If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Howcroft J, Klejman S, Fehlings D, Wright V, Zabjek K, Andrysek J, Biddiss E. Active video game play in children with cerebral palsy: potential for physical activity promotion and rehabilitation therapies.
To evaluate the potential of active video game (AVG) play for physical activity promotion and rehabilitation therapies in children with cerebral palsy (CP) through a quantitative exploration of energy expenditure, muscle activation, and quality of movement.
Single-group, experimental study.
Human movement laboratory in an urban rehabilitation hospital.
Children (N=17; mean age ± SD, 9.43±1.51y) with CP.
Participants played 4 AVGs (bowling, tennis, boxing, and a dance game).
Main Outcome Measures
Energy expenditure via a portable cardiopulmonary testing unit; upper limb muscle activations via single differential surface electrodes; upper limb kinematics via an optical motion capture system; and self-reported enjoyment via the Physical Activity Enjoyment Scale (PACES).
Moderate levels of physical activity were achieved during the dance (metabolic equivalent for task [MET]=3.20±1.04) and boxing (MET=3.36±1.50) games. Muscle activations did not exceed maximum voluntary exertions and were greatest for the boxing AVG and for the wrist extensor bundle. Angular velocities and accelerations were significantly larger in the dominant arm than in the hemiplegic arm during bilateral play. A high level of enjoyment was reported on the PACES (4.5±0.3 out of 5).
AVG play via a low-cost, commercially available system can offer an enjoyable opportunity for light to moderate physical activity in children with CP. While all games may encourage motor learning to some extent, AVGs can be strategically selected to address specific therapeutic goals (eg, targeted joints, bilateral limb use). Future research is needed to address the challenge of individual variability in movement patterns/play styles. Likewise, further study exploring home use of AVGs for physical activity promotion and rehabilitation therapies, and its functional outcomes, is warranted.
Two extremely impactful and high-potential applications of AVGs are: (1) promoting increased daily physical activity for children with and without disabilities, and (2) increasing engagement and participation in physical rehabilitation therapies. This article focuses on assessing the potential of AVGs in these 2 areas for children with cerebral palsy (CP).
Ambulatory children with CP are less likely to be physically active than their typically developing peers.
As such, these children are at an increased risk for overweight/obesity, diabetes, cardiovascular disease, cancer, joint disease, and musculoskeletal pain, in addition to reduced psychological and emotional health.
indicated that typically developing children and youth increase their energy expenditure (EE) by 222% (±100%) during AVG play. Metabolic equivalents for task (METs) varied from 2 (for XaviX Bowling) to 5 (for XaviX J-Mat and EyeToy Knockout).
found that adults with CP can achieve moderate levels of activity during Wii boxing (5 MET) and Wii tennis (4.5 MET). No research to our knowledge has examined EEs during AVG play in children with CP. EE during activities such as walking or running have been shown to differ between adults and children, likely because of cocontraction of agonist and antagonist muscle groups that have not developed at the same level of coordination and efficiency as in healthy adolescents or adults.
The strength of AVGs lies in their potential to provide an entertaining and immersive virtual reality in which repetitive movements and/or complex motor tasks can be encouraged with increasing levels of in-game difficulty and rewards. These characteristics are key to eliciting neuroplastic changes that are so critical to motor rehabilitation.
While considered promising in rehabilitation circles, very little is actually known regarding the movement patterns and muscle activations elicited by AVGs, particularly low-cost, commercially available systems like Nintendo's Wii or Microsoft's Kinect. While these systems may extend the availability and use of virtual reality therapy (VRT) to home settings and therapy clinics, their potential therapeutic value is not well understood, particularly relative to dedicated VRT systems that are equipped with more accurate (and expensive) body capture technologies and offer more detailed performance metrics. Most investigations to date have focused on customized VRT technologies with highly specific equipment and therapeutic goals
found that Wii Sports games could improve visual perceptual processing, postural control, and functional mobility in an adolescent with spastic diplegic CP. While contributing important findings, the data collected in this study were not sufficient to identify the specific elements of game play that resulted in these functional improvements. A better understanding of the kinematics of AVG play is therefore needed to identify games with therapeutic potential, to aid in prescription guidelines, and to design improved games for rehabilitation therapies.
