Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Jan 1.
Published in final edited form as: J Aging Phys Act. 2013 Dec 11;23(1):24–33. doi: 10.1123/japa.2013-0126

A comparison of older adults' subjective experience with virtual and real environments during dynamic balance activities

Rachel Proffitt 1,2, Belinda Lange 2, Christina Chen 3, Carolee Winstein 4,5
PMCID: PMC4053487  NIHMSID: NIHMS546010  PMID: 24334299

Abstract

The purpose of this study was to explore the subjective experience of older adults interacting with both virtual and real environments. Thirty healthy older adults engaged with real and virtual tasks of similar motor demands: reaching to a target in standing and stepping stance. Immersive tendencies and absorption scales were administered before the session. Game engagement and experience questionnaires were completed after each task, followed by a semi-structured interview at the end of the testing session. Data were analyzed respectively using paired t-tests and grounded theory methodology. Participants preferred the virtual task over the real task. They also reported an increase in presence and absorption with the virtual task, describing an external focus of attention. Findings will be used to inform future development of appropriate game-based balance training applications that could be embedded in the home or community settings as part of evidence-based fall prevention programs.

Keywords: fall prevention, virtual reality, healthy aging, task engagement

Introduction

More than one-third of people over the age of 65 experience a fall every year (Sherington et al., 2008) and people who experience a fall are two to three times more likely to fall again (Tinetti & Williams, 1997). Systematic reviews of fall prevention studies have established that prevention programs with a physical activity component can reduce falls (Sherrington et al., 2008). Furthermore, there is consistent and high quality evidence to support prevention programs that include home based exercise, home safety assessment and modifications, medication, and optical review. Specifically, there is high quality evidence from a systematic review which included 159 trials with 79,193 participants, comparing a fall prevention intervention for older people living in the community with either no intervention or an intervention not expected to reduce falls (Gillespie et al., 2012). One major outcome of this review was that exercise interventions significantly reduced the risk of sustaining a fall-related fracture (RR 0.34, 95% CI 0.18 to 0.63; 6 trials; 810 participants).

Most of these exercise intervention programs include a focus on improving and maintaining balance and muscular strength. Older adults continue to desire participation in similar activities they have pursued throughout their life as they move into the “older adult” stage of life (Einholf, 2009). These activities include those that have been shown efficacious for fall prevention including aerobic classes, yoga, and strength building. Many maintain an active membership at a local fitness center or YMCA. Some community-based classes are specifically tailored to meet the needs of seniors and many are frequently populated by only those in the older age categories (Avers, 2010; Belza, Snyder, Thompson & LoGerfo, 2010; Gerling, Schlid & Masuch, 2010). However, far fewer classes available in the community are designed specifically for fall prevention and include appropriate balance training exercises that challenge static and dynamic balance abilities (US Department of Health and Human Services).

Video games are increasingly being investigated to assess their potential in a wide variety of applications including fall prevention, rehabilitation and as an engaging social and physical activity for older adults (Garcia, Felix Navarro, Schoene, Smith & Pisan, 2012; Lamoth, Caljouw & Postema, 2011; Rademaker et al., 2013). Specific games and platforms include the Nintendo Wii, Nintendo Wii Balance board and the dance mat (used across several video game consoles) (Laver, George, Thomas, Deutsch, & Crotty, 2011). Video games, in general, provide a fun, engaging and interactive environment for exercises and activities (Przybylski, Rigby & Ryan, 2010). They are also accessible to a wide variety of populations and locales, allowing a person to play in their home environment and have feedback provided for individual performance. The newly released Microsoft Kinect movement-sensing device for the Xbox 360 has potential applications in many areas. As it can track multiple joints and limbs in space, researchers have explored the use of off-the-shelf Kinect games for increasing exercise in children and have adapted the Kinect camera for use in computer interactions (Chang, Chen & Huang, 2011; Gallo, Placitelli, & Ciampi, 2011; Lange et al., 2011a; Lange et al., 2011b; Zafrulla, Brashear, Starner, Hamilton & Presti, 2011). However, researchers only recently have begun to assess the potential beneficial effects, such as improved postural control during static standing with perturbations, of customized games integrated with the Kinect device (Gerling et al., 2010; Lamoth et al., 2011). Furthermore, little is known about the perceptions of virtual environments by the older adult population. Most of the studies examining video game engagement have been completed with younger adults or adolescents (Chen et al., 2011; Cowley, Charles, Black & Hickey, 2008; Przybylski et al., 2010; Whitton, 2011). Associated rating scales have also been developed for younger populations (Brockmeyer et al., 2009). The purpose of this study was to explore the subjective experience of older adults interacting with virtual and real environments and to systematically compare those responses. Findings will be used to inform future development of appropriate game-based dynamic balance training applications that could be embedded as a module in home or community, evidence-based exercise and fall prevention programs currently in use (Thomas, Mackintosh, & Halbert, 2010).

Methods

Participants

Thirty older adults participated in this study based on the following inclusion criteria: 1) over the age of 50; 2) no physical or mental limitations that would prevent them from interacting with a video game; and 3) no medical conditions that prevent watching TV or playing video games. The first 15 participants were recruited exclusively from the Center for Successful Aging Program at the California State University- Fullerton campus. The Program serves approximately 135 older adults in the area surrounding the campus. We contacted persons in the Program previously recruited for a related study via telephone. Those persons were also asked to help recruit persons from the Program via word of mouth. Approximately 75% of those contacted agreed to participate. The second 15 were recruited using convenience sampling methods via telephone in the Los Angeles area and 100% of those contacted agreed to participate.

This study was approved by the Institutional Review Board at the University of Southern California.

