Children With Developmental Coordination Disorder Play Active Virtual Reality Games Differently Than Children With Typical Development Leandra Gonsalves, Amity Campbell, Lynn Jensen and Leon Straker PHYS THER. Published online October 9, 2014 doi: 10.2522/ptj.20140116
The online version of this article, along with updated information and services, can be found online at: http://ptjournal.apta.org/content/early/2014/10/08/ptj.20140116 E-mail alerts
Sign up here to receive free e-mail alerts
Online First articles are published online before they appear in a regular issue of Physical Therapy (PTJ). PTJ publishes 2 types of Online First articles: Author manuscripts: PDF versions of manuscripts that have been peer-reviewed and accepted for publication but have not yet been copyedited or typeset. This allows PTJ readers almost immediate access to accepted papers. Page proofs: edited and typeset versions of articles that incorporate any author corrections and replace the original author manuscript.
Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Running head: Children With Developmental Coordination Disorder
Innovative Technologies Special Issue
Children With Developmental Coordination Disorder Play Active Virtual Reality Games Differently Than Children With Typical Development
Leandra Gonsalves, Amity Campbell, Lynn Jensen, Leon Straker
L. Gonsalves, BSc(hons), School of Physiotherapy and Exercise Science, Curtin University, Perth, Western Australia, Australia.
A. Campbell, BSc(hons), PhD, School of Physiotherapy and Exercise Science, Curtin University.
L. Jensen, BAppSc, PostgradDip, MSc, School of Physiotherapy and Exercise Science, Curtin University.
L. Straker, BAppSc, MSc, PhD, School of Physiotherapy and Exercise Science, Curtin University, GPO Box U1987, Perth, Western Australia, Australia 6845. Address all correspondence to Dr Straker at: [email protected].
1 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
[Gonsalves L, Campbell A, Jensen L, Straker L. Children with developmental coordination disorder play active virtual reality games differently than children with typical development. Phys Ther. 2014;94:xxx–xxx.] © 2014 American Physical Therapy Association
Published Ahead of Print: xxx Accepted: October 2, 2014 Submitted: March 17, 2014
2 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Abstract Background. Active Virtual Reality Gaming (AVG) may be useful for children with Developmental Coordination Disorder (DCD) to practice motor skills if their movement patterns are good quality whilst playing AVG. Objective. This study aimed to examine: 1) the quality of motor patterns of children with DCD playing AVG by comparing them to typically developing children (TD), and 2) whether differences existed in the motor patterns utilized on two AVG types: Sony PlayStation 3 Move and Microsoft Xbox 360 Kinect. Design. Quasi-experimental biomechanical laboratory based study. Methods. Twenty-one children with DCD, aged 10 to 12 years, and 19 age and sex matched TD children played a match of table tennis on each AVG type. Hand path, wrist angle and elbow angle were recorded using the Vicon motion analysis system. Linear mixed-model analyses were utilised to determine differences between DCD and TD groups and Move and Kinect AVG type for forehands and backhands. Results. Children with DCD utilized a slower hand path speed (backhand Mean Difference (MD): 1.20ms-1, 95% Confidence Interval (CI): 0.41 to 1.98); greater wrist extension (forehand MD: 34.3°, 95% CI: 22.6 to 47.0); and greater elbow flexion (forehand MD: 22.3°, 95% CI: 7.4 to 37.1) than TD children when playing AVG. There were also differences in movement patterns utilized between AVG types. Limitations. Only simple kinematic measures were compared and no data regarding movement outcome were assessed. Conclusions. If a therapeutic treatment goal is to promote movement quality in children with DCD clinical judgement is required to select the most appropriate AVG type and determine whether movement quality is adequate for unsupervised practice.