Functional outcomes of physical activity and rehabilitation therapies are linked in many ways. Improved aerobic capacity and endurance achieved through physical activity can improve a disabled child's ability to participate in physical therapies. Likewise, improved muscle strength, coordination, and range of motion obtained through rehabilitation may heighten performance and self-efficacy in physical activities. The quality of physical activities and therapies depends on many factors including: aerobic intensity, range and quality of movements, and levels of muscle activity and strengthening, not to mention psychological motivators such as enjoyment.
As such, this study uses a multifaceted assessment protocol to explore the potential of AVGs for the promotion of physical activity and therapy in children with CP. Specifically, the goals of this study are: (1) to determine the intensity of physical activity that can be achieved during AVG play by monitoring levels of EE, (2) to investigate the therapeutic potential of AVG play by quantifying upper limb kinematics, and (3) to explore practical considerations surrounding the use of AVGs for physical activity promotion and therapy by assessing: potential risks for muscle strain injuries, individual variability during AVG play, and the extent of user enjoyment.
Children (N=17; 10 boys and 7 girls) participated in the study. Demographic and anthropometric information are presented in table 1. Fourteen children had previous experience with the Wii Sportsa and 2 were familiar with Dance Dance Revolution (DDR).b To ensure that all participants could independently play the AVGs in a standing position, only children who were categorized as level I or II on the Gross Motor Function Classification System (GMFCS)
and the following exclusion criteria: (1) fractures or orthopedic surgery in the last 6 months, (2) botulinum toxin treatment in the last 3 months, (3) diagnoses of epilepsy or chronic asthma, (4) any injury/disability that would make moderate exercise unsafe, and (5) any visual, cognitive, or auditory disability that would interfere with game play. All children were recruited from Holland Bloorview Kids Rehabilitation Hospital and the ErinoakKids Centre for Treatment and Development. Approval for this study was granted by local ethics review boards. Informed assent and consent was obtained from the child and guardian, respectively.
Table 1Participant Characteristics
All Subjects (N=17)
GMFCS level I (n)
GMFCS level II (n)
NOTE. Values are mean ± 1SD or as otherwise indicated.
was calibrated prior to each session according to manufacturer protocols. Specifically, calibration of oxygen and carbon dioxide sensors was completed using a sample gas of known concentrations (16% oxygen and 5% carbon dioxide). Respiratory volume was calibrated using a 3-liter volume calibration syringe. A delay calibration was also performed to ensure adequate response time to a breathing cycle. To establish baseline measures, the child sat quietly and watched a self-selected, low stimulus, G-rated DVD for 20 minutes, as recommended in previous studies.
A near-resting, baseline-steady state was established in the first 10 minutes of rest, while baseline energy levels were extracted from data recorded in the final 10 minutes. EE was determined based on measures of V̇o2 and expired carbon dioxide (V̇co2) using the following equation:
Motion data were collected using a 7-camera Vicon MX motion capture system.f Light-weight, reflective markers (14-mm spheres) were placed at anatomical landmarks bilaterally on the upper extremities (3rd metacarpophalangeal joint, medial and lateral wrist, lateral aspect of the forearm, medial and lateral condyles of the humerus, lateral aspect of the upper arm) and shoulder girdle (left and right acromions, C7, and midpoint of the clavicles).
Muscle Activity and Establishment of Maximum Voluntary Exertions
Electrical activity of the upper trapezius, triceps, biceps, flexor carpi radialis (FCR), and wrist extensor bundle (WE) of the dominant limb was recorded during maximum voluntary exertions (MVEs) and during AVG play via surface electromyography. Of note, the dominant arm (defined as the one used for writing) is primarily used during typical AVG play in an unstructured environment (ie, at home).
