JMIR SERIOUS GAMES Estrella et al Original Paper A Serious Game to Self-Regulate Heart Rate Variability as a Technique to Manage Arousal Level Through Cardiorespiratory Biofeedback: Development and Pilot Evaluation Study Tony Estrella1,2, MSc; Carla Alfonso1,2, MSc; Juan Ramos-Castro3, PhD; Aitor Alsina4, PhD; Lluis Capdevila1,2, PhD 1 Laboratory of Sport Psychology, Department of Basic Psychology, Universitat Autónoma de Barcelona, Barcelona, Spain 2 Sport Research Institute, Universitat Autònoma de Barcelona, Barcelona, Spain 3 Group of Biomedical and Electronic Instrumentation, Department of Electronic Engineering, Universitat Politècnica de Catalunya (UPC), Barcelona, Spain 4 Department of Information and Communications Engineering, Universitat Autònoma de Barcelona, Barcelona, Spain Corresponding Author: Lluis Capdevila, PhD Sport Research Institute Universitat Autònoma de Barcelona Edifici N, Planta 1 Barcelona, 08193 Spain Phone: 34 93 581 3329 Email: lluis.capdevila@uab.cat Abstract Background: Heart rate variability biofeedback (HRVB) is an established intervention for increasing heart rate variability (HRV) in the clinical context. Using this technique, participants become aware of their HRV through real-time feedback and can self-regulate it. Objective: The aim of this study was 2-fold: first, to develop a serious game that applies the HRVB technique to teach participants to self-regulate HRV and, second, to test the app with participants in a pilot study. Methods: An HRVB app called the FitLab Game was developed for this study. To play the game, users must move the main character up and down the screen, avoiding collisions with obstacles. The wavelength that users must follow to avoid these obstacles is based on the user’s basal heart rate and changes in instantaneous heart rate. To test the FitLab Game, a total of 16 participants (mean age 23, SD 0.69 years) were divided into a control group (n=8) and an experimental group (n=8). A 2 × 2 factorial design was used in each session. Participants in the experimental condition were trained in breathing techniques. Results: Changes in the frequency and time domain parameters of HRV and the game’s performance features were evaluated. Significant changes in the average RR intervals and root mean square of differences between adjacent RR intervals (RMSSD) were found between the groups (P=.02 and P=.04, respectively). Regarding performance, both groups showed a tendency to increase the evaluated outcomes from baseline to the test condition. Conclusions: The results may indicate that playing different levels leads to an improvement in the game’s final score by repeated training. The tendency of changes in HRV may reflect a higher activation of the mental system of attention and control in the experimental group versus the control group. In this context, learning simple, voluntary strategies through a serious game can aid the improvement of self-control and arousal management. The FitLab Game appears to be a promising serious game owing to its ease of use, high engagement, and enjoyability provided by the instantaneous feedback. (JMIR Serious Games 2023;11:e46351) doi: 10.2196/46351 KEYWORDS serious game; heart rate variability; biofeedback; mobile health; mHealth; app; mobile phone https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 1 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al HRVB, they can be a way to teach participants to control HRV Introduction through breathing techniques in an interactive environment. Background Objectives Heart rate variability (HRV) represents the temporal variation With the information provided in the Background section, the between successive heartbeats (RR interval), and it is measured aim of this study was 2-fold: first, to develop a serious game in milliseconds [1]. HRV is a biomarker of cardiovascular [2] that applies the HRVB technique to teach participants to and mental [3] health, reflecting the balance between the self-regulate HRV and, second, to test the app with participants parasympathetic and sympathetic branches of the autonomic in a pilot study. nervous system (ANS) [4]. HRV has been considered as an objective and noninvasive health indicator, with higher resting Methods HRV being associated with lower cardiovascular risk [5], reduction of negative emotions through acute stress [6], flexible This study has two phases: (1) the development of a serious and adaptive response to environmental demands [7], and game biofeedback app called the FitLab Game and (2) the improved fitness [8]. This reflects the numerous applications testing of the app with participants in a pilot study. of HRV in various fields. Phase 1: App Development HRV biofeedback (HRVB) is an established intervention that increases HRV in the clinical context and has proven to be Visual Environment and Game Explanation effective as a treatment for anxiety and depression [9,10]. With General Design this technique, participants become aware of their HRV through In