction effect was found Signifi-cant differences were found between the backgroundobjects named after scanning the office via the GX-monitor (48 24 for participants from both agegroups) as compared to the number named using anHMD (56 23 for participants from both agegroups) (t(46) 21 p 05)

Scan TimemdashA main effect was found only for theplatform (F(1 44) 148 p 0001) while no signifi-cant interaction effect was found A significant differ-ence was found between the time participants took toscan the office presented via the GX-monitor (33

12 s for the whole population) compared to the HMD(41 12 s for participants from both age groups)(t(47) 38 p 0001)

323 Side Effects A significant difference (z

26 p 01) was found in the total score of the SSQfor the GX-monitor (17 306) as compared to thetotal score of the SSQ for the HMD (248 282)Here too the scores were very low for both platformsbut it must be noted that only 6 participants filled in theSSQ after both platforms

324 Exertion A significant main effect wasfound for age group (F(1 41) 93 p 0004) butno significant interaction effect was found For both theGX-monitor and the HMD the younger participantsperceived significantly more exertion while using theGX-monitor (t(38) 27 p 001) and for the HMD(t(43) 36 p 001)

4 Discussion

The objective of this study was to compare the senseof presence level of performance side effects and per-ceived exertion of healthy participants for two different VRplatforms when they played three virtual games or whenthey scanned the Virtual Office In addition the relation-ship or effect of age and gender on these variables was ex-amined The results of this study have highlighted a num-ber of significant differences in some of the keycharacteristics of the VR platforms and environments aswell as user characteristics as they affect usersrsquo sense ofpresence performance side effects and exertion

41 Presence

A significant main effect for the sense of presenceduring the three virtual games was found only for typeof platform That is presence was significantly higherfor the GX-monitor whereas for the Virtual Office amain effect between the participantsrsquo sense of presencewas found for gender and an interaction effect wasfound between age and type of platform These findingshighlight the fact that a participantrsquos sense of presence isnot influenced only by attributes of the VR platformrather by features of the virtual environment (eg typeof games extent of functionality) as well as characteris-tics of the individual user (eg age) and the task itself(eg scanning movement) as suggested by Nash Ed-wards Thompson and Barfield (2000) At least twoother studies have compared the sense of presence re-ported by healthy participants when they used differentVR platforms (Mania amp Chalmers 2001 Lo PrioreCasterlnuovo Liccione amp Liccione 2003) Thesestudies did not find differences in the sense of pres-ence between VR platforms since they did not testfor difference in other variables as was the case in thepresent study further comparison of the results arenot possible

In contrast to the PQ that was completed for eachplatform after participants had experienced all environ-ments the Scenario Feedback questionnaire was com-pleted following each virtual environment (ie after

156 PRESENCE VOLUME 14 NUMBER 2

as other outcome measures for the Virtual Office arelisted in Table 4

When the PQ was divided into its four subscales foreach of the age groups significant differences betweenthe platforms were found for the elderly group for theInterfacequality subscale (t(15) 25 p 05) Dif-ferences in the natural PQ subscale were found betweenthe age groups within the Virtual Office when viewedvia an HMD and GX-monitor whereas the involvementsubscale was found to have a significant difference be-tween the age groups when viewed via the HMD onlyThus the older group had higher scores on the naturalsubscale than did the younger group for both the GX-monitor (t(47) 20 p 05) and for HMD (t(47)

26 p 01) Moreover the older group had a greater

sense of involvementcontrol when using the VirtualOffice with an HMD than did the younger group(t(47) 42 p 0001)

Scenario Feedback QuestionnairemdashA significant maineffect for age was found (F(1 30) 50 p 05) butthere was no significant interaction T-tests were carriedout to analyze the main effect and the results showedthat for the HMD a significant difference was foundbetween the younger and the older participants(t(33) 262 p 05) but no significant differencewas found when using the GX-monitor

322 Performance Visual ScanningmdashA maineffect was found for age (F(1 45) 183 p 0001)and for the type of platform (F(1 45) 117 p