3 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Introduction Developmental Coordination Disorder (DCD) is a condition in which children have a marked impairment in motor development impacting on their daily life1. It has been reported to affect between 5-10% of children worldwide1,2. Presentations can vary substantially and may involve both gross and fine motor control,3 which may impact on overall movement quality. Reported impairments include lack of body position awareness, delayed reaction times and reduced movement speed.2,4-6 Joint movements lacking precision and fluency, coupled with increased variability in task performance have also been reported.2,4 The increasingly accepted ‘internal model deficit (IMD) theory’ suggests these deficits result from sub optimal motor planning and sensoriperceptual integration.1,2,4,7,8 More specifically, this theory hypothesises that children with DCD have difficulty correcting movements in real time due to an inability to use or generate predictive estimates of body position.2 Failure to manage such impairments may affect a child’s physical and psychological health.9 The lack of age appropriate motor skills, and low confidence in motor skills, may lead children with DCD to participate less in physical activities.8 Lack of participation in sport and play can affect a developing child’s ability to learn and practice their motor skills.4 This may eventuate into a cycle in which psychosocial and physical factors feedback on each other to further impact on the child’s ability to learn and develop normally.1 Thus there is an argument for health care professionals to manage these impairments of movement quality in early life to optimise short and long term health.10 A poor understanding of DCD aetiology2 has led to diverse intervention styles, with over 30 different approaches being reported.10 No one style of intervention has gained universal acceptance.11 There is debate whether participation in physical activity alone is sufficient as an intervention for children with DCD,11 or whether promoting good
4 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
quality movement should be the focus.2 It has been suggested that interventions addressing movement quality may precipitate neuroplastic changes and adjust deficient internal modelling processes, which is critical to motor rehabilitation.2,12 To be most effective in doing this it is thought that interventions should consider the heterogeneity in presentations of children with DCD, and additionally be engaging and involve many opportunities for motor practice.1,10,13 Thus, it is important to consider interventions that are appealing to children to enhance compliance with motor practice opportunities from an early age. Electronic games may be a useful means for achieving this. Electronic games have been defined as an interactive activity involving manipulation of figures on a screen.14 A 2007 study indicated that 94% of American school-aged children had played some form of electronic game in the 6 months prior.15 There is strong evidence that electronic games increase motivation and self-confidence in children.14,16,17 They also provide immediate feedback about performance which has been shown to be important for motor skill development, and may be particularly important in children with DCD who are reported as having deficits in internal feedback modelling.4 Additionally, electronic game play is often unsupervised, allowing for a considerable volume of use within a home setting.16,18 Traditionally electronic gaming has involved the use of a gamepad, joystick, computer keyboard or mouse game controller.19,20 These only involved fine hand movements with little gross movement.21 There has been an emerging trend towards active virtual reality gaming (AVG).20 Two gaming platforms that have recently adopted AVG play are the PlayStation 3 Move (Sony, Tokyo, Japan) and Xbox 360 Kinect (Microsoft, Redmond, USA). Both use a video camera placed near the television screen to capture the movements of the player to create a simulated character that mimics the movements of the player.
5 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
In recognition of the potential therapeutic benefits of AVG, a growing area of research is examining the training of motor skills through AVG.18,22-26 Recent studies have implied relationships between AVG intervention and improved endurance, coordination and balance in children with neurodevelopmental conditions such as Cerebral Palsy and Down Syndrome.25,27-29 Further, AVG have been shown to effectively improve balance in children with DCD 24. This suggests that AVG might also help children with DCD improve other aspects of gross motor skill development; an area future research aims to confirm. 