accordance with definition of serious games by Djaouti et al real-time feedback and can self-regulate it, restoring its [25], the FitLab Game is meant for physical and mental training homeostasis [11]. HRVB is based on the physiological in the health care sector and intended for the general public. phenomenon of respiratory sinus arrhythmia, which is defined The overall goal of the game is to help the user self-regulate as the variation in heart rate (HR) caused by respiration [12]. their HRV by providing biofeedback in an attempt to follow or The analysis of respiratory sinus arrhythmia shows that the mimic a preset iHR waveform. The architecture and design of instantaneous HR (iHR) increases during inhalation and the game allows the incorporation of new real iHR curves (RR) decreases during exhalation. Thus, the original HRVB procedure to imitate and can be adapted to different levels of complexity proposed by its promoters emphasizes the importance of in the different existing scenarios (levels). These real recordings breathing training to maximize HRV and achieve the maximum can be chosen by the player or a professional accompanying benefits of this technique [12]. In a recent systematic review the player. Depending on the level and the preset iHR waveform [13], a series of methodological guidelines were proposed to chosen, the game will lead to a state of relaxation (increasing improve the effectiveness of HRVB interventions. This review cardiac variability) or a state of activation in the participant highlighted the importance of the breathing protocol as the core (decreasing cardiac variability), according to the needs of the HRVB interventions, concluding that concrete aspects demanded by each particular situation. Different gamification such as the inhalation, holding, or exhalation ratio should be techniques were used in its design to increase the players’ explained in detail to the participants who apply the technique. engagement with the tasks. More general techniques include Furthermore, HRVB can help establish a connection between having a goal to aim toward, a set of rules to follow, a challenge the mind and body through voluntarily controlled physiological to complete, and real-time feedback to tell the player how he behaviors, such as breathing, facilitating the learning process is doing and how close he is to completing the goal. More and self-control. specific gamification techniques were added to further engage HRVB is becoming more widely used owing to its ease of the player, including Bonus points, increasing challenge, and application and high cost-benefit potential, particularly because performance graphs. of the current accessibility of health technologies through mobile Menus health devices [14,15]. The growth of mobile health technology has allowed the development of game-based mobile apps, which The game’s menus were designed in landscape mode and are known to increase the transfer of learning and motivation presented in Spanish. The first screen that appears when opening toward practice [16]. The use of games either in health or the app is the main screen, which allows users to choose the educational interventions allows the participants to immerse sensor used to input data: either the device’s camera (“Cámara themselves in the activity that is required by the game, de vídeo”) or a HR band (“Banda Pectoral”). The latter was promoting a flow state [17,18]. Similarly, other studies have used in this study. Once connected to the sensor, the app allows shown that when playing serious games, the student’s role access to the game’s mode, where the player can select between changes from a passive receiver of information to an active the option “Jugar” (“Play”), in which the user overcomes entity interacting with the game, facilitating the learning process different levels in an orderly manner so that the difficulty of [19,20]. Serious games, rather than being geared toward the game gradually increases, or the option “Práctica” entertainment, are aimed at educating, training, and informing (“Practice”), in which the user can activate an independent the player about a specific topic [21]. They are being used in session. The tab “Opciones” (“Options”) refers to the game’s various health domains [22-24]. For instance, in the field of settings (Figure 1). https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 2 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al Figure 1. Screenshots of the game’s menus. Game’s transition from the main screen to the screen that appears when you select “Banda pectoral” (“Chest Band”) and which allows to connect a heart rate sensor to the game. On the right, the third menu shows where the player can select “Jugar” (“Play”) to play. As a result of the initial calibration, all users start the game Screens under similar conditions, while also having the level of difficulty The FitLab Game has 9 levels distributed in 3 visually different of each scenario of the game adjusted to their