Electrode sites were prepared by light abrading and cleaning with alcohol. Reusable single differential surface electrodesg with a 1-cm interelectrode distance were placed centrally over the muscle belly, which was located via anthropometrics and palpation. Of note, electrodes were placed and connection wires secured such that the range of movement was not constrained or limited in any way. Raw surface electromyography signals were sampled at 1250Hz via the Bagnoli Desktop EMG Systemg with 1-k gain. MVEs were obtained using the Biodex System 2,h which was calibrated before each use and positioned for each MVE according to manufacturer instructions. Three isometric MVEs for each of the 5 muscle groups were obtained. Each contraction lasted 5 seconds with a 45-second rest between contractions. The largest contraction was used as the MVE. Verbal encouragement was provided throughout the contractions.
Active Video Games
Four age-appropriate and popular AVGs (Wii bowling, Wii tennis, Wii boxing, DDR Disney Dance Grooves) were investigated in this study. Wii bowling and Wii tennis are played with the Wii remote held in the dominant hand. In Wii boxing, players use both the Wii remote and an additional hand controller called the nunchuck. Lastly, DDR Disney Dance Grooves is played with the Wii remote, nunchuck, and dance mat.i These games were selected to provide a representative sample of current AVGs involving a range of upper and lower limb movements. The interested reader is referred to Deutsch et al
for a detailed qualitative description of the movements associated with each game. Previous studies with able-bodied children also indicate that the games selected should elicit a range of activity intensity.
Once all the sensors were attached, as shown in figure 1, and baseline tests were completed as described, the children were allowed to familiarize themselves with the games for a maximum of 5 minutes before playing. Instructions were kept as simple as possible (ie, When you move the remote, the arm/tennis racket in the game will copy that movement). The child played each game in a randomized order for 8 minutes on a preselected beginner level with a rest period of 5 minutes between each game. Eight minutes was a sufficient period of time for the children to meaningfully engage in several (2−3) matches/rounds of each game with brief (<10s) intervals between each match. Energy, muscle, and motion data were collected for the entire 8 minutes of game play. The child completed the OMNI Scale of Perceived Exertion (OMNI) after playing each game.
For energy measures, the first 30 seconds of data associated with each game was eliminated to allow for subject acclimatization. The mean ± SD were then calculated for the remaining data. The MET is computed by normalizing the average V̇o2 during AVG play to the baseline V̇o2.
Upper limb kinematics for each game were established based on a typical 1-minute segment of play chosen such that rest periods (ie, instant replays of game activities and pauses between rounds) were minimized. The positions of the retro-reflective markers were manually identified using Vicon Workstation.f Marker trajectories were filtered using a second-order Butterworth filter with a cutoff frequency of 6Hz.
The upper body biomechanical model consisted of 7 segments (shoulder girdle, left upper arm, right upper arm, left lower arm, right lower arm, left hand, and right hand). Vicon Bodybuilderf was used to calculate angular displacements, velocities, and accelerations according to the biomechanical model described by Rab et al.
Specifically, the following were calculated: wrist flexion-extension, forearm pronation-supination, elbow flexion-extension, shoulder flexion-extension, shoulder abduction-adduction, and shoulder internal-external rotation. Angles for each upper limb joint (ie, wrist, elbow, and shoulder) had to be present for at least 70% of the 1-minute segment to be included in the subsequent analyses.
For muscle activation data, the root mean square (with a moving average window of 75ms and a 50% overlap) of all raw electromyographic signals were taken and normalized to the MVEs. Outliers that exceeded 2 SD from the mean were removed as noise associated with motion artifacts. Participants were divided according to whether they used realistic or adapted movements during play to allow for an exploratory assessment of the effect of movement style on muscle activity. A realistic style used movements that could be successful in the real-world version of the game (eg, a traditional tennis swing). Alternatively, the adapted style was characterized by movements that could only simulate and achieve similar outcomes to realistic movements in the virtual game (eg, a wrist flick). Movement styles were classified by 2 independent reviewers from video recordings. Interrater reliability of observations were assessed with Cohen kappa. Any differences between interrater observations were discussed until an agreement was reached.
Lastly, the overall PACES score was determined by averaging the scores for each individual item of the scale according to established guidelines.