Table 4 Summary of Results of the Virtual OfficemdashGX-Monitor versus HMD

Age group

GX-Monitor HMD

16ndash35 yearsN 33

60ndash75 yearsN 16

16ndash35 yearsN 33

60ndash75 yearsN 16

PQ total (19ndash133) 857 135 898 181 835 127 984 116

Involvementcontrol (11ndash77) 555 78 57 92 524 85 593 84

Natural (3ndash21) 106 44 135 47 109 31 155 42

Resolution (2ndash14) 76 33 906 32 75 31 102 26Interface quality (3ndash21) 12 37 102 35 126 307 133 35

Scenario feedback questionnaire (6ndash30) 238 31 220 36 214 39 251 28

PerformanceVisual scanning 97 28 68 25 82 22 54 15

Naming of background objects 5 22 47 27 55 24 6 203Scan time 36 12 29 113 41 11 41 14

Borg scale of perceived exertion (0ndash20) 912 30 72 14 102 22 75 18

significantNote differences (t-test or paired t-test)

Rand et al 155

001) but no significant interaction effect was foundSignificant differences between age groups were foundfor the number of objects named by the younger partic-ipants when compared to the older participants for boththe GX-monitor (t(47) 34 p 001) and the HMDt(47) 44 p 0001) Participants from both agegroups were able to name significantly more objectswhen using the GX-monitor platform (youngert(32) 29 p 01 and older t(15) 24 p 01)

Naming of Background ObjectsmdashA main effect wasfound only for the platform (F(1 43) 49 p 05)but no significant interaction effect was found Signifi-cant differences were found between the backgroundobjects named after scanning the office via the GX-monitor (48 24 for participants from both agegroups) as compared to the number named using anHMD (56 23 for participants from both agegroups) (t(46) 21 p 05)

Scan TimemdashA main effect was found only for theplatform (F(1 44) 148 p 0001) while no signifi-cant interaction effect was found A significant differ-ence was found between the time participants took toscan the office presented via the GX-monitor (33

12 s for the whole population) compared to the HMD(41 12 s for participants from both age groups)(t(47) 38 p 0001)

323 Side Effects A significant difference (z

26 p 01) was found in the total score of the SSQfor the GX-monitor (17 306) as compared to thetotal score of the SSQ for the HMD (248 282)Here too the scores were very low for both platformsbut it must be noted that only 6 participants filled in theSSQ after both platforms

324 Exertion A significant main effect wasfound for age group (F(1 41) 93 p 0004) butno significant interaction effect was found For both theGX-monitor and the HMD the younger participantsperceived significantly more exertion while using theGX-monitor (t(38) 27 p 001) and for the HMD(t(43) 36 p 001)

4 Discussion

The objective of this study was to compare the senseof presence level of performance side effects and per-ceived exertion of healthy participants for two different VRplatforms when they played three virtual games or whenthey scanned the Virtual Office In addition the relation-ship or effect of age and gender on these variables was ex-amined The results of this study have highlighted a num-ber of significant differences in some of the keycharacteristics of the VR platforms and environments aswell as user characteristics as they affect usersrsquo sense ofpresence performance side effects and exertion

41 Presence

A significant main effect for the sense of presenceduring the three virtual games was found only for typeof platform That is presence was significantly higherfor the GX-monitor whereas for the Virtual Office amain effect between the participantsrsquo sense of presencewas found for gender and an interaction effect wasfound between age and type of platform These findingshighlight the fact that a participantrsquos sense of presence isnot influenced only by attributes of the VR platformrather by features of the virtual environment (eg typeof games extent of functionality) as well as characteris-tics of the individual user (eg age) and the task itself(eg scanning movement) as suggested by Nash Ed-wards Thompson and Barfield (2000) At least twoother studies have compared the sense of presence re-ported by healthy participants when they used differentVR platforms (Mania amp Chalmers 2001 Lo PrioreCasterlnuovo Liccione amp Liccione 2003) Thesestudies did not find differences in the sense of pres-ence between VR platforms since they did not testfor difference in other variables as was the case in thepresent study further comparison of the results arenot possible