own HR values. worlds (Figure 2). Each world has 3 scenarios or levels of Overall, the configuration of each game is a dynamic and difficulty that the user must overcome in a set order to be able individual process, setting it apart from “one-size-fits-all” to access the next one when in “Jugar” (“Play”) mode. To games. The individually tailored nature of the FitLab Game was overcome a scenario, the user must move the main character up designed to increase motivation and the learning process [26]. and down the screen to avoid colliding with obstacles that keep appearing. Touching an obstacle results in the loss of a life, and The left side of Figure 3 illustrates the algorithm that defines losing all lives means the game is over and needs to be restarted. the ideal pathway of the character. The blue line represents the The movement of the main character reflects cardiac variability waveform defined by the prerecorded iHR curve (from a specific and can be controlled by different voluntary strategies such as RR series), and the dark gray boxes represent the upper and breathing. lower limits of this curve, where the obstacles would be positioned. The right side of Figure 3 shows an example of how The upper and lower limits and the obstacles in each scenario this algorithm would be seen in the game (see a dynamic of the game are not placed randomly. They follow a specific example in Multimedia Appendix 1). Once the obstacles are waveform determined by previously recorded RR intervals, placed, the objects and landscape of the scene are selected. For corresponding to a real recording. Each scenario represents a each world, there is a bank of images that can be used real RR series transformed and visualized as iHR, with different (mountains, animals, houses, etc) and placed by distributing difficulty to imitate depending on the particular and distinct RR them according to their size. Larger objects form the background series (represented as a curve of iHR values) on which each plane, such as the mountains. Medium objects, such as houses, scenario is based. Before starting the game, the data were gradually fill the spaces that the first type of images could not calibrated to establish an initial HR range for the user (based fill because of their size, and small objects, such as birds, trees, on iHR from RR intervals). An algorithm was created or other details, are placed in the spaces that are missing to be specifically for the game, which takes this initial HR and filled in, which in Figure 3 are the dark gray boxes. Notably, displays it on a “grid” (invisible to the user) where each box is the placement of obstacles is not exhaustive, meaning that it is 1 second wide per 1 second high. With this algorithm, the screen not intended to fill all the space reserved for obstacles; is configured to place the character in the central area vertically, otherwise, the screen will be saturated with objects. The goal and the width of the screen is adjusted such that it is proportional is to place objects in the most harmonious way possible but to the range of variability presented by the user. In addition, the clearly marking the path that the main character should follow. upper and lower limits and the obstacles, based on the previously When the user reaches the end of the level or loses a life owing recorded RR series, are placed above and below the defined to touching obstacles, a final page appears. On this screen waveform, leaving a space for the character to pass in between (Figure 4), a “Level Completed” message is shown, together (Figure 3). with the total points collected, and the average stress index and average HR of the player and of all participants who have played the game is also shown. https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 3 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al Figure 2. Screenshots of worlds and scenarios of the game: “Campo” (“Rural area”), “Mar” (“Sea”), and “Espacio” (“Space”). Figure 3. Example of wave based on an instantaneous heart rate curve from a particular RR interval series. Figure 4. Example of the final page, appearing after a word of the FitLab Game is played. https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 4 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al App Architecture the user can move the main character up and down the screen using the variation of each iHR beat with respect to the previous Programing Language and Functionalities one. This app container was implemented using a cross-platform The architecture of the app consists of two parts: (1) the base environment for writing applications for Android and iOS of the game and (2) the container of the app and the operating systems called Titanium SDK (TiDev, Inc). communication with another device. The base of the game refers to the visual part, the game’s engine, screen management, The game engine is written in JavaScript and is based on a menus, and navigation. This entire part has been implemented constant