Procedures

After obtaining consent and initial demographic information, the participants engaged in two task conditions: Virtual and Real. The order of presentation of the task conditions was counterbalanced across participants.

Description of the virtual task condition

The “Jewel Mine” virtual task was specifically developed to encourage static and dynamic balance activities. The player travels through a “Jewel Mine” and collects gems located at various points in the mine shaft. The gems are placed in positions that are customized for the individual player. This application was originally developed for balance training activities for people following neurological impairment (Lange et al., 2011a). It was developed using the game engine Unity (www.unity3d.com). The virtual task was run on a laptop connected to a Microsoft Kinect camera mounted on a stand. The application used the Open NI/NITE Framework to communicate with the Kinect camera and to translate user movements into game actions. The user was provided with feedback through seeing the full body skeleton on the screen. The first 15 participants viewed the virtual task on a 32” television screen connected to the laptop. The second 15 participants viewed the virtual task on a large projector screen (57.5” high x 92” wide) connected to the laptop. If necessary, participants were asked to use their usual method for vision correction to view the screen. The user movements during the virtual task were recorded with the Microsoft Kinect camera. The initial set-up of the virtual task involves a user calibration to place the virtual targets. The participant was instructed to stand a certain distance from the Kinect camera. This distance varied for each participant depending on the participant's height and was marked with tape on the floor for reference for the participant. A clinician researcher guided the participant through the calibration. There were postural demands within the Virtual task interaction: standing and stepping. For the standing postural demand, the reaching targets were placed in 8 locations (Figure 1) that challenged the participant's balance from a standing position without requiring movement outside the locus of control. For the stepping postural demand, the reaching targets were placed in 8 locations that challenged the participant's balance outside the locus of control, requiring a step. For the stepping postural demand, the participant was instructed to return to the initial starting point after reaching each target. The postural demand tasks were designed to mimic tasks that are typically part of a fall prevention or rehabilitation program. The participants completed 7 sets of 8 targets totaling 56 reaches for each condition (Figure 2). The amount of time each participant was engaged in each task condition varied between participants and ranged between 5 and 10 minutes. The participants were not given any instructions on time limits nor were they told to complete the task as quickly as possible. The participants received visual feedback from the virtual environment. The targets were jewels or gems and when the target was successfully reached, it would disappear with resulting “sparks” on the screen in the location where the target was and a sound indicated that the correct gem was chosen. The participants received no feedback from the researchers on task performance. If a participant had difficulty completing the task in a reasonable amount of time, the directions were repeated, reminding participants that the targets were in the same location as the calibration step.

Figure 1.

Figure 1

Open in a new tab

The location of the 8 virtual targets relative to the player.

Figure 2.

Figure 2

Open in a new tab

The two postural demands within each task condition. Upper left: Standing virtual, Upper Right: Stepping Virtual, Lower Left: Standing Real, Lower Right: Stepping Real.

Description of the real task condition

The real task condition used the same Kinect camera to track and record the user movements. The only feedback provided to the person on the screen was the same visual of the full body skeleton. Instead of reaching for virtual targets, a researcher clinician provided a target with a tennis ball mounted on the end of an aluminum pole. Both standing and stepping postural demands were included with the researcher clinician presenting the targets in similar locations to the virtual targets (Figure 2). The accuracy and order of placement by the researcher clinician were held constant across task conditions and postural demand for all targets (Wade, Proffitt, Requejo, Mulroy & Winstein, 2012).

Outcome Measures

Prior to engagement in the tasks, demographic information was collected using the Health Activity Information Form and included questions about age, gender, ethnicity, education, falls, and video game history. Tendencies for engagement in tasks were assessed using the Immersive Tendencies Questionnaire (Whitmer & Singer, 1998) and the Tellegen Absorption Scale (Tellegen & Atkinson, 1974). The Immersive Tendencies Questionnaire is a 29 item scale that measures the capability or tendency of an individual to be involved or immersed in a task. It has high internal consistency (r = 0.75). Possible scores can range from a minimum of 18 to a maximum of 126. However there are no normative values and the Immersive Tendencies Questionnaire is typically used in studies with adolescent populations and has not been validated for the older adult population. The Tellegen Absorption Scale is a 34 item true-false scale with an average score of 20 (Glisky et al., 1991) that assesses an individual's openness to experience, emotional and cognitive alterations across a variety of situations. It has high internal reliability (r = 0.88), high test-retest reliability (r = 0.91) and moderate construct validity (r = 0.57). Heart rate was monitored with an ePulse 2™ monitor (Impact Sports Technologies) during the task. Baseline, minimum and maximum heart rate were recorded for each task and balance condition. To assess perceived exertion, the Borg Scale of Perceived Exertion (Borg, 1976) was administered following each task condition. The Borg Scale has good test-retest reliability (r = 0.50-0.75) and moderate to high criterion-related validity (r = 0.57-0.72). The participants completed two questionnaires and a semi-structured interview following the two task conditions. The questionnaires were designed to understand the level of engagement, immersion, presence and absorption of the participants during both interactions. The Game Engagement Questionnaire (Brockmeyer et al., 2009) consists of 15 items and 4 subscales measuring the components of engagement including presence, absorption, immersion and flow. It has good short-term test-retest reliability (r = 0.72) and good convergent validity (r = 0.95). Absorption is total engagement in the present experience and includes an altered state of consciousness. Presence involves a normal state of consciousness (awareness of time passing and things in the real environment) and involves having the experience of being in a virtual environment. Immersion is a deep engagement and for this particular questionnaire, is related to media. Flow involves a state of “just right challenge” for the individual that maximizes task enjoyment. The Game Engagement Questionnaire is considered a measure of engagement while playing video-games and was developed and validated on adolescent and young adult populations but not the older adult population. It was also developed for traditional video games designed for entertainment. The questionnaire was modified from the original version for the interactions in this study in order to elicit responses appropriate for the tasks. The Game Experience Questionnaire is a set of 8 Likert Scale questions designed to elicit feedback on the interactions and future use of the game and technology. It was developed from system and computer usability questionnaires (Bangor, Kortum & Miller, 2008; Lewis, 1995) and each question had a 5 point Likert scale rating. There are currently no data on the psychometric properties of the questionnaire.