22 Active Virtual Reality Gaming may be useful therapeutically to simply gain participation in some form of physical activity, for psychological benefits.30 Alternatively, the therapeutic aim may be to promote good quality movement, thereby facilitating neuroplastic changes and the normal development of other body systems.1,12,18 Whilst AVG seems promising in a motor rehabilitation setting, little is known regarding the movement patterns utilized during AVG play. To be most effective as an intervention for enhancing movement quality, a greater understanding of the movement patterns of children with DCD during AVG play is required. This will allow therapists to form targeted goals and prescription guidelines to address the movement quality issues, thereby optimising the benefit gained from an AVG intervention. The primary aim of this study was to determine whether differences existed in the motor patterns of children with DCD and children with typical development during AVG table tennis game play. It was hypothesised that children with DCD would utilize different movement patterns compared to typically developing (TD) children. In order to establish whether clinicians can plan standardised treatment across systems, the secondary aim of this study was to compare the movement patterns required between two popular AVG systems (Move and Kinect) during table tennis. It was hypothesised
6 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
that differences in the required movement patterns would exist between the two AVG types. Method: Design: We conducted a quasi-experimental laboratory study comparing hand path measures and upper limb joint angles of two groups of children (DCD and TD) playing AVG table tennis on two consoles (Move and Kinect). Participants: This study collected data from 40 male and female participants between 10 and 12 years old (refer to Table 1 for participant information). Twenty-one children with DCD were recruited into a randomized control trial (RCT)22 examining the impact of traditional electronic games and AVG on motor skill. Baseline data from the RCT (collected in early 2011) were utilised in the current study. Nineteen age and sex matched typically developing children (TD group) were recruited and tested in late 2012. Eleven of the 21 participants comprising the DCD group were classified as at risk of being obese/obese based on sex and age corrected body mass index percentile, compared with only 3 of the 17 participants in the TD group (Table 1).31 Recruitment to the study was through school and community notices and networks. Children were included in the ‘DCD group’ if they scored < 16th percentile on the Movement Assessment Battery for Children Second Edition (MABC-2)32 and their motor impairments impacted on their activities of daily living based on the parent-reported Developmental Coordination Disorder Questionnaire (DCDQ) (a score of ≤15th percentile). These scores are reflective of children considered ‘at risk’ of DCD.33,34 The children had no other obvious disorder likely to impact on their coordination and participation in the study in compliance with the diagnostic criteria listed in the Diagnostic and Statistical Manual of
7 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Mental Disorders 4th edition (DSM-IV).11,32,35 Children were included in the TD group if they scored >16th percentile in the MABC-2 and had no reports from parents or teachers of movement problems impacting on activities of daily living or academic performance. Active Virtual Reality Games: Two AVG types were used: 1) Sony’s PlayStation 3 Move (Move) which uses a single motion sensing camera to track the movements of a handheld wand together with sensors contained within the wand for translation, acceleration and rotation,18 2) Microsoft’s Xbox Kinect (Kinect) tracks whole body movement in three dimensions using an infrared laser with dual camera sensors, eliminating the need for a handheld controller.36 The game used was table tennis, from PlayStation Move Sports Champions (Sony Computer Entertainment, Tokyo, Japan) and Xbox Kinect Sports (Microsoft Studios, Redmond, USA). Procedures: Volunteers expressing interest in the study were provided with participant information sheets outlining the study details. Informed written consent/assent from parent/child participant was obtained. Children attended the XXXXXXXXXX Motion Analysis Laboratory for two hours of testing. Each child performed the MABC-2 test. Parents or guardians of those children scoring < 16th percentile were asked to complete the DCDQ. Participants were fitted with a set of retro-reflective body markers on their preferred hand, forearm and upper limb. Body markers were fixed to specific anatomical locations in compliance with the protocol of the International Society of Biomechanics.7 A three dimensional motion analysis system (Vicon; Oxford Metrics, Los Angeles, USA) was utilised to track the position of the markers throughout data collection at a sampling rate of 250Hz. This
8 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