loop where, at each step, the current position of the with HTML (WHATWG) assisted by CSS (World Wide Web character is evaluated with respect to its environment (screen Consortium) and JavaScript (Oracle Corporation) technologies, boundaries, obstacles, and prizes) and the input of the user (iHR so that it can be run in any web browser, whether hosting the data; Figure 5). The game engine, menu screens, and navigation application on a web server or running it directly on a local are written in web languages compatible with any browser. computer or mobile device. Note that opening the game in a Finally, both operating systems (Android and iOS) allow the web browser will not allow the use of the sensors and hardware; introduction of web components within the app to reproduce therefore, on a web browser, the game can only run in the test web code (ie, HTML+CSS+JavaScript) as if it were part of the mode. Nonetheless, in the test mode, it is still fully functional same app. The current app takes advantage of this feature to using the cursor on a computer keyboard or by tapping on the separate the game from the app’s container, which allows screen of a mobile device, which allows testers to move the communication via Bluetooth with other devices. When the character up and down the screen manually. container receives a command via Bluetooth, the app container code sends the corresponding interpretation of this command The container refers to what allows the game to be installed as to the game part. For example, when instantaneous HR data are a mobile app on an Android or iOS operating system. In received over Bluetooth, the app container unpacks the received addition, it provides the necessary functionality to access the frame, obtains specific iHR data, and sends it to the game, which camera of the device and process the images as well as to interprets it and moves the character. Figure 5 shows a game establish Bluetooth connections with HR sensors and process flow representation, where the user wearing a chest band sends the collected data. In this case, data processing is used in the iHR signals via Bluetooth and the game algorithm evaluates game as input commands, replacing the usual input methods, iHR, time playing, Bonus achieved, and position on a constant including the keyboard or touch screen (for testing purposes). loop. That is, it is with iHR data (from real-time RR intervals) that Figure 5. Game flow representation. iHR: instantaneous heart rate. standard attribute profiles (Generic Attribute Profile). This Data Collection and Bluetooth Connection allows, for example, Bluetooth communication in the Heart Before communication occurs between the sensor (HR sensor) Rate Profile mode to be established between the data sensor and the FitLab Game, a connection must be established between (the sender) and the app (the receiver). In this way, the receiver the entities. Such a connection is established via Bluetooth and can interpret the data in whichever format it is packaged and is known as pairing. In this type of connection, one of the can obtain the iHR information regardless of the manufacturer devices acts as a host or receiver of data and the other acts as a and model of the sending device, as long as it uses Bluetooth client or sender. The device acting as a receiver can establish low energy with the Heart Rate Profile mode. connections with multiple sending devices, but a sending device can only connect to a single receiver. Currently, the present app At present, when the app is opened, the user is asked to select allows only 1 sender to be connected to the device because each the data entry method. When Bluetooth is selected, a screen game is calibrated to a user’s iHR. with detected nearby Bluetooth devices is displayed. The list of displayed devices is filtered to show only those in the Heart It should be noted that the iHR reading device used for the Rate Profile mode. The user can select a device from the list of FitLab Game could be any, but it must offer a Bluetooth low available devices and establish a connection. This process may energy connection so that the app can obtain the data. This take 1 or 2 seconds, and the user receives a confirmation if the version of Bluetooth can define specific data formats using https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 5 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al connection has been successfully established or if an error has application manager. To export data, the manager needs to select occurred. Once the connection is established, the application the desired sessions and indicate an email address to which the takes the user to the game’s playing menu. anonymized data will be sent, that is, identified by user codes and never by names or other identifying data. Data Processing For each game carried out in the app, iHR data are collected Phase 2: Pilot Study together with other metadata, including the start and end times Study Sample of the game, a user’s