Qualitative data were obtained using semi-structured interview questions following traditional qualitative methodologies, specifically a grounded theory approach (Strauss and Corbin, 1997). The semi-structured interviews were conducted following the completion of the tasks and outcome measures by members of the research team trained in qualitative methods. The overarching question that guided the development of the interview questionnaire was:

The questions focused on perceptions about the task (virtual or real), expectations, specific task characteristics (rules, visual, audio) that helped or hindered completion of the task, clarity of directions, and recommendations about for whom the task could be beneficial and where the task could be used. Each semi-structured interview lasted approximately 10 minutes. The length of time varied depending on the participant's answers and the number of probe questions asked by the researcher.

Data Analysis

The raw data from the questionnaires were analyzed using SPSS (HIMB-Corp-Released-2012IMB Corp., Released 2012). For the Game Engagement Questionnaire, student's t-tests were used to analyze within subjects between task conditions for the total score as well as the scores for the four respective subscales. The Game Experience Questionnaire was analyzed by question using within subjects student's t tests. Due to a small sample size and multiple comparisons, Bonferroni correction was applied to minimize Type I error.

To analyze the qualitative data, two trained researchers coded the interview data separately, allowing the codes to emerge from the data. In order to increase rigor, the researchers continually met to ensure authenticity of meaning of the codes. For example, a code of “difficulty interacting with system” could be further broken down into more detailed codes such as “difficulty seeing virtual targets” and “difficulty understanding task instructions”. The codes were continually refined as new meanings emerged. Under the broad research question, themes emerged from the data. A theme was developed from the codes as those with similar meanings were grouped together. The more frequently a code appeared, the more the theme was strengthened. The themes were refined with the addition of new data.

Results

Demographics

Basic demographic and health-related information are presented in Table 1. The mean age of the participants was 74.37 ± 8.7 years with the majority of participants in the 66-80 year old cohort. Most participants were Caucasian with about 27% of participants identifying as Asian. None identified as Hispanic or Latino. As expected with this population, the majority were married (57%) and 23% were widowed. All participants had completed at least a high school education and most rated their current health as “good” or better. The reported health conditions are common for this demographic and the majority reported that their health was either unchanged or better than the prior year. As expected for older adults in relatively good physical health, the majority had little or no fear of falling. Only one third of participants reported that they were moderately concerned about falling. Two participants reported having fallen in the past 6 months. These falls were due to uneven surfaces outdoors and a wet floor at the local YMCA pool. Three participants reported falls in the past year. Two of those reporting falls in the past year were the recent fallers and one additional person had become lightheaded while standing in a line and fainted. Eleven out of 30 participants indicated that they played video or computer games. All 11 listed computer games such as solitaire or MahJong. None reported playing games with an off-the-shelf console (eg. Nintendo Wii).

Table 1. Participant Demographics.

Variable n
Age (years)
74.37 ± 8.7
Gender
 Male 10
 Female 20
Current Living Situation
 Alone 10
 With a spouse 18
 With a family member 2
Years of Education
 12+ 30
Current Health Rating
 Excellent 5
 Very Good 13
 Good 11
 Fair 1
 Poor 0
Health Compared to 1 year ago
 Much Better 4
 Somewhat Better 3
 About the Same 19
 Somewhat Worse 4
 Much Worse 0
Health Conditions
Orthopedic Conditions 21
Neurological Conditions 6
Cardiovascular Conditions/Events 19
 Endocrine Conditions 19
Open in a new tab

Quantitative

Heart rate and perceived exertion

The baseline, minimum and maximum heart rate levels for the participants for each task condition are reported in Table 2. All three heart rate measures were similar across the task conditions. There were no significant differences between the two task conditions or between the standing and stepping tasks of each environment. The perceived level of exertion for each task condition is listed in Table 2. Both of these average ratings fall into the “endurance training zone” on the Borg Scale of Perceived Exertion. A level of 11-12 corresponds to fairly light effort. There was no significant difference in rating of perceived exertion between the two task conditions.

Table 2. Heart Rate and Perceived Exertion for the Virtual and Real Task Conditions.
Mean ± SE Mean ± SE
Virtual Real
Outcome Measure Standing Stepping Standing Stepping
Baseline Heart Rate 83.3 ± 13.4 88 ± 12.5 87.8 ± 12.4 85.4 ± 11.7
Minimum Heart Rate 73.7 ± 10.6 75.3 ± 8.1 74.5 ± 7.4 73.4 ± 6.7
Maximum Heart Rate 113 ± 20.3 118 ± 17.1 115 ± 26.2 111.5 ± 17.9
Perceived Level of Exertion 11.97 ± 2.1 10.48 ± 2.3
Open in a new tab

Task engagement as measured by questionnaires

The mean total score on the Immersive Tendencies Scale for the participants was 108.6 ± 17.8, indicating that the participants were on the higher end of the scale. The participants on average scored slightly below the average score for the dichotomous version of the Tellegen Absorption Scale (Glisky et al., 1991) with a mean total score of 17.4 ± 6.5. Therefore, the participants rated themselves as less likely to be open to absorbing experiences (such as hypnosis).