In contrast to the PQ that was completed for eachplatform after participants had experienced all environ-ments the Scenario Feedback questionnaire was com-pleted following each virtual environment (ie after

156 PRESENCE VOLUME 14 NUMBER 2

001) but no significant interaction effect was foundSignificant differences between age groups were foundfor the number of objects named by the younger partic-ipants when compared to the older participants for boththe GX-monitor (t(47) 34 p 001) and the HMDt(47) 44 p 0001) Participants from both agegroups were able to name significantly more objectswhen using the GX-monitor platform (youngert(32) 29 p 01 and older t(15) 24 p 01)

Naming of Background ObjectsmdashA main effect wasfound only for the platform (F(1 43) 49 p 05)but no significant interaction effect was found Signifi-cant differences were found between the backgroundobjects named after scanning the office via the GX-monitor (48 24 for participants from both agegroups) as compared to the number named using anHMD (56 23 for participants from both agegroups) (t(46) 21 p 05)

Scan TimemdashA main effect was found only for theplatform (F(1 44) 148 p 0001) while no signifi-cant interaction effect was found A significant differ-ence was found between the time participants took toscan the office presented via the GX-monitor (33

12 s for the whole population) compared to the HMD(41 12 s for participants from both age groups)(t(47) 38 p 0001)

323 Side Effects A significant difference (z

26 p 01) was found in the total score of the SSQfor the GX-monitor (17 306) as compared to thetotal score of the SSQ for the HMD (248 282)Here too the scores were very low for both platformsbut it must be noted that only 6 participants filled in theSSQ after both platforms

324 Exertion A significant main effect wasfound for age group (F(1 41) 93 p 0004) butno significant interaction effect was found For both theGX-monitor and the HMD the younger participantsperceived significantly more exertion while using theGX-monitor (t(38) 27 p 001) and for the HMD(t(43) 36 p 001)

4 Discussion

The objective of this study was to compare the senseof presence level of performance side effects and per-ceived exertion of healthy participants for two different VRplatforms when they played three virtual games or whenthey scanned the Virtual Office In addition the relation-ship or effect of age and gender on these variables was ex-amined The results of this study have highlighted a num-ber of significant differences in some of the keycharacteristics of the VR platforms and environments aswell as user characteristics as they affect usersrsquo sense ofpresence performance side effects and exertion

41 Presence

A significant main effect for the sense of presenceduring the three virtual games was found only for typeof platform That is presence was significantly higherfor the GX-monitor whereas for the Virtual Office amain effect between the participantsrsquo sense of presencewas found for gender and an interaction effect wasfound between age and type of platform These findingshighlight the fact that a participantrsquos sense of presence isnot influenced only by attributes of the VR platformrather by features of the virtual environment (eg typeof games extent of functionality) as well as characteris-tics of the individual user (eg age) and the task itself(eg scanning movement) as suggested by Nash Ed-wards Thompson and Barfield (2000) At least twoother studies have compared the sense of presence re-ported by healthy participants when they used differentVR platforms (Mania amp Chalmers 2001 Lo PrioreCasterlnuovo Liccione amp Liccione 2003) Thesestudies did not find differences in the sense of pres-ence between VR platforms since they did not testfor difference in other variables as was the case in thepresent study further comparison of the results arenot possible

In contrast to the PQ that was completed for eachplatform after participants had experienced all environ-ments the Scenario Feedback questionnaire was com-pleted following each virtual environment (ie after

156 PRESENCE VOLUME 14 NUMBER 2

each game and the Virtual Office) The participants whoplayed the virtual games felt similar enjoyment successcontrol and so forth regardless of which VR platformthey used Gender influenced these variables only forthe Snowboard game which indicates that at least forSnowboard the environment or task appeared to have agreater impact than did the type of platform

