This iterative study consisted of four phases: (1) prototype design and personnel training, (2) round 1 pilot-testing, (3) prototype modifications, and (4) round 2 pilot-testing. A detailed timeline can be found in Figure 1. The study was led by a multidisciplinary team trained in computer science, kinesiology, and developmental psychology. Prior to data collection, the research team determined a priori the study end points (metrics of success) that would inform the number of iterative testing and modifications. Specifically, the prototype would be considered complete once the product (1) engaged users in physical activity (ranging from light to moderate to vigorous) for at least 50% of the time spent completing an online academic assignment on average and (2) achieved an average score of at least 60 on the System Usability Scale (SUS) [23,24]. While developing the prototype, the research team had several discussions regarding the movements that the prototype would use, with the limitations of the Meta Quest 3 (Reality Labs) in mind. The team aimed to encourage movement of the upper and lower body while being mindful to minimize repetitive movements as they could cause injury. The Meta Quest 3 does not support feet tracking, but it does support hand and head tracking. On the basis of these constraints, the research team determined that punching would be used to elicit upper-body movements, enabled via hand tracking, and squatting and jumping would elicit lower-body movements, enabled via head tracking.

From November 2023 to November 2024, members of the research team designed a mixed reality prototype program that, when paired with a Meta Quest 3 headset, would allow students to use their body as a “mouse” to complete an online academic assignment. Personnel training took place in November 2024. During this time, 4 middle school students used the prototype to allow the research team to refine the study protocol and prototype program. Informal conversations with the students indicated that the prototype was enjoyable and worked as expected but could benefit from the addition of audio cues. This feedback was incorporated into the prototype prior to pilot-testing. Round 1 of pilot-testing took place in February 2025 at a local middle school. On the basis of the results of round 1, minor modifications were made to the prototype in March 2025. In April 2025, round 2 of pilot-testing took place at a laboratory space on the Louisiana State University campus.

To be eligible to participate in the pilot-testing of the prototype, students had to be aged between 10 and 15 years. Students were excluded if they did not have parental consent to participate.

The first round of testing took place at a local school that serves as a center for educational innovation and research. Prior to recruiting any participants, members of the research team met with the school administrators to provide them with details of the study and a demonstration of the prototype. After this initial meeting, the principal spoke with the sixth-grade teachers and determined that a math class would be the best option to participate in the study. Consent forms were sent home to parents. Those who agreed to allow their children to participate were asked to return a signed copy of the consent form to the school. Members of the research team collected consent forms from the school and then worked with the principal and math teacher to schedule a date and time to conduct the test. A total of 22 students returned signed consent forms and were eligible to participate in round 1 of pilot-testing.

The second round of testing took place when many of the local schools were on spring break. In an effort to not retest the same participants, local middle school students were recruited via word of mouth and flyers. In total, 10 parents returned signed consent forms, and their children were eligible to participate in round 2 of pilot-testing. After receiving signed consent forms, research team members worked with parents to schedule a testing day, which took place at a university laboratory space.

During both rounds of pilot-testing, members of the research team explained to the students, in an age-appropriate manner, what participation in the study would entail. If students were interested in participating in the study, they were asked to sign an assent form. Students were then provided with a short video tutorial created by the research team that introduced the assignments they would be working on and instructed them how to scroll and click. Members of the research team asked a few questions to ensure that the students knew how to complete the proper movements. A demonstration was provided if needed. Research team members then fitted the students with accelerometers and a Meta Quest 3 headset. Research team members recorded start and stop times. Additionally, the research team kept note of any exceptional factors that could impact testing (ie, motion sickness or ill-fitting headsets).

A modified virtual reality web browser was used for the online environment during pilot-testing. Upon putting on the Meta Quest 3 headset, students were presented with a large web page (Figure 2) with a yellow button to the left of the screen and a red button to the right. To begin the assignment, students had to press the yellow button, which set their default height. The default height value was used to detect when students jumped or squatted. Once the default height was established, participants could scroll down by jumping and scroll up by squatting. Clicking was a multistep process. Students first aimed with their left hand, which would move a clicker over the screen (Figure 3A). Once the clicker was hovering over the link or button, the participant would make a pinching motion (index finger and thumb closing) to lock the clicker in place. If the student accidentally placed the clicker in the wrong spot, they could press the red button on the right side of the screen to reset it. Once the clicker was locked, a large target appeared (Figure 3B); participants were encouraged to punch the target 3 times to finalize the click. Students were instructed to make hard, fast punches. However, interactions were triggered through collision, not speed. Thus, students could trigger the click with slower punches. Despite these activity-focused interactions, the virtual reality browser otherwise functioned as a normal web browser. It could display arbitrary web content, but for the purpose of this study, the browser was restricted to only showing the standardized academic assignments and the SUS.