Significant within-subjects differences were found for the absorption (p = 0.009) and presence (p = 0.000) subscales of the Game Engagement Questionnaire between the two task conditions (Table 3). The total Game Engagement score was also significant (p = 0.016) within subjects between the two task conditions. Within subjects differences on the flow and immersion subscales were non-significant. Participants rated their perceived level of absorption and presence significantly higher in the virtual task condition compared to the real task condition, as well as their overall task engagement between the two conditions.

Table 3. Within Subjects t-test for the Game Engagement Questionnaire Responses.
Scale Mean ± SE p-value
Virtual Real
Absorption 7.0 ± 2.9 6.4 ± 2.5 0.009‡
Flow 19.2 ± 5.1 18.7 ± 5.1 0.396
Immersion 3.5 ± 1.2 3.2 ± 1.3 0.111
Presence 9.0 ± 2.8 7.6 ± 3.1 0.000‡
Total 38.7 ± 10.0 35.9 ± 10.7 0.016†
Open in a new tab

Note.

†

Statistically significant at the p < 0.05 level.

‡

Statistically significant at the p < 0.01 level.

The individual questions on the Game Experience Questionnaire were analyzed within subjects between task conditions (Table 4). Subjects responded significantly more positively to the virtual task condition on the questions regarding the use of the task as exercise (p = 0.035), the task being more engaging than the exercises done previously (p = 0.001), future interaction with the task (p = 0.001), and motivation to keep interacting with the task (p = 0.010). Subjects reported more frustration interacting with the virtual task than the real task (p = 0.002).

Table 4. Within Subjects t-test of Game Experience Questionnaire Responses.
Game Experience Questionnaire Question Mean ± SE p-value
Virtual Real
I would like to use the [task condition] as exercise. 2.79±2.54 3.21±1.36 p = 0.035†
The [task condition] was more engaging than typical exercises I have done before. 2.79±2.03 3.57±1.51 p = 0.001‡
The [task condition] was more strenuous than typical exercises I have done before. 4.14±0.87 4.32±0.67 p = 0.129
I could see myself interacting with the [task condition] in the future. 2.75±2.19 3.42±1.37 p =0.001‡
It was hard to understand the directions for interacting with the [task condition].* 4.32±1.12 4.57±0.92 p = 0.083
I felt frustrated while interacting with the [task condition].* 4.00±1.93 4.75±0.42 p = 0.002‡
I was motivated to keep interacting with the [task condition]. 2.07±1.40 2.64±1.65 p = 0.010‡
I feel as though I would benefit from interacting with the [task condition]. 2.57±1.74 2.89±1.14 p = 0.071
Open in a new tab

Note.

*

higher number rating corresponds with a more positive response.

†

Statistically significant at the p < 0.05 level.

‡

Statistically significant at the p < 0.01 level.

Qualitative: Perceptions of the Virtual and Real Task Conditions

Locus of attention

When describing the real task condition, participants focused on the task movements. More than half of the participants stated that they enjoyed the “reaching” part of the real task. For example, one participant described the challenge of the task as “…being able to reach each one without… controlling my balance, not falling over and being able to reach forward and come back.” In comparison, only one third of participants mentioned a movement aspect of the virtual task condition. Participants also felt that the real task condition had similar movements to exercises that focused on reaching, stretching, upper body motion and balance. They felt that because of this, the real task condition could potentially be beneficial for those who need to work on reaching and balance. When asked to compare the real task condition to activities that they had done or currently do for exercise, participants felt that it was similar to balance exercises and Tai Chi. For example, one participant stated: “Same thing with tai chi and regular exercises. Doing the reaching exercises and balance exercises”.

In contrast, participants focused on aspects of the virtual environment when describing the virtual task condition. In particular, 40% of the participants stated that they felt that the on-screen visuals were their favorite part of the virtual task condition as a whole. Participants also felt that because of the visual environment, the virtual task was beneficial for making improvements on coordination, eye-hand coordination, vision and cognition.

Perceived differences

Most participants had no expectations prior to engaging in either task condition. The majority of participants had no least favorite part of either task condition and enjoyed both. Participants responded unanimously when asked about the similarities between the two task conditions: As stated by one participant: “They're both the same in as much as I was reaching for. It was all physical activity reaching for an object.” Only one participant mentioned that the level of challenge was the same between the task conditions. The other similarities mentioned by two participants were the goal to be reached and the fact that they had to follow directions in both task conditions.

When asked to describe the differences between the two task conditions, participants focused on a variety of aspects of the tasks. Nearly two-thirds of participants commented on the differences between the two task conditions. Table 5 summarizes the major themes that emerged from the qualitative data.

Table 5. Summary of Major Themes of Task Condition Differences.
Real Task Condition Virtual Task Condition
Ball is a real object Feedback on the screen when the target was hit
Depth is easier to understand More challenging
Involved a second person Required more concentration
Gave a sense of control Required more coordination
Simplistic More exercise
The second person provided feedback Less patterned/more random
Is just a task, not a game Feeling of interacting with the figure on the screen
Open in a new tab

As evidenced by the thematic differences summarized in Table 5, participants had differing perceptions of the level of challenge of the two task conditions. Participants enjoyed the challenge of the virtual task condition and some stated that they tried to “do it fast” or employ a strategy to touch the targets as efficiently as possible. This is interesting considering that the testers did not give any instructions to the participants with regards to speed or timing of the movements required for the task. The enjoyment associated with the cognitive challenge of the virtual task condition was the main reason that participants stated as to what motivated them to continue with the virtual task condition. In contrast, the main reason for continuing with the real task condition was “because I was part of a research study”.