identifier code, an ID assigned to the session, the level of the game played, the Bonus achieved, and A total of 16 participants (9 men and 7 women; mean age 23, the iHR calibration. All these data are packaged, saved in the 0.69 years) were divided into 2 groups: a control group (4 men container of the app, and can later be exported for analysis. and 4 women; mean age 23.25, SD 0.99 years) and an Multiple user sessions can be stored locally, that is, in the experimental group (5 men and 3 women; mean age 22.75, SD application, but an internet connection is required to export the 1.01 years). All participants were aged >18 years and did not data to a server. Note that the game does not need an internet present any cardiovascular disorder or take any medication that connection to work, and a wireless Bluetooth connection could affect HRV. The descriptive statistics of the participants between the sender and the receiver in proximity is sufficient. are presented in Table 1. All participants, selected via a For research purposes, the sessions that are temporarily saved convenience sample of university students, were volunteers, on the device are accessible from a configuration menu by an and all participants provided written informed consent. Table 1. Descriptive data of the participants (N=16). Variable Control group (n=8) Intervention group (n=8) All participants (N=16) P value Gender (woman), n (%) 4 (50) 3 (38) 7 (44) .61 Age (years), mean (SD) 23.25 (2.82) 22.75 (2.87) 23 (2.76) .73 Height (cm), mean (SD) 170 (10.53) 173.88 (8.34) 171.94 (9.39) .43 Weight (kg), mean (SD) 66.5 (13.86) 66 (9.41) 66.25 (11.45) .93 BMI (kg/m2), mean (SD) 22.84 (3.38) 21.78 (2.50) 22.31 (2.92) .49 information about the following factors (see item number in Material and Instruments brackets): (1) usefulness, (2) ease, (3) enjoyment of the game, Control Measures Self-Report (4) personal ability, (5) intention to use, (6) clarity of the goal Before attending the session, participants were recommended, of the game, (7) controllability of the game, (8) strategic by email, to avoid taking nonessential drugs (up to 24 hours approach, (9, 10, and 11) flow, (12) feedback, (13) educational before the session) as well as caffeine, smoking, or any other aspect, (14 and 15) motivation, and (16) design aspects. Each psychostimulant (up to 2 hours before); alcohol (10 hours factor was represented by a single item, except for flow and before); heavy meals (3 hours before); or eating in general (1 motivation, which consisted of 3 and 2, respectively. The hour before). They were also asked to avoid high-intensity participants responded to the questionnaire using a Likert scale physical activity or any unusual exercise (20 hours before), to ranging from 1 (strongly disagree) to 5 (strongly agree). Mean sleep for at least 6 hours, and to wear comfortable clothes to scores >3 can be considered as positive, 3 as neutral, and <3 as the laboratory. A brief questionnaire was administered negative. Items related to the techniques used to modulate HR immediately before the laboratory sessions to record these were not included in the control group. The average total score conditions. The participants were also asked to report their was computed based on participants’ responses, providing a weight and height. In addition, the BMI was computed based general satisfaction score. on the participants’ reported height and weight. This measure Procedure was used as a control variable between the groups because of The study consisted of 1 individual session per participant, its possible influence on HR [27]. which lasted for approximately 35 minutes and took place in Polar Band H10 the Basic Psychology Laboratory of the Universitat Autònoma An H10 cardiac chest band (Polar Electro) was used to record de Barcelona. The day before this session, each participant was the RR interval signal (transformed to iHR), as validated provided a list of recommendations, based on the Control previously [28]. The signal was sent and collected to the FitLab Measures Self-report, to record variables that may affect HRV, Game App (iOS version 1.0, build 1; Health&SportLab SL), including physical activity and intake of caffeine, which was active on an iPad Air 2 (version 12.1.4). psychostimulants, alcohol, drugs, and food. At the beginning of the session, the H10 cardiac band was placed on the Questionnaire of Satisfaction With FitLab Game participant’s chest so that they could become familiarized with An ad hoc satisfaction questionnaire based on the proposals of the sensation of the band. A few drops of conductive gel were different works [29-31] was tailored to our research purpose. applied to the band to facilitate good contact with the skin. Once The questionnaire was created to collect participants’ the band was placed, the participant answered the Control https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 6 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al Measures Self-report questionnaire. The