During both rounds, students were required to complete academic assignments. For round 1, they completed a vocabulary flash card exercise matching 30 terms to their meaning. If terms were connected to the incorrect meaning, students were required to try again. In addition to the vocabulary assignment, students completed 5 reading comprehension prompts, answering questions related to each prompt. In round 2, they completed the same 2 exercises, but the vocabulary exercise was changed from a flash card format, which did not require scrolling, to a web questionnaire format, which did require scrolling. The updated vocabulary quiz only consisted of 15 questions and did not require students to correct mistakes. This change was made to encourage more physical activity and de-emphasize the notion that students needed to select a “correct” answer, which was not the purpose of this iterative pilot study. In both rounds, the order of these 2 exercises was randomized across participants.

The statistical software SPSS (version 29; IBM Corp) was used. Descriptive statistics were calculated for all variables. Mann-Whitney U tests were used to identify differences in SUS scores (ie, boys vs girls and rounds 1 vs 2). The α level was set at .05 a priori.

This study was approved by the institutional review boards of the Pennington Biomedical Research Center (2024-052; approved on October 3, 2024) and Louisiana State University (IRBAM-24-1094; approved on December 5, 2024) and conformed to the latest Declaration of Helsinki. Written parental consent and child assent were obtained before data collection. To maintain privacy and confidentiality, students were assigned a study-specific identification number and a randomly generated code to connect physical activity data with headset use. Students were provided with small tokens of appreciation (ie, notebooks, small fidget toys, stickers, and lip balm) and snacks (ie, fruit, water, granola bars, and fruit snacks) as compensation for their participation.

Results from round 1 and 2 of pilot-testing are presented separately to highlight the changes that occurred during this iterative development process. Demographic characteristics of users in each round are reported in Table 1.

A total of 22 middle school students (mean age 12.3, SD 0.4 years; n=15, 68.2% boys) participated in round 1 of pilot-testing. Students spent 6.9 (SD 2.7) minutes (46.0%) of the session, which lasted a mean 15 (SD 0) minutes, engaging in physical activity, as shown in Table 2. The average acceptability score was 80.7 (SD 16.1), which was in the highest range for acceptability. Figure 4 [23,24], used with permission, shows how SUS scores are usually interpreted. The average acceptability score of 80.7 lies in the “excellent” range. There were no significant differences (P=.79) in acceptability scores between boys (mean 83.17, SD 11.47) and girls (mean 74.58, SD 24.67).

Semistructured interviews and direct observation revealed several key insights. As suggested by the high SUS score, students indicated that they would be willing to use the prototype to complete school assignments both at home and at school. Students enjoyed most of the features (eg, punching, jumping, sound effects, and ease of use), but some design aspects could be improved. For example, the cursor (pinching and aiming) required precise aim to use; even a small movement could cause a student to miss a button or link. Students suggested making a larger pointer or expanding the selection area to fix this issue. Proctors noticed that some students, instead of punching the target, slapped the target 3 times by making a small movement with their hand.

Following round 1 of pilot-testing, the research team made several adjustments to the prototype to increase physical activity and improve usability. First, the mechanism to scroll was modified to encourage more movement to meet the physical activity goal. In round 1, participants held a squat to scroll the page up. This was modified in round 2 so that, for every squat, the modified web browser would only scroll up a fixed distance, thereby requiring participants to squat repeatedly to scroll up. Second, to require full arm movement, the software was modified to require students’ hands to return to their body after punching before their next punch. Finally, an internal clock was added to allow researchers to precisely analyze physical activity data. In addition to these modifications, researchers changed the vocabulary flash cards to a vocabulary quiz that would require students to scroll the page more. The reading comprehension questions did not change in round 2.

Round 2 participants comprised 10 local middle school students (mean 11.82, SD 0.63 years; n=6, 60% boys). Students were physically active for 5.8 (SD 3.1) minutes (62.4%) of the session, which lasted a mean 9.3 (SD 2.41) minutes (Table 2). The average acceptability score for round 2 was 73.8 (SD 17.2; Figure 4), which remained in the highest category for acceptability and was not statistically significantly different from that in round 1 (P=.16). There was no significant difference (P=.61) in round 2 acceptability scores between boys (mean 78.75, SD 8.48) and girls (mean 66.25, SD 25.37). Nearly half (4/10, 40%) of the participants reported the user-friendliness of the prototype program as “best imaginable,” which was the highest score. Similarly to round 1 of pilot-testing, the students stated during the semistructured interview that they would use the prototype to complete school assignments. However, a few students stated that, due to the weight of the headset, they would only use the prototype to complete short (15-20 minutes) assignments. Jumping and punching remained the favorite features of both boys and girls. Additionally, the issues with aiming the cursor remained a common complaint.