When participants were comparing both task conditions to activities that they had previously done or currently did for exercise, most thought that in either case, the task condition was not as physically challenging: “I found it not nearly as difficult as my general exercise”. For the real task condition, two participants did state they felt the task required more reaching and stretching than they had previously done. For the virtual task condition, some participants thought that it was more cognitively challenging: “Well I think it was mentally challenging, but as far as physical exercise goes I don't think it's as challenging”. Interestingly, some people stated that they expected a high level of challenge from the virtual task condition as well as differences in the technology. Although neither task condition was described as being very challenging, slightly more participants (10%) rated their confidence lower in the virtual task condition. However this was primarily due to visual perception of the 3-D location of the targets. The main problems reported in the virtual task condition were in regard to “finding the Jewel in space”. More participants who interacted with the smaller screen (32”) had difficulty locating the virtual target. Participants also mentioned difficulty seeing the objects on the screen in addition to perceiving the 3-D location of the virtual target. Similarly, more participants who interacted with the smaller screen reported this problem. Participants did not mention any problems with the real task condition.

Future use

Participants were asked to think about the use of the tasks beyond the research study. Over three quarters of participants stated that they would not engage in the real task condition again. In contrast, two thirds of participants stated that they would engage with the virtual task condition again. Participants listed locations such as their home, exercise classes or community centers as places in which they could see the virtual task being used. Family members and friends were mentioned as others to interact with while engaging with the virtual task. When asked to think about recommending the task conditions to others, the locus of attention for the task was apparent. Participants would recommend the real task condition to people who are sedentary or inactive and those who want to participate in a task that involves exercise or reaching movements: “People who want to have some movement exercise without it being too heavy or difficult”. In contrast, participants recommended the virtual task condition to specific family members or friends, “people who have an interest in technology” and “someone who wants to have fun”. As one participant stated: “I would love to see my brother in law do it. Because he's really athletic. He's older.”

Task condition preference

Eighteen of the thirty participants (60%) preferred the virtual task over the real task. Five participants preferred the real task over the virtual task (17%). The remaining seven participants (23%) did not prefer either task or thought that both tasks were equally preferable. Furthermore, participants had no negative perceptions of the virtual task condition. Five of the 30 participants felt that the real task was boring, poor quality and less exciting than the virtual task. Some felt that the virtual task was more interesting and motivating than other activities that they had done previously for exercise. When asked to explain their preference for either the virtual or real task condition, participants listed a number of reasons summarized in Table 6.

Table 6. Participant Stated Reasons for Task Condition Preference.
Real Task Condition Virtual Task Condition
I like to work with people The task has a technology aspect
I had more control over the task Visuals/Visual Feedback
I could relate to it It is fun
It provided a good reminder that I want to exercise It is more challenging
It is more exciting
It is more engaging
It is satisfying to pop the gems
Open in a new tab

Feedback on the virtual task

The participants in the current study had several suggestions for modifying the virtual task condition for improved playability and engagement. Participants desired a more lifelike avatar and instant feedback on success when reaching for targets. The participants also wanted to see more rewards in the game for success and rewards placed throughout the game. Although the participants had positive responses to the virtual environment, they wanted options for customization of the graphics such as the colors of the targets, background and target objects. Lastly, the participants wanted multiplayer options so that they could engage in the task with their grandkids, friends and during group classes (both local and remote).

Discussion

The purpose of this study was to explore the subjective experience of older adults interacting with virtual and real environments and compare the responses in order to develop appropriate game-based balance training applications for fall prevention in the future. To our knowledge, this is the first study examining the subjective experience of the participant in both real world and virtual environments. Most previous research comparing virtual and real environments focuses on specific aspects of performance, such as subcomponents of reaching (Stewart, Gordon & Winstein, 2013). The use of outcome measures designed to elicit information on task engagement allowed us to make comparisons between the two environment conditions. Qualitative methods allowed for an in-depth exploration of how the participants perceived the two environments. The findings add richness to the literature on engagement in virtual environments and explore a sub-population (older adults) that receives little attention in the virtual reality and video game literature. Our major findings include differences in the focus of attention brought about by the two environments and a significant preference for the virtual compared with the real environment. Each is discussed below.

A Focus on Movement

Although the physical movements required were held constant across the task conditions (Wade et al., 2012), participants focused more on the physical movement of the task when speaking about the real task condition and less so for the virtual task condition. Participants mentioned the reaching movement as their favorite part of the real task condition and focused on other aspects of the virtual task condition besides movement, such as the visuals. They also compared the real task condition to other movement activities, such as specific exercises and balance activities. The focus on movement was clearly seen when the participants were asked what aspects of the task would be beneficial for them. One possible explanation for the focus on movement in the real task condition is that the participants were more absorbed and immersed in the virtual task condition compared to the real task condition as indicated by a significant difference between the two task conditions on the immersion and absorption subscales of the Game Engagement Questionnaire. When someone is fully “absorbed”, attention is focused on the task and task-related environment and an awareness of the real environment is minimized. In contrast, presence involves a normal state of consciousness (awareness of time passing and things in the real environment) and involves having the experience of being in a virtual environment (Brockmeyer et al., 2009). Although the Game Engagement Questionnaire has not been validated for the older adult population, the concepts of absorption, flow, immersion and presence are not age dependent.