system has been reported to be one of the most accurate and reliable systems, with reconstruction errors < 0.5mm.37 Following a standardised set of instructions, participants played table tennis on Move for five practice points followed by one full game to 11 points. This process was then repeated using Kinect. During practice the investigator was able to provide feedback about technique (for example full forehand/backhand stokes were encouraged); however no feedback was given during the formal assessment. All tasks were performed in the same order against a computer generated opponent using the beginner setting and all children were allowed rest between tasks to minimise the effects of fatigue on subsequent tasks. A consistent order was used to minimise participant burden given that a number of tasks were performed on each AVG type as part of the RCT study. Data Processing: Vicon motion analysis software (Nexus; Oxford Metrics, Los Angeles, USA) was utilised to check marker trajectories for breaks that can result from occlusion of the markers. Gaps were filled using algorithmic interpolation between trajectory end points, with no break greater than 20 frames in duration. The data were then filtered using a quintic spline filter using a mean square error of 3, as determined by a residual analysis.38 A valid upper limb three dimensional mathematical model,39 that utilised previously published upper limb segment parameters40 and followed recommended biomechanical procedures was applied in order to calculate upper limb kinematics.7 Three forehands and three backhands were randomly identified from both Kinect and Move game play data. A stroke was defined as the end of the backswing until the end of the forward swing for both backhand and forehand. A stroke was considered a backhand when the dominant hand was on the non-dominant side of the body, with the palm
9 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
facing away from the screen at the start of the forward swing; while a forehand was defined by the dominant hand being on the dominant side of the body, with palm facing towards the screen at the start of the forward swing. Two participants in the TD group (one female and one male) had motion analysis data that were missing or corrupted and were therefore not included in analyses. A custom LabVIEW program (National Instruments, Austin, USA) was used to output hand path distance and speed, and wrist and elbow range of motion for each stroke. Statistical analyses were performed with SPSS (version 21; IBM, Armonk, USA). Following assumption testing, linear mixed-model analyses were utilised to determine differences between DCD and TD groups (between groups comparison) and Move and Kinect AVG type (repeated measures) and any interaction between group and AVG type. Alpha probability was set at 0.05. Results: Hand Path Measures: Children with DCD utilized a significantly slower maximum hand speed than TD children during the backhands irrespective of game type (Mean Difference (MD): 1.20ms-1, 95% Confidence Interval (CI): 0.41 to 1.98, p=0.04) (Table 2), although there was no significant differences in hand path distance. Post hoc power calculations for hand path distance found observed power for detecting a difference in hand path distance between two independent groups (DCD n=20 and TD n=20) using the observed between group difference (Move 0.49; Kinect 0.36; Table 2) and averaged within group standard deviation (Move 0.51; Kinect 0.53; Table 2) was Move 0.841; Kinect 0.549 (Using PS Power and Sample Size Calculations v3.1.2, Vanderbilt University, 2009). Power estimates were for a main effect, not a group (DCD, TD) by game type interaction effect.
10 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Differences between AVG type were also detected. Specifically, children’s average hand path speed using Move was slower than when using Kinect (forehand MD: 0.82ms-1, 95% CI: 0.52 to 1.12, p
The online version of this article, along with updated information and services, can be found online at: http://ptjournal.apta.org/content/early/2014/10/08/ptj.20140116 E-mail alerts
Sign up here to receive free e-mail alerts
Online First articles are published online before they appear in a regular issue of Physical Therapy (PTJ). PTJ publishes 2 types of Online First articles: Author manuscripts: PDF versions of manuscripts that have been peer-reviewed and accepted for publication but have not yet been copyedited or typeset. This allows PTJ readers almost immediate access to accepted papers. Page proofs: edited and typeset versions of articles that incorporate any author corrections and replace the original author manuscript.
Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Running head: Children With Developmental Coordination Disorder
Innovative Technologies Special Issue
Children With Developmental Coordination Disorder Play Active Virtual Reality Games Differently Than Children With Typical Development
Leandra Gonsalves, Amity Campbell, Lynn Jensen, Leon Straker
L. Gonsalves, BSc(hons), School of Physiotherapy and Exercise Science, Curtin University, Perth, Western Australia, Australia.
A. Campbell, BSc(hons), PhD, School of Physiotherapy and Exercise Science, Curtin University.
L. Jensen, BAppSc, PostgradDip, MSc, School of Physiotherapy and Exercise Science, Curtin University.
L. Straker, BAppSc, MSc, PhD, School of Physiotherapy and Exercise Science, Curtin University, GPO Box U1987, Perth, Western Australia, Australia 6845. Address all correspondence to Dr Straker at: [email protected].