In contrast in the Virtual Office a significant maineffect for age was found the older group preferred theHMD as compared to the younger group This findingwas unexpected since some of the elderly participantsreported difficulty in focusing on the images while wear-ing the HMD a problem that was particularly notice-able for the participants who wore bifocal glasses Froma clinical point of view the ready acceptance of both theHMD and the video-capture technologies by elderlyparticipants was encouraging since they are a segment ofthe population often in need of rehabilitation StanneyMourant and Kennedy (1998) have suggested that usercharacteristics that significantly influence the VR experi-ence should be considered Characteristics such as priorVR experience cognitive ability personality and agehave been found to influence the virtual experience(Stanney et al) The results of the present study indicatethat gender differences are also important Since appli-cations of VR as an intervention tool for rehabilitationare of great interest (Rizzo and Kim 2005 Weiss Ki-zony Feintuch amp Katz in press) future studies shouldalso examine the effect that disability and subsequentimpairment (eg motor cognitive) have on participantperformance in different VR platforms

42 Performance

Performance during each of the three virtual gamesdiffered not only between the platforms (for Birds amp Balls)but was also influenced by the gender of the participants(for Soccer) This finding again points to the importanceof both the virtual environment and task characteristics forparticipant performance It is interesting to consider whythe male participants performed so much better duringSoccer than did their female counterparts One possibleexplanation is that this scenario was more meaningful for

the male participants and they were therefore more moti-vated to achieve higher scores The literature on the im-pact of meaningfulness of a virtual environment on VRperformance is sparse (eg Hoffman Prothero Wells ampGroen 1998) and should be further investigated

Performance for the Virtual Office environment did dif-fer between the platforms Moreover for this environ-ment there was an interaction effect on performance be-tween the type of platform and age After scanning theoffice via the GX-monitor participants from both agegroups were able to name more objects from the list of 16embedded items but fewer background objects than whilescanning the office via the HMD The flat-screen monitorappeared to encourage quicker and easier visual scanningperhaps because the objects were located horizontallywithin the officersquos panoramic 2D on-screen image In con-trast visual scanning via the HMD was slower perhaps dueto the greater complexity of the 3D image

The decreased performance of the older group in bothtypes of platforms points to the validity of the tasks per-formed within these VR systems since we expect to seedifferences as a function of age A decline in visual acuity(eg Stanney et al 1998) as well as perceptual-cognitiveabilities such as memory (eg Josman amp Hartman-Maeir2000) could lead to the decreased performance demon-strated in this study The performance of the older groupmay also be influenced by the task characteristics as indi-cated above

43 Side Effects

Since the potential for VE-related side effects (cyber-sickness) is considered to be a key potential ethical consid-eration for the use of VR for rehabilitation (Rizzo et al2002b) different studies have addressed this issue inhealthy and in patient populations (eg Pugnetti et al1998) The extent of the side effects experienced in thisstudy on both platforms was quite low and no participantsrequested to terminate their participation in the studyThis fact is encouraging especially when one is interestedin using the various platforms in rehabilitation Althoughthe exposure times in this study were consistent with typi-cal treatment protocols (successive 2- to 4-minute trials

Rand et al 157

separated by 2- to 3-minute breaks) they were still rela-tively short Thus our negative findings with respect to sideeffects should be considered carefully as it may be that forlonger exposure times the likelihood of side effects wouldincrease For example participants in studies conducted byHowarth amp Finch (1999) were assessed for side effectsevery 2ndash5 minutes during a 20-minute exposure An in-crease in nausea as time progressed was found approxi-mately 40 of the participants reported at least one cyber-sickness symptom after 20 minutesrsquo exposure

Although overall the incidence of side effects in thepresent study was low there were significant differencesbetween platforms Participants reported side effects morefrequently while using the HMD as compared to the GX-monitor This finding is in accordance with the literaturewhich suggests that HMDs have a greater potential forcausing short-term side effects especially oculomotorsymptoms (Lo Priore et al 2003) Our findings also dem-onstrated that side effects experienced while wearing anHMD depended upon the type of environmentmdashfewerparticipants complained of side effects while experiencingthe Virtual Office as compared to the virtual games Thisresult may be explained in part by differences in the ex-tent of active movement within the two environmentsThe virtual games encouraged participants to move theirlimbs and even their entire bodies while attending to thetraveling stimuli In contrast stimuli within the VirtualOffice environment were static and participants scannedthe environment with small head movements rather thanmoving within it The differences in occurrence in sideeffects in this study may also be due to the fact that totalexposure times to the virtual games (3ndash4 min each for 6games with 2ndash3 min breaks between games) was longerthan for the Virtual Office (up to 5 min) It should benoted that for the virtual games the HMD was connectedto the GX platform to create a combined platform (GX-HMD) The HMD was not built originally for this usewhich may have caused more side effects (see Table 2 forfurther description of this platform)