This iterative study resulted in an acceptable mixed reality prototype program that engaged students in physical activity while they completed online academic assignments. To our knowledge, this is the first study to test mixed reality technology as a method for increasing physical activity while completing schoolwork. During round 2 of pilot-testing, students were physically active for most of the time it took them to complete the online assignments, and the students also provided high scores for user satisfaction. Semistructured interviews with students further emphasized that the mixed reality prototype program was a digital learning tool that they would be willing to use at school and at home. Of note, teachers and school administrators were also interested in deploying the mixed reality prototype program in their classes. By conducting 2 rounds of testing with minor modification in between, researchers were able to demonstrate that this prototype can engage students in physical activity 46% to 62.4% of the time it takes to complete online assignments. This ultimately equates to approximately 6 to 7 minutes of physical activity in a 10- to 15-minute academic task. Overall, the results of this study are congruent with those of popular virtual reality exergames such as Beat Saber or Gorilla Tag [21].

Traditionally, school-based physical activity interventions have been designed to incorporate moderate to vigorous physical activity into classroom settings. These methods often result in modest changes in moderate to vigorous physical activity during intervention lessons [28]. Unfortunately, teachers and students might be hesitant to engage in moderate to vigorous physical activity in classroom settings. Teachers often have a limited time to incorporate physical activity into lessons [29], and the classroom size might make it difficult to engage in higher-intensity physical activity. Qualitative data indicate that students are hesitant to engage in higher-intensity physical activity because they do not want to sweat or change clothes [30], and teachers are reluctant due to limited space [31] and competing priorities [32].

Instead, replacing sedentary computer time that is already part of classroom instruction with physical activity computer time may be a less disruptive and more feasible strategy to increase minutes of physical activity. In this study, the developed prototype program elicited similar physical activity responses to those elicited by traditional interventions [28], all while working with a digital learning tool that can be easily incorporated into the learning environment.

These results contribute promising evidence to the nascent research that suggests that mixed reality technology can be used to increase physical activity [20]. Most school-based trials to date that have tested mixed reality, augmented reality, or virtual reality have focused on incorporating activity-promoting technology into physical education classes [22], whereas this prototype was designed to be integrated into a traditional academic subject matter classroom with academic assignments such as reading comprehension. One mixed reality trial integrated into physical education focused on improving children’s motivation to be active and their motor skills and physical fitness but did not evaluate minutes spent engaged in physical activity [33]. A trial that did focus on increasing physical activity was conducted by Ahn et al [34] and involved a virtual fitness buddy ecosystem that would allow children aged 6 to 11 years to interact with a virtual dog to set and meet self-determined physical activity goals. Results indicated that the ecosystem was effective at significantly increasing light physical activity and decreasing sedentary behavior during an after-school program [34]. Similarly to the work by Ahn et al [34], this study found that using a mixed reality prototype primarily increased light physical activity. The integration of movement, even light physical activity, has been shown to benefit physical health and academic outcomes [16-18].

The findings of this study indicate that mixed reality technology might encourage more physical activity during a traditionally sedentary school day. The broader evidence base for exergaming and active video games is mixed. Many have found that these methods are effective at promoting physical activity and decreasing sedentary behavior [20,33-35]. Furthermore, there is evidence suggesting that exergaming and active video games might improve sleep hygiene and health-related quality of life [36]. However, one small trial of 49 children found significant increases in sedentary behavior but no significant increases in other screen time [37]. The overall quality of evidence in active video game research is low, and researchers should aim for increasing methodological rigor and more balanced reporting of findings [38].

This study was designed to examine physical activity while wearing the headset; thus, the difference between typical student classroom activity and activity using our prototype is not known. Although typical in-class assignments are generally sedentary, future trials should examine students’ physical activity levels with and without the prototype, such as comparing physical activity while using a traditional computer vs the mixed reality headset. Additionally, researchers should examine whether integration of the prototype contributes to a meaningful increase in physical activity throughout the school day and whether the effect can be sustained over weeks and months. The assignment time was set for 15 minutes. However, with the addition of an internal clock in round 2, it was evident that students completed the assignment in approximately 9 minutes, and some indicated that the weight of the headset precluded longer sessions. The difference in assignment completion time is most likely attributed to the change in the academic assignment from 30 vocabulary flash cards to a 15-item vocabulary quiz. Future studies should examine the duration of time that students will tolerate wearing the headset while balancing engagement in physical activity and acceptability. Finally, future research should examine acceptability from the classroom teachers and school administrators to ensure that the prototype is helpful and not disruptive to classroom instruction. Anecdotal information from the pilot test indicated that the school administrators were positive about the prototype and the math and science teachers were interested in integrating the mixed reality headset into the classroom, but this should be explored in a structured interview in future trials.

The continued use of digital learning tools in educational settings has led to increases in screen time and sedentary behavior [7,8,11]. These trends have challenged educators to identify new ways to incorporate movement into classroom settings. The results of this iterative study indicate that this mixed reality prototype program might be a pathway to increase physical activity in the classroom in a way that is acceptable and enjoyed by students. Future research is needed to determine longer-term usability and effectiveness.