1 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
[Gonsalves L, Campbell A, Jensen L, Straker L. Children with developmental coordination disorder play active virtual reality games differently than children with typical development. Phys Ther. 2014;94:xxx–xxx.] © 2014 American Physical Therapy Association
Published Ahead of Print: xxx Accepted: October 2, 2014 Submitted: March 17, 2014
2 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Abstract Background. Active Virtual Reality Gaming (AVG) may be useful for children with Developmental Coordination Disorder (DCD) to practice motor skills if their movement patterns are good quality whilst playing AVG. Objective. This study aimed to examine: 1) the quality of motor patterns of children with DCD playing AVG by comparing them to typically developing children (TD), and 2) whether differences existed in the motor patterns utilized on two AVG types: Sony PlayStation 3 Move and Microsoft Xbox 360 Kinect. Design. Quasi-experimental biomechanical laboratory based study. Methods. Twenty-one children with DCD, aged 10 to 12 years, and 19 age and sex matched TD children played a match of table tennis on each AVG type. Hand path, wrist angle and elbow angle were recorded using the Vicon motion analysis system. Linear mixed-model analyses were utilised to determine differences between DCD and TD groups and Move and Kinect AVG type for forehands and backhands. Results. Children with DCD utilized a slower hand path speed (backhand Mean Difference (MD): 1.20ms-1, 95% Confidence Interval (CI): 0.41 to 1.98); greater wrist extension (forehand MD: 34.3°, 95% CI: 22.6 to 47.0); and greater elbow flexion (forehand MD: 22.3°, 95% CI: 7.4 to 37.1) than TD children when playing AVG. There were also differences in movement patterns utilized between AVG types. Limitations. Only simple kinematic measures were compared and no data regarding movement outcome were assessed. Conclusions. If a therapeutic treatment goal is to promote movement quality in children with DCD clinical judgement is required to select the most appropriate AVG type and determine whether movement quality is adequate for unsupervised practice.
3 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Introduction Developmental Coordination Disorder (DCD) is a condition in which children have a marked impairment in motor development impacting on their daily life1. It has been reported to affect between 5-10% of children worldwide1,2. Presentations can vary substantially and may involve both gross and fine motor control,3 which may impact on overall movement quality. Reported impairments include lack of body position awareness, delayed reaction times and reduced movement speed.2,4-6 Joint movements lacking precision and fluency, coupled with increased variability in task performance have also been reported.2,4 The increasingly accepted ‘internal model deficit (IMD) theory’ suggests these deficits result from sub optimal motor planning and sensoriperceptual integration.1,2,4,7,8 More specifically, this theory hypothesises that children with DCD have difficulty correcting movements in real time due to an inability to use or generate predictive estimates of body position.2 Failure to manage such impairments may affect a child’s physical and psychological health.9 The lack of age appropriate motor skills, and low confidence in motor skills, may lead children with DCD to participate less in physical activities.8 Lack of participation in sport and play can affect a developing child’s ability to learn and practice their motor skills.4 This may eventuate into a cycle in which psychosocial and physical factors feedback on each other to further impact on the child’s ability to learn and develop normally.1 Thus there is an argument for health care professionals to manage these impairments of movement quality in early life to optimise short and long term health.10 A poor understanding of DCD aetiology2 has led to diverse intervention styles, with over 30 different approaches being reported.10 No one style of intervention has gained universal acceptance.11 There is debate whether participation in physical activity alone is sufficient as an intervention for children with DCD,11 or whether promoting good
4 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
quality movement should be the focus.2 It has been suggested that interventions addressing movement quality may precipitate neuroplastic changes and adjust deficient internal modelling processes, which is critical to motor rehabilitation.2,12 To be most effective in doing this it is thought that interventions should consider the heterogeneity in presentations of children with DCD, and additionally be engaging and involve many opportunities for motor practice.1,10,13 Thus, it is important to consider interventions that are appealing to children to enhance compliance with motor practice opportunities from an early age. Electronic games may be a useful means for achieving this. Electronic games have been defined as an interactive activity involving manipulation of figures on a screen.14 A 2007 study indicated that 94% of American school-aged children had played some form of electronic game in the 6 months prior.15 There is strong evidence that electronic games increase motivation and self-confidence in children.14,16,17 They also provide immediate feedback about performance which has been shown to be important for motor skill development, and may be particularly important in children with DCD who are reported as having deficits in internal feedback modelling.4 Additionally, electronic game play is often unsupervised, allowing for a considerable volume of use within a home setting.16,18 Traditionally electronic gaming has involved the use of a gamepad, joystick, computer keyboard or mouse game controller.19,20 These only involved fine hand movements with little gross movement.21 There has been an emerging trend towards active virtual reality gaming (AVG).20 Two gaming platforms that have recently adopted AVG play are the PlayStation 3 Move (Sony, Tokyo, Japan) and Xbox 360 Kinect (Microsoft, Redmond, USA). Both use a video camera placed near the television screen to capture the movements of the player to create a simulated character that mimics the movements of the player.