44 Exertion

The participantsrsquo perceived exertion while playingthe virtual games was not influenced by either platform or

gender Thus any encumbrance associated with the HMDdue to its weight or cables did not add to the effort expe-rienced by the participants as compared to their effort viathe GX-monitor In contrast the type of gametask playeddid affect the level of perceived exertion a finding that islikely due to difference in the physical effort required bythe different games The level of exertion ranged fromldquovery lightrdquo for the Snowboard game to ldquosomewhat hardrdquofor the Soccer game as measured with Borgrsquos scale (Borg1990) Thus participants responded to the virtual stimuliwith large relatively rapid movements of the whole bodyand extremities during the Soccer scenarios but onlyshifted their weight gently from side to side during Snow-board The level of perceived exertion for Birds amp Ballsand Soccer was higher than that perceived for scanning theVirtual Office which ranged from ldquoextremely lightrdquo forthe younger group to ldquovery lightrdquo for the older group Itmay be that the scores were lower since scanning the Vir-tual Office environment did not require much movementof the body A significant main effect was found betweenage and level of perceived exertion Surprisingly Borg-scale scores from the younger participants while scanningthe Virtual Office were higher than the scores for the oldergroup a finding that may be related to the fact that theolder group enjoyed this task more than did their youngercounterparts and thus did not pay as much attention totheir effort

45 Conclusions

These results highlight the need to investigate theuse of alternate platforms environments and tasks whencreating protocols for rehabilitation to target therapeuticobjectives (eg motivation performance and effort) It isanticipated that continued cross-platform and cross-envi-ronment comparison studies will provide further insightinto their different characteristics especially when ex-panded to include participants across a range of impair-ments The results of this study underscore the importanceof collecting data regarding the performance of differenthealthy populations (gender age groups etc) in order todetermine which tasks and platforms have the best usabilityand relevance for patients during the rehabilitation processFuture work should be directed at examining the re-

158 PRESENCE VOLUME 14 NUMBER 2

sponses of patients with different cognitive or motor defi-cits as well as a greater number of elderly participants inorder to clarify their needs and behaviors when using bothplatforms

Acknowledgments

We thank Meir Shahar and Yuval Naveh for programming ad-aptations to the virtual environments The third-year studentsin the Department of Occupational Therapy at the Universityof Haifa are thanked for helping with recruitment of partici-pants and data collection The financial support from the Uni-versity of Haifa the Israeli Ministry of Health and theKoniver Foundation is gratefully acknowledged

References

AOTA (2002) The occupational therapy practice frameworkDomain and process The American Journal of OccupationalTherapy 56 609ndash639

Borg G (1990) Psychophysical scaling with applications inphysical work and the perception of exertion The Scandina-vian Journal of Work Environment and Health 16 (Suppl1) 55ndash58

Howarth P A amp Finch M (1999) The nauseogenicity oftwo methods of navigating within a virtual environmentApplied Ergonomics 30 39ndash45

Hoffman H G Prothero J Wells M amp Groen J (1998)Virtual chess The role of meaning in the sensation of pres-ence International Journal of Human-Computer Interac-tion 10 251ndash263

Jack D Boian R Merians A Tremaine M Burdea G Ad-amovich A et al (2001) Virtual reality-enhanced stroke re-habilitation IEEE Transactions on Neural Systems and Rehabil-itation Engineering 9 308ndash318

Josman N amp Hartman-Maeir A (2000) Cross cultural as-sessment of the contextual memory test (CMT) Occupa-tional Therapy International 7 246ndash258