All statistical analyses were performed using SPSS version 17j with a significance level of P<.05. Comparisons (ie, of energy measures, electromyographic activity, kinematic parameters, and enjoyment scores) between the 4 AVGs were conducted via analysis of variance with a Bonferroni correction. A Friedman test was used to determine if significant difference existed between OMNI scores for different games. If significant, a Wilcoxon signed-rank test was performed.
Potential for Physical Activity Promotion: EE During AVG Play
Of the 17 participants, 13 children (GMFCS level I) wore the Cosmed K4b
cardiopulmonary testing unit, while the remainder found the mask too uncomfortable to wear. Mean levels for the energy measures (eg, V̇o2, MET, EE, and heart rate) and median scores for perceived exertion on the OMNI are presented in table 2.
Table 2Energy Measures at Rest and During AVG Play
DDR Disney Dance Grooves
% increase in EE from rest
Heart rate (bpm)
% increase in heart rate from rest
NOTE. Values are mean ± 1SD or median (interquartile range).
Abbreviations: bpm, beats per minute; NA, not applicable.
As expected, all energy measures were significantly lower for the baseline state than during AVG play (P≤.002). Energy measures were significantly higher for games involving the lower body (eg, DDR, P≤.001) and/or those requiring high frequency, bilateral movements of the upper limb (eg, Wii boxing, P≤.039) when compared with Wii bowling. Of note, Wii bowling is a unilateral game characterized by low frequency movements. The MET for Wii tennis (high frequency, unilateral movements) was also significantly higher than for Wii bowling (P=.041). Moderate levels of physical activity were achieved by 15% (n=2) of participants in Wii bowling; 23% (n=3) in Wii tennis; and 62% (n=8) in DDR. Fifty-four percent (n=7) of participants played Wii boxing at a moderate intensity, while 8% (n=1) reached a vigorous level (MET=6.69).
The median OMNI score for perceived exertion during AVG play ranged from 2 (very easy) for Wii bowling to 4 (just feeling a strain) for DDR and Wii boxing. Perceived exertions mirrored EE data. A Friedman test revealed significant differences in OMNI scores between the AVGs (P<.023). Wilcoxon comparisons indicated that perceived exertion for Wii bowling was significantly lower than for Wii tennis (P=.031) and Wii boxing (P=.022).
Potential for Therapy: Movement Patterns During AVG Play
Upper limb kinematics were assessed for the subset of participants with hemiplegic CP (n=14). Data associated with 1 participant could not be analyzed because of a corrupted static frame. The AVGs eliciting the largest ranges of movements, angular velocities, and accelerations are summarized in table 3. Of note, Wii bowling required the least wrist activity with significantly lower degrees of extension, flexion, and lateral deviation than in Wii tennis (flexion and extension, P≤.021) and DDR (flexion and lateral deviation, P≤.006). Meanwhile, Wii boxing elicited higher angular velocities and accelerations for wrist movements (ie, extension/flexion and wrist lateral/medial deviations) compared with the other games (P<.05). Elbow extension (of the dominant limb) was greatest during Wii bowling, while DDR required significantly less elbow extension than the other games (P≤.037).
Table 3AVGs That Elicited the Largest Ranges of Movements, Angular Velocities, and Accelerations
Practical Considerations in the Use of AVGs for Physical Promotion and Therapy
Muscle activity and potential risk for overstrain
Electromyographic data were collected and analyzed for all 17 participants to characterize muscle activity and potential risk for overstrain of the dominant limb. Figure 2 depicts mean muscle activation levels (normalized to the MVE) during AVG play. Mean muscle activations for all muscles were below 20% of the MVE. Peak muscle activations did not exceed 90% MVE at any time during game play. In general, the WE was the most active muscle for all games. Wii boxing consistently elicited the highest activation levels for all muscles groups (P<.05).