The almost exclusive reliance on visual feedback during the virtual task condition may explain the difference in levels of absorption and presence. The participants had to synchronize the corresponding movement of the on-screen avatar to their movements in 3-dimensional space and rely on the avatar for visual feedback, including the accuracy of their performance. There was no provision of tactile feedback, such as that which would be provided when the hand actually contacts a tennis ball. The focus of visual attention was primarily on the screen, as this was necessary to successfully complete the task, rather than on the body in space. Other studies have shown that shifts from center of pressure increase when visual feedback (in a Head Mounted Display) is inconsistent with vestibular and somatosensory feedback but does not lead to an increased sense of presence (Lott Bisson, Lajoie, McComas & Sveistrup, 2003). The study by Lott and colleagues (2003) had limitations in the technology used for the virtual environments but the findings are indicative of how the perception of the environment and the feedback within the environment, virtual or real, can change behavior and perception of the task.

Studies completed in real-world settings, particularly in sport-related tasks, have shown that an external focus of attention can lead to improved task performance and learning (Chiviacowsky, Wulf & Wally, 2010; Porter, Wulf, Nolan & Ostrowski, 2010; Wulf, McConnel, Gartner & Schwarz, 2002). In the virtual task condition, the participant's responses in the semi-structured interview suggest that they focused on aspects of the virtual environment, such as the colored jewels, and were less aware of the actions of their body in space. The participants did see an avatar on the screen that represented their actions in real-time. The full body avatar is one of several options for on-screen representation of participant movements. A study is currently being conducted to understand the difference in participant performance with respect to different on-screen representations of participant actions. These findings will provide further evidence and understanding of how to structure feedback within the virtual environment to promote maximum task engagement as well as performance.

Preference for the Virtual Task Condition

Not only were participants more absorbed in the virtual task condition, there was also a clear preference for the virtual task over the real task. Participants reported that they had fun interacting with the virtual task. Furthermore, the participants felt that the real task was boring and not engaging. These qualitative findings are consistent with the significantly different total score on the Game Engagement Questionnaire between the two task conditions, with participants rating the virtual task condition as being more engaging. Among the many reasons that the participants listed for preferring the virtual task condition over the real task condition was the perceived level of challenge. Some participants stated that they thought the virtual task was more challenging and even “more random” in the placement of targets. This is an interesting finding considering that the researchers were able to hold the 3-dimensional target position constant across task conditions (Wade et al., 2012). The extent of reach was held constant across conditions so the actual physical demand was not different. Some participants explicitly stated that they felt the virtual task provided more of a cognitive challenge; others simply mentioned challenge without identification of type. The rating of perceived exertion was not significantly different between the two task conditions as well as the measures of heart rate. Furthermore, there was no significant difference within subjects on the flow subscale of the Game Engagement Questionnaire. Flow, as defined by the authors of the questionnaire, is the feelings of enjoyment that occur when a balance between skill and challenge is achieved in the process of performing an intrinsically rewarding activity (Brockmeyer et al., 2009). Given the nature of both task conditions and the level of physical challenge, it is understandable that the participants did not perceive entering a flow state during either task condition. It is possible that a person might perceive a virtual task to be more difficult because of a physical condition (such as osteoarthritis or multiple sclerosis) compared to someone who does not have a limiting physical condition. However, these data did not emerge from the qualitative findings. Other results from our group that compared the attentional demand of the target/environment conditions using a dual-task paradigm showed significantly higher attentional demand in the virtual compared to the real condition (Chen et al., 2013). This provides complimentary support for the greater perceived challenge in the virtual condition derived from the qualitative preference analysis.

Limitations

There were several limitations in this study. First, we recruited a relatively small, convenience sample of active older adults. The first 15 participants were recruited from the Center for Successful Aging. The members of the Center participate in a wide range of physical, mental and social activities throughout the month. Many members are involved in weekly exercise classes and volunteer on a regular basis. The second 15 participants were also very active in their local community. Many participants were still regularly employed or held volunteer positions. Most of the participants had not experienced a fall or rehabilitation of any kind. Some had rehabilitation following elective surgeries, such as a knee replacement. It is difficult to generalize the findings of this study to a population that would be challenged by a dynamic balance task. However we argue that these findings provide useful information for designing a game that appeals to the older adult population and to people who might engage in fall prevention exercises in the community or home. Related work by our group and others has emphasized that fear of falling and the ensuing disablement and inability to care for oneself (loss of autonomy) with age are concerns for this population. Furthermore, older adults have expressed interest in using similar technologies in community settings, such as group exercise classes or as a station in a gym circuit. (Proffitt & Lange 2013). Another study examined the safety and feasibility of the Nintendo Wii as an unsupervised balance intervention for older adults in the home setting and with positive results (Agmon, Perry, Phelan, Demiris & Nguyen, 2011). Interactive technologies focused on fall prevention exercises are likely an acceptable and useful strategy in this population (Cisneros, Dyer-Chamberlerlain, & Hickie, 2012; Proffitt & Lange, 2013).

Another limitation is that the “game” or virtual task condition was somewhat artificial for the purposes of determining “game engagement”. The movements were very controlled through markings on the floor and instructions from the testers. The participants had to return to a “home” foot location during the stepping tasks and they also had to bring their reaching arm back to their side after each target was achieved. Typical games (those not designed for dynamic balance control) allow for more choice on the part of the player through minimal restrictions on acceptable movements. The specific constraints used here were chosen to allow systematic comparability between the two conditions to directly address the research questions. Recommendations for future design of a game to engage dynamic balance control include: consideration of player agency throughout the process, progressive and performance-based adaptation of the challenge level, and an engaging virtual environment that supports maximum player immersion in the task (including screen size no less than 32 inches and both auditory and visual feedback).