5 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
In recognition of the potential therapeutic benefits of AVG, a growing area of research is examining the training of motor skills through AVG.18,22-26 Recent studies have implied relationships between AVG intervention and improved endurance, coordination and balance in children with neurodevelopmental conditions such as Cerebral Palsy and Down Syndrome.25,27-29 Further, AVG have been shown to effectively improve balance in children with DCD 24. This suggests that AVG might also help children with DCD improve other aspects of gross motor skill development; an area future research aims to confirm. 22 Active Virtual Reality Gaming may be useful therapeutically to simply gain participation in some form of physical activity, for psychological benefits.30 Alternatively, the therapeutic aim may be to promote good quality movement, thereby facilitating neuroplastic changes and the normal development of other body systems.1,12,18 Whilst AVG seems promising in a motor rehabilitation setting, little is known regarding the movement patterns utilized during AVG play. To be most effective as an intervention for enhancing movement quality, a greater understanding of the movement patterns of children with DCD during AVG play is required. This will allow therapists to form targeted goals and prescription guidelines to address the movement quality issues, thereby optimising the benefit gained from an AVG intervention. The primary aim of this study was to determine whether differences existed in the motor patterns of children with DCD and children with typical development during AVG table tennis game play. It was hypothesised that children with DCD would utilize different movement patterns compared to typically developing (TD) children. In order to establish whether clinicians can plan standardised treatment across systems, the secondary aim of this study was to compare the movement patterns required between two popular AVG systems (Move and Kinect) during table tennis. It was hypothesised
6 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
that differences in the required movement patterns would exist between the two AVG types. Method: Design: We conducted a quasi-experimental laboratory study comparing hand path measures and upper limb joint angles of two groups of children (DCD and TD) playing AVG table tennis on two consoles (Move and Kinect). Participants: This study collected data from 40 male and female participants between 10 and 12 years old (refer to Table 1 for participant information). Twenty-one children with DCD were recruited into a randomized control trial (RCT)22 examining the impact of traditional electronic games and AVG on motor skill. Baseline data from the RCT (collected in early 2011) were utilised in the current study. Nineteen age and sex matched typically developing children (TD group) were recruited and tested in late 2012. Eleven of the 21 participants comprising the DCD group were classified as at risk of being obese/obese based on sex and age corrected body mass index percentile, compared with only 3 of the 17 participants in the TD group (Table 1).31 Recruitment to the study was through school and community notices and networks. Children were included in the ‘DCD group’ if they scored < 16th percentile on the Movement Assessment Battery for Children Second Edition (MABC-2)32 and their motor impairments impacted on their activities of daily living based on the parent-reported Developmental Coordination Disorder Questionnaire (DCDQ) (a score of ≤15th percentile). These scores are reflective of children considered ‘at risk’ of DCD.33,34 The children had no other obvious disorder likely to impact on their coordination and participation in the study in compliance with the diagnostic criteria listed in the Diagnostic and Statistical Manual of
7 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Mental Disorders 4th edition (DSM-IV).11,32,35 Children were included in the TD group if they scored >16th percentile in the MABC-2 and had no reports from parents or teachers of movement problems impacting on activities of daily living or academic performance. Active Virtual Reality Games: Two AVG types were used: 1) Sony’s PlayStation 3 Move (Move) which uses a single motion sensing camera to track the movements of a handheld wand together with sensors contained within the wand for translation, acceleration and rotation,18 2) Microsoft’s Xbox Kinect (Kinect) tracks whole body movement in three dimensions using an infrared laser with dual camera sensors, eliminating the need for a handheld controller.36 The game used was table tennis, from PlayStation Move Sports Champions (Sony Computer Entertainment, Tokyo, Japan) and Xbox Kinect Sports (Microsoft Studios, Redmond, USA). Procedures: Volunteers expressing interest in the study