Individual variability during AVG play
It was observed that an adapted (nonrealistic) movement style was used by 47% (n=8) of children. The kappa coefficient for interrater reliability of distinguishing between movement styles (ie, realistic or adapted) was .841±.089, indicating excellent agreement. Realistic and adapted movement styles were associated with varying levels of muscle activation. For instance, upper trapezius muscle activity was greater when children punched realistically (14.0%±2.6%) as opposed to when an adapted (ie, a small punching movement primarily involving the wrist) technique was used (9.0%±3.4%, P=.004). Secondly, FCR activation levels were highest when an adapted movement style was adopted. For example, FCR activations during Wii tennis reached 11.8%±4.4% for the adapted movement (ie, a flick of the wrist) compared with 7.4%±3.5% for the realistic swing (P=.038).
As expected, intraindividual variability was also observed when comparing dominant and hemiplegic limb kinematics. Maximum angles of wrist lateral deviation were significantly greater for the dominant arm compared with the hemiplegic arm in all bilateral games (P≤.027). A similar trend was observed for wrist extensions. Likewise, the degree of elbow extension was generally greater in the dominant arm than in the hemiplegic arm, but only reached statistical significance for Wii boxing (P=.031).
Enjoyment of AVG play
The average PACES score for all AVGs was 4.5±0.3 (out of 5), indicating a high level of enjoyment. The positive statement most highly rated by participants was “I enjoy it” at 4.8±0.4. The negative statement that received the most agreement (3.1±1.2) was “I feel as though I'd rather be doing something else.”
This study provides a multifaceted evaluation of AVG play to assess its potential for physical activity and therapy. Table 4 provides a descriptive comparison of the AVGs investigated with respect to activity levels and suitability for different therapeutic goals. We summarize our key findings in the subsequent paragraphs.
Children with mild CP can attain moderate levels of physical activity during AVG play with games that require full body movements (ie, Wii boxing and DDR). These results are inline with studies of able-bodied children.
AVGs should not, at this stage, be regarded as a replacement for more vigorous forms of physical activity, nor do they appear to be of value for muscle strengthening of the dominant upper limb given the low mean levels of muscle activation observed (ie, <20% of MVE). For strengthening to occur, electromyography activity must exceed 40% of the MVE for a given muscle, while muscular endurance may be improved when electromyographic levels are between 20% and 40% MVE.
While all AVGs have the potential for increased motor learning, different games may be strategically preferable depending on the goals of the therapy. While ineffective for building endurance or strength, the range of mean muscle activation observed in this study (ie, 7%−18% MVE) can result in neuromuscular reeducation, which is important for physical rehabilitation.
Features of the AVGs that support suitability for motor learning and therapy are as follows. All games encouraged repetitive movements of varying frequencies from low (Wii bowling) to high (Wii boxing). Repetition is a mainstay in fostering neuroplastic change as is the provision of feedback. The Wii provides vibrotactile (ie, through the Wii remotes), visual (ie, by way of the on-screen avatar), auditory, and cognitive (ie, through game scores and performance) feedback to the user. The high levels of enjoyment reported suggest that motivation, the third ingredient to enacting neuroplasticity, is also established. Finally, while this study did not investigate the kinematics of advancing game levels specifically, experiential observations would suggest that movement complexity increases with respect to hand-eye coordination and speed/frequency of movements as the game advances and the player improves. Additionally, different games can be selected based on the complexity of the movement desired. A recent article by Damiano
calls for increased emphasis on activity in conjunction with structured therapy sessions, noting potential benefits such as improved physical functioning, increased cognitive, social, and emotional development, and development/restoration of neural pathways. Damiano emphasizes the importance of providing the support for patients to incorporate activity into their everyday lifestyles. AVGs may indeed offer 1 avenue toward achieving this goal.