Finally, this study examined the perceptions of older adults based on a single session experience. Personal preference and perceptions of tasks are likely to change over time including both the general perception of each task condition as well as perceptions of the task conditions relative to one another. Efficacious fall prevention programs require long term interventions and education in order to maintain gains or the current level of function. The results of this single session study may not generalize to a longer term intervention in older adults. Longitudinal research examining the perceptions of older adults with virtual environments is necessary to validate the findings of this study.

Conclusion

The purpose of this study was to explore the subjective experience of older adults interacting with both virtual and real environments and compare the responses in order to develop appropriate game-based balance training applications for fall prevention in the future. The results of the study show that even though the virtual task was simple, the participants reported higher levels of absorption and presence as compared to a real environment. The participants also had less of a focus on the movement during the virtual task and reported being engaged with and preferred the virtual environment. In order to create an effective game for fall prevention in older adults that includes a movement component, these findings should be incorporated with findings from related studies so that engagement is maximized.

What is the experience of the older adult with games, both technology-based and non technology-based?

Acknowledgments

This work was support by a NIH T32 grant: Training in Rehabilitation Efficacy and Effectiveness Trials (TREET) (PI: Clark) (5T32HD064578-02) and by a National Institute on Disability and Rehabilitation Research (NIDRR) grant: Optimizing Participation Through Technology: Rehabilitation Engineering Research Center (OPTT:RERC) (PI: Winstein) (H133E080024).