Kennedy R S Lane N E Berbaum K S amp LilienthalM G (1993) Simulator sickness questionnaire An en-hanced method for quantifying simulator sickness Interna-tional Journal of Aviation Psychology 3 203ndash220

Kizony R Katz N Weingarden H amp Weiss P L (2002)Immersion without encumbrance Adapting a virtual reality

system for the rehabilitation of individuals with stroke andspinal cord injury In P Sharkey C S Lanyi P Stanton(Eds) Proceedings of the Fourth International Conference onDisability Virtual Reality and Associated Technology (pp55ndash61) University of Reading Vresprem Hungary

Kizony R Katz N amp Weiss P L (2003) Adapting an im-mersive virtual reality system for rehabilitation Journal ofVisualization and Computer Animation 14 261ndash268

Kizony R Raz L Katz N Weingarden H Weiss P L(2003) Using a video projected VR system for patientswith spinal cord injury In G C Burdea D Thalmann ampJ A Lewis (Eds) Proceedings of the Second InternationalWorkshop on Virtual Rehabilitation (pp 82ndash88) Piscat-away NJ Rutgers University

Lo Priore C Casterlnuovo G Liccione D amp Liccione D(2003) Experience with V-STORE Considerations on pres-ence in virtual environments for effective neuropsychologicalrehabilitation of executive functions CyberPsychology and Be-havior 6 281ndash287

Mania K amp Chalmers A (2001) The effects of levels of immer-sion on memory and presence in virtual environments A realitycentered approach Cyber Psychology and Behavior 4 247ndash264

Nash E B Edwards G W Thompson J A amp BarfieldW (2000) A review of presence and performance in virtualenvironments International Journal of Human-ComputerInteraction 12 1ndash41

Pugnetti L Mendozzi L Barbieri E Motta A AlpiniD Attree E A et al (1998) Developments of a collabo-rative research on VR applications for mental health In PSharkey D Rose amp J Lindstrom (Eds) Proceedings of theSecond European Conference on Disability Virtual Realityand Associated Techniques (pp 77ndash84) Skove SwedenUniversity of Reading

Rizzo A A Bowerly T Buckwalter J G SchultheisM T Matheis R Shahabi C et al (2002a) Virtual envi-ronments for the assessment of attention and memory pro-cesses The virtual classroom and office Proceedings of theFourth International Conference on Disability Virtual Real-ity and Associated Technology 3ndash12

Rizzo A amp Kim G J (2005) A SWOT analysis of the fieldof virtual-reality rehabilitation and therapy Presence Tele-operators and Virtual Environments 14(2) 119ndash146

Rizzo A A Schultheis M T amp Rothbaum B O (2002b)Ethical issues for the use of virtual reality in the psychologi-cal sciences In S Bush amp M Drexler (Eds) Ethical Issuesin Clinical Neuropsychology Lisse NL Swets amp ZeitlingerPublishers

Rand et al 159

Schultheis M T amp Rizzo A A (2001) The application ofvirtual reality technology in rehabilitation RehabilitationPsychology 46 296ndash311

Stanney K M Mourant R R amp Kennedy R S (1998)Human factors issues in virtual environments A review ofthe literature Presence Teleoperators and Virtual Environ-ments 7(4) 327ndash351

Sveistrup H McComas J Thornton M Marshall S Fine-stone H McCormick A et al (2003) Experimentalstudies of virtual reality-delivered compared to conventionalexercise programs for rehabilitation Cyber Psychology amp Be-havior 6 245ndash249

Weiss P L Naveh Y amp Katz N (2003) Design and test-ing of a virtual environment to train CVA patients with uni-lateral spatial neglect to cross a street safely OccupationalTherapy International 10 39ndash55

Weiss P L Kizony K Feintuch U amp Katz N (inpress) Virtual reality in neurorehabilitation In M ESelzer L Cohen F H Gage S Clarke amp P W Dun-can (Eds) Textbook of Neural Repair and Neurorehabili-tation

Witmer B G amp Singer M J (1998) Measuring presence invirtual environments A presence questionnaire PresenceTeleoperators and Virtual Environments 7 225ndash240

160 PRESENCE VOLUME 14 NUMBER 2