Table 4Summary of AVG Characteristics With Respect to Intensity of Physical Activity, Nature of Movements, Most Active Muscle Groups, Frequency of Movements, and Suitability for Potential Therapeutic Goals
This study presents data to help guide the selection of games for targeted therapies focused on specific joints or movements. For example, kinematic analyses suggest that Wii boxing may be a good choice for encouraging and training faster wrist movements, particularly wrist extensions. While this finding may not be intuitive when considering movements used in real-life boxing (eg, primarily elbow and shoulder driven), it does emphasize the importance of conducting kinematic analyses to examine the actual movements elicited during natural play of AVGs, wherein adapted movement styles and simulated play conditions (eg, absence of force of impact or glove) may lead to largely different biomechanics. Conversely, Wii bowling required significantly less wrist activity than other games. This may be important information to note when selecting appropriate games at the start of a new therapy program for children with very limited wrist mobility. Of note, for all games, the WE was the most active muscle. The WE is of particular importance for children with CP, because their natural wrist position often involves a higher degree of wrist flexion.
poses a significant challenge to rehabilitation and the advancement of functional skills, which hinges on consistent practice and the use of the hemiplegic limb. Children were observed to functionally engage both upper limbs while playing Wii boxing and DDR. Hemiplegic limb movements were carried out at a lower velocity and acceleration than those of the dominant limb, which is not surprising considering that reduced movement speed is a common consequence of CP.
Wii boxing, or comparable AVGs, may be an effective motivational environment for encouraging and training increased movement speed of the hemiplegic limb, in addition to bilateral use of the limbs, because in-game success is strongly linked to these 2 metrics.
Practical Considerations in the Use of AVGs for Physical Promotion and Therapy
Risk for injury
To date, at least 5 case studies have reported on injuries occurring in able-bodied populations during AVG play.
As such, investigations into the safety of these games for children with CP is warranted, both with respect to their dominant and hemiplegic limbs. In this study, we observed relatively low levels of muscle activity in the dominant limb (<20% MVE), that at no point exceeded 90% of the MVE. Additionally, range of motion of all joints of the dominant limb was well within typical norms associated with upper limb movements in able-bodied individuals.
This biomechanical exploration into the safety of AVGs, while preliminary, is particularly important given the added reliance of children with hemiplegia on their dominant limb for activities of daily living. This study was not able to assess risks associated with extensive periods of play, nor should it be regarded as a replacement for stringent epidemiologic assessments. However, it does suggest that, inherently, AVG play should be a relatively low impact activity for the child's dominant limb. Future studies focused on the hemiplegic limb are essential to further explore potential risks of AVG play.
Similar to traditional physical and occupational therapies, AVGs do encourage repetitive, goal-oriented movements. Conversely, AVG play is largely unstructured and considerable variations in movements can be observed.
One of the key difficulties in quantifying the physical activity or therapeutic benefits gained from AVGs is the considerable variability that exists in participant strategies to succeed in-game (ie, realistic or adapted movements) and in different in-game parameters (ie, opponent difficulty, sequence of in-game events). Therefore, while an AVG may have a specific therapeutic benefit in the majority of cases, this may not be the case for all individuals or occur every time the game is played. In this study, 47% of participants were observed adopting an adapted movement style during AVG play. This adapted style likely minimized physical effort by favoring stronger, more distal muscles (eg, FCR), while maximizing in-game rewards. In determining the potential of AVGs for rehabilitation purposes, it is therefore not only important to consider the type of games prescribed, but also to train and provide in-game rewards for appropriate movement styles. The interested reader is directed to Berry et al
Results of this study suggest that AVGs can be highly enjoyable for children with mild CP. This is an encouraging result supporting the potential of AVGs for self-directed use and play in the home environment.
Future studies are needed to track player enjoyment in the home and over extended periods of time.