References

  1. Agmon M, Perry CK, Phelan E, Demiris G, Nguyen HQ. A pilot study of Wii Fit exergames to improve balance in older adults. Journal of Geriatric Physical Therapy. 2011;34:161–167. doi: 10.1519/JPT.0b013e3182191d98. [DOI] [PubMed] [Google Scholar]
  2. Aronson RE, Oman RF. Views on exercise and physical activity among rural-dwelling senior citizens. The Journal of Rural Health. 2004;20(1):76–79. doi: 10.1111/j.1748-0361.2004.tb00010.x. [DOI] [PubMed] [Google Scholar]
  3. Avers D. Community-based exercise programs for older adults. Topics in Geriatric Rehabilitation. 2010;26(4):275–298. [Google Scholar]
  4. Bangor A, Kortum PT, Miller JT. An empirical evaluation of the system usability scale. Intl Journal of Human–Computer Interaction. 2008;24(6):574–594. [Google Scholar]
  5. Belza B, Snyder S, Thompson M, LoGerfo J. EnhanceFitness, an innovative community-based senior exercise program. Topics in Geriatric Rehabilitation. 2010;26(4):299–309. [Google Scholar]
  6. Borg G. Simple rating methods for estimation of perceived exertion. Physical work and effort. 1976:39–46. [Google Scholar]
  7. Brockmyer JH, Fox CM, Curtiss KA, McBroom E, Burkhart KM, Pidruzny JN. The development of the Game Engagement Questionnaire: A measure of engagement in video game-playing. Journal of Experimental Psychology. 2009;45(4):624–634. [Google Scholar]
  8. Chang YJ, Chen SF, Huang JD. A kinect-based system for physical rehabilitation: A pilot study for young adults with motor disabilities. Research in Developmental Disabilities. 2011;32(6):2566–70. doi: 10.1016/j.ridd.2011.07.002. [DOI] [PubMed] [Google Scholar]
  9. Chen M, Kolko BE, Cuddihy E, Medina E. Modeling but NOT measuring engagement in computer games. 2011, June; Proceedings of the 7th international conference on Games+ Learning+ Society Conference; ETC Press. pp. 55–63. [Google Scholar]
  10. Chen YA, Chung YC, Proffitt R, Wade E, Winstein C. Is there more benefit than fun and enjoyment when engaging in a VR-based target reaching exercise in non-disabled elders? 2013 Manuscript in preparation. [Google Scholar]
  11. Chiviacowsky S, Wulf G, Wally R. An external focus of attention enhances balance learning in older adults. Gait & posture. 2010;32(4):572–575. doi: 10.1016/j.gaitpost.2010.08.004. [DOI] [PubMed] [Google Scholar]
  12. Cisneros H, Dyer-Chamberlain M, Hickie J, editors. Independent for Life: Homes and Neighborhoods for an aging America. Austin, TX: University of Texas Press; 2012. [Google Scholar]
  13. Cowley B, Charles D, Black M, Hickey R. Toward an understanding of flow in video games. Computers in Entertainment (CIE) 2008;6(2):20. [Google Scholar]
  14. Einholf CJ. Will the Boomers volunteer during retirement? Comparing the Baby Boom, Silent and Long Civic cohorts. Nonprofit and Voluntary Sector Quarterly. 2009;38(2):181–199. [Google Scholar]
  15. Gallo L, Placitelli AP, Ciampi M. Computer-Based Medical Systems (CBMS), 2011 24th International Symposium on. IEEE; 2011. Jun, Controller-free exploration of medical image data: Experiencing the Kinect; pp. 1–6. [Google Scholar]
  16. Garcia JA, Felix Navarro K, Schoene D, Smith ST, Pisan Y. Exergames for the elderly: Towards an embedded Kinect-based clinical test of falls risk. Studies in Health Technology and Informatics. 2012;178:51–57. [PubMed] [Google Scholar]
  17. Gerling KM, Schild J, Masuch M. Exergame design for elderly users: the case study of SilverBalance. 2010, November; In Proceedings of the 7th International Conference on Advances in Computer Entertainment Technology; pp. 66–69. ACM. [Google Scholar]
  18. Gillespie LD, Robertson MC, Gillespie WJ, Sherrington C, Gates S, Clemson LM, Lamb SE. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev. 2012;9 doi: 10.1002/14651858.CD007146.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Glisky ML, Tataryn DJ, Tobias BA, Kihlstrom JF, McConkey KM. Absorption, openness to experience, and hypnotizability. Journal of Personality and Social Psychology. 1991;60(2):263. doi: 10.1037//0022-3514.60.2.263. [DOI] [PubMed] [Google Scholar]
  20. Lamoth CJ, Caljouw SR, Postema K. Active video gaming to improve balance in the elderly. Stud Health Technol Inform. 2011;167:159–164. [PubMed] [Google Scholar]
  21. Lange B, Chang C, Suma E, Newman B, Rizzo A, Bolas M. EMBC 2011: Integrating Technology and Medicine for a Healthier Tomorrow. Boston, MA: 2011a. Development and evaluation of low cost game-based balance rehabilitation tool using the Microsoft Kinect sensor. [DOI] [PubMed] [Google Scholar]
  22. Lange B, Suma EA, Newman B, Phan T, Chang C, Rizzo A, Bolas MT. Leveraging unencumbered full body control of animated virtual characters for game-based rehabilitation. Proceedings of the 2011 international conference on Virtual and mixed reality: systems and applications - Volume Part II; 2011b. pp. 243–252. [Google Scholar]
  23. Laver KE, George S, Thomas S, Deutsch JE, Crotty M. Virtual reality for stroke rehabilitation (review) Cochrane Database of Systematic Reviews. 2011;9 doi: 10.1002/14651858.CD008349.pub2. [DOI] [PubMed] [Google Scholar]
  24. Lewis JR. IBM computer usability satisfaction questionnaires: psychometric evaluation and instructions for use. International Journal of Human-Computer Interaction. 1995;7(1):57–78. [Google Scholar]
  25. Lott A, Bisson E, Lajoie Y, McComas J, Sveistrup H. The effect of two types of virtual reality on voluntary center of pressure displacement. CyberPsychology & Behavior. 2003;6(5):477–485. doi: 10.1089/109493103769710505. [DOI] [PubMed] [Google Scholar]
  26. Porter JM, Wulf G, Nolan R, Ostrowski E. Instructions that promote an external focus of attention benefit agility performance. Journal of Sport and Exercise Psychology. 2010;32:S119–120. [Google Scholar]
  27. Proffitt R, Lange B. User centered design and development of a game for exercise in older adults. International Journal of Technology, Knowledge and Society. 2013;8(5):95–112. [Google Scholar]
  28. Przybylski AK, Rigby CS, Ryan RM. A motivational model of video game engagement. Review of General Psychology. 2010;14(2):154–166. [Google Scholar]
  29. Rademaker M, Fox EJ, Lange B, Thigpen M, Fox C. Made for Rehab: Feasibility and effect of a clinician adapted video game using the Kinect™ on balance and falls in a woman with parkinsonism. American Physical Therapy Association, Combined Sections Meeting; San Diego, CA. 2013. Jan, [Google Scholar]
  30. Sherrington C, Whitney JC, Lord SR, Herbert RD, Cumming RG, Close JC. Effective exercise for the prevention of falls: A systematic review and meta-analysis. Journal of the American Geriatrics Society. 2008;56(12):2234–2243. doi: 10.1111/j.1532-5415.2008.02014.x. [DOI] [PubMed] [Google Scholar]
  31. Stewart JC, Gordon J, Winstein CJ. Planning and adjustments for the control of reach extent in a virtual environment. Journal of NeuroEngineering and Rehabilitation. 2013;10(1):27. doi: 10.1186/1743-0003-10-27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Strauss AL, Corbin JM. Grounded theory in practice. Thousand Oaks, CA: Sage Publications, Inc; 1997. [Google Scholar]
  33. Tellegen A, Atkinson G. Openness to absorbing and self-altering experiences (“absorption”), a trait related to hypnotic susceptibility. Journal of Abnormal Psychology. 1974;83(3):268. doi: 10.1037/h0036681. [DOI] [PubMed] [Google Scholar]
  34. Tinetti ME, Williams CS. Falls, injuries due to falls, and the risk of admission to a nursing home. New England Journal of Medicine. 1997;337(18):1279–1284. doi: 10.1056/NEJM199710303371806. [DOI] [PubMed] [Google Scholar]
  35. U.S. Department of Health and Human Services. Healthy People 2020. Washington, DC: [Accessed October 21, 2013]. Office of Disease Prevention and Health Promotion. [Google Scholar]
  36. Wade E, Proffitt R, Requejo P, Mulroy S, Winstein C. Visuomotor researching behavior to virtual and real targets depends on postural requirements in healthy elders; 2012, October; Poster presented at the American Congress of Rehabilitation Medicine Annual Conference; Vancouver, Canada. [Google Scholar]
  37. Whitton N. Game engagement theory and adult learning.  Simulation & Gaming. 2011;42(5):596–609. [Google Scholar]
  38. Witmer BG, Singer MJ. Measuring presence in virtual environments: A presence questionnaire. Presence. 1998;7(3):225–240. [Google Scholar]
  39. Wulf G, McConnel N, Gärtner M, Schwarz A. Enhancing the learning of sport skills through external-focus feedback. Journal of Motor Behavior. 2002;34(2):171–182. doi: 10.1080/00222890209601939. [DOI] [PubMed] [Google Scholar]
  40. Zafrulla Z, Brashear H, Starner T, Hamilton H, Presti P. American Sign Language recognition with the Kinect. Proceedings of the 13th International Conference on Multimodal Interfaces (ICMI); 2011. pp. 279–286. [Google Scholar]

RESOURCES