were provided with participant information sheets outlining the study details. Informed written consent/assent from parent/child participant was obtained. Children attended the XXXXXXXXXX Motion Analysis Laboratory for two hours of testing. Each child performed the MABC-2 test. Parents or guardians of those children scoring < 16th percentile were asked to complete the DCDQ. Participants were fitted with a set of retro-reflective body markers on their preferred hand, forearm and upper limb. Body markers were fixed to specific anatomical locations in compliance with the protocol of the International Society of Biomechanics.7 A three dimensional motion analysis system (Vicon; Oxford Metrics, Los Angeles, USA) was utilised to track the position of the markers throughout data collection at a sampling rate of 250Hz. This
8 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
system has been reported to be one of the most accurate and reliable systems, with reconstruction errors < 0.5mm.37 Following a standardised set of instructions, participants played table tennis on Move for five practice points followed by one full game to 11 points. This process was then repeated using Kinect. During practice the investigator was able to provide feedback about technique (for example full forehand/backhand stokes were encouraged); however no feedback was given during the formal assessment. All tasks were performed in the same order against a computer generated opponent using the beginner setting and all children were allowed rest between tasks to minimise the effects of fatigue on subsequent tasks. A consistent order was used to minimise participant burden given that a number of tasks were performed on each AVG type as part of the RCT study. Data Processing: Vicon motion analysis software (Nexus; Oxford Metrics, Los Angeles, USA) was utilised to check marker trajectories for breaks that can result from occlusion of the markers. Gaps were filled using algorithmic interpolation between trajectory end points, with no break greater than 20 frames in duration. The data were then filtered using a quintic spline filter using a mean square error of 3, as determined by a residual analysis.38 A valid upper limb three dimensional mathematical model,39 that utilised previously published upper limb segment parameters40 and followed recommended biomechanical procedures was applied in order to calculate upper limb kinematics.7 Three forehands and three backhands were randomly identified from both Kinect and Move game play data. A stroke was defined as the end of the backswing until the end of the forward swing for both backhand and forehand. A stroke was considered a backhand when the dominant hand was on the non-dominant side of the body, with the palm
9 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
facing away from the screen at the start of the forward swing; while a forehand was defined by the dominant hand being on the dominant side of the body, with palm facing towards the screen at the start of the forward swing. Two participants in the TD group (one female and one male) had motion analysis data that were missing or corrupted and were therefore not included in analyses. A custom LabVIEW program (National Instruments, Austin, USA) was used to output hand path distance and speed, and wrist and elbow range of motion for each stroke. Statistical analyses were performed with SPSS (version 21; IBM, Armonk, USA). Following assumption testing, linear mixed-model analyses were utilised to determine differences between DCD and TD groups (between groups comparison) and Move and Kinect AVG type (repeated measures) and any interaction between group and AVG type. Alpha probability was set at 0.05. Results: Hand Path Measures: Children with DCD utilized a significantly slower maximum hand speed than TD children during the backhands irrespective of game type (Mean Difference (MD): 1.20ms-1, 95% Confidence Interval (CI): 0.41 to 1.98, p=0.04) (Table 2), although there was no significant differences in hand path distance. Post hoc power calculations for hand path distance found observed power for detecting a difference in hand path distance between two independent groups (DCD n=20 and TD n=20) using the observed between group difference (Move 0.49; Kinect 0.36; Table 2) and averaged within group standard deviation (Move 0.51; Kinect 0.53; Table 2) was Move 0.841; Kinect 0.549 (Using PS Power and Sample Size Calculations v3.1.2, Vanderbilt University, 2009). Power estimates were for a main effect, not a group (DCD, TD) by game type interaction effect.
10 Downloaded from http://ptjournal.apta.org/ by Dell Schramm on October 15, 2014
Differences between AVG type were also detected. Specifically, children’s average hand path speed using Move was slower than when using Kinect (forehand MD: 0.82ms-1, 95% CI: 0.52 to 1.12, p