Although this study is an important first step toward evaluating the potential of AVGs in children with CP, it has a number of limitations. GMFCS level II (n=1) and diplegia (n=2) were not equally represented in this study. As such, our results likely cannot extend to these subpopulations. Additional studies are needed with larger samples to increase validity and allow for differences between CP subpopulations and demographics (sex, age) to be evaluated. All participants played the AVGs at a beginner level in an attempt to ensure consistency. While enjoyment levels (as measured on the PACES) and interest in the activity did not appear to be affected by the level of play selected, the beginner level may have been too easy for some of the more experienced players, potentially reducing their energy and muscle activation levels and altering movement patterns. Equations used to calculate EE were based on data from adults. Inaccuracies are likely present in these calculations when applied to children with CP. Nonetheless, this equation is currently used to determine EE in children during AVG play
Also of note, the degree of wrist radial and ulnar deviation could not be assessed with the current model. Additionally, we did not compare the range of motion achieved during AVG play to the active range of motion of the hemiplegic limb. Lastly, it is noted that children were wearing a number of sensors while playing the AVGs. The measurement equipment did not appear to impede their movements or cause additional fatigue, nor were complaints voiced pertaining to the equipment with the exception of the 3 children who did not wish to wear the cardiopulmonary testing unit. Regardless, the presence of the equipment and observer are certainly different from conditions that would be experienced in the home setting. It is important to note that the experimenters did not provide motivational encouragement to the players or attempt to alter their natural play. Nevertheless, future studies are needed to further study AVG play in natural home settings.
This study provides preliminary insights into performance variables (eg, EE, muscle activity, and kinematics) associated with natural, undirected AVG play. Future work is needed to outline how/if these games can be used in rehabilitation therapies and to establish an evidence base to support or refute their efficacy in terms of functional improvements, enjoyment, and adherence. Strategies to optimize the therapeutic value of the Wii and other AVG platforms are needed. Future studies with alternative AVG systems that use camera-based, motion-capture technologies (ie, Microsoft's Kinect) are of interest in order to observe if these systems can minimize movement variability during AVG play and provide more detailed performance measures. The ability to better control and track in-game movements may lead to the development of more effective games for targeted rehabilitation outcomes. It is also important to assess whether and how movements change as the difficulty of games increase, and whether AVGs are appropriate to engage and challenge children over time. The ability to increase the intensity of a therapeutic exercise is an important requirement for rehabilitation therapy
and key to the successful use of AVGs in therapeutic exercises.
This study assessed electromyography activity of muscles in the dominant arm of children with CP. A study that compares muscle activation levels in the dominant and hemiplegic arms would be vastly interesting. Inclusion of key leg muscles would also be of interest. Finally, although 1 study has assessed muscle activation levels in typically developing children during AVG play, this study
did not include the MVEs, making comparisons impossible. The importance of normalizing electromyographic data to the MVEs in future studies is emphasized in order to enable comparisons between studies and populations.
One of the most attractive features of AVGs for physical activity promotion and physical therapy is that they can be played in the home and are largely unaffected by barriers (ie, weather, accessibility, transportation) that commonly reduce participation in other forms of physical activity. Home studies investigating how/if AVGs are integrated into daily activity patterns and their use over time are much needed. While children's enjoyment of the games appears to be high, it is not clear if factors such as boredom or frustration would prevent long-term efficacy of AVGs or what measures should be taken to promote continued interest in AVGs for physical activity promotion and rehabilitation. Additionally, it is unclear how AVGs affect current activity patterns. If AVGs supplant sedentary screen time, they may positively contribute to a child's health and activity. If, however, they displace participation in more active pastimes, then they may be a negative, competing influence.
While not a replacement for structured exercise and physical therapy regimes, AVGs can offer an enjoyable opportunity for light to moderate physical activity in children with CP and the practice of complex motor activities. There are many opportunities for future research in this emerging field to better understand and manage individual variability and to quantify functional outcomes and use of AVGs in the home. Future development and optimization of AVG technologies may usher in a new age in physical rehabilitation where virtual environments provide an arena for neuroplastic change in the comfort of one's home.
SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.
We thank Elizabeth Han, MASc, Delbert Hung, MASc, Ajmal Khan, BASc, and Jomy Varghese, MSc for aiding with data collections. We also thank Holland Bloorview Kids Rehabilitation Hospital and ErinkoakKids Centre for Treatment and Development. The results presented in this article were presented at Jennifer Howcroft's Master's defense and thesis in August 2011.
Active gaming: definitions, options, and implementation.
Supported by the Canadian Institutes of Health Research (grant no. MOP-102485 ), Natural Sciences and Engineering Research Council of Canada, Ontario Ministry of Training, Colleges and Universities, and the Holland Bloorview Kids Rehabilitation Hospital Foundation.
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.