participants were then with the functioning of the app, without time constraints. In part introduced to the FitLab Game for the first time. Play time was 2, the practice phase, participants practiced on a different level divided into 3 parts, with the first and third parts being recorded for 6 minutes, or a maximum of 3 levels. Finally, in part 3, the for the study (Figure 6). In part 1, the baseline phase, each test phase, each participant played the same 2 levels as in part participant played 2 levels of the game to familiarize themselves 1. Figure 6. Example of screens of the game that participants played. Testing consisted of 3 parts (part 1: baseline phase, part 2: practice phase, and part 3: test phase). A few techniques to consciously control iHR (and thus manage calculated as the average of the 2 games played in part 3 (Figure HRV) were explained to the participants in the experimental 6). This statistical analysis was carried out to analyze the group but not to those in the control group. To increase HRV baseline versus test data obtained from HRV parameters, as (and reduce iHR), it was recommended to either breathe at a well as time and Bonus, all separately. Statistical analysis was pace where the exhalation was longer than the inhalation [32-34] performed using SPSS (version 28.0; IBM Corp). or to hold one’s breath [35]. To reduce HRV (and increase iHR), The raw RR interval values recorded by the Polar H10 cardiac participants were instructed to slightly hyperventilate [36], to chest band were processed to remove artifacts owing to false perform muscle movements [37], or to swallow saliva positive or false negative detections; afterward, the RR intervals voluntarily all at once [38]. were filtered, and the corresponding iHR values were shown in For each level and session that the participants played, the real time on the screen when the participant was playing. The FitLab Game calibrated a baseline iHR, which allowed the same processed RR interval values were saved and analyzed to game’s character to be placed in the center of the screen, in such calculate the HRV parameters. A time domain HRV analysis a way that the starting point and the difficulty of the iHR series was performed to calculate average RR intervals (RR mean), were equal for all participants. Then, participants had to follow SD of RR intervals, root mean square of differences between the same pre-established route corresponding to each level and adjacent RR intervals (RMSSD), and percentage of consecutive session, collecting apples (Bonus), which marked the correct RR intervals that differ by >50 ms between them. A frequency route, and avoiding other elements such as unhealthy foods, domain analysis was used to calculate the low-frequency band tobacco, or the upper and lower limits marked by the floor and (low frequency [LF]; 0.04-015 Hz) and high-frequency band ceiling of the screen. To follow the route (or avoid certain (high frequency [HF]; 0.15-0.4 Hz). HRV analysis followed elements), the main character had to move up or down the the guidelines recommended by the Task Force of the European screen, which was controlled by changes in iHR. Participants Society of Cardiology and the North American Society of Pacing in the control group were given no instructions on how to do and Electrophysiology (1996). HRV analysis was performed so, whereas participants in the experimental group were using MATLAB (R2021 update 3 for 64 bits Windows; explained the techniques mentioned in the paragraph above. All MathWorks). participants started the game with 20 lives, and the game ended when they ran out of lives or when they reached the end of the Ethics Approval level without losing all these lives. This study was conducted according to the local ethics committee for human experimentation (protocol code Data Analysis CEEAH-5745). Descriptive statistics are reported as means and SDs. A 2 × 2 multivariate analysis of variance following a general linear model was used to analyze the differences between the baseline and test situations. The baseline values were calculated as the average of the 2 games played in part 1, and the test values were https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 7 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al time played in seconds (Time) were considered, as presented Results in Table 2. No significant changes were found in the baseline Change in Gaming Performance (Pilot Study) versus test situation when comparing the experimental and control groups. However, there was a tendency for both groups To analyze the participants’ performance in the game, the to have increased Bonus and Time from the baseline to the test change in the number of apples collected (Bonus) and the total condition. Table 2. Game outcomes in baseline and test phases in the control and experimental groups. Variables Baseline, mean (SD) Test, mean (SD) Bonus Control 16.75 (10.91) 18.06 (9.54) Experimental 15.75 (5.34) 20 (10.15) Time Control 91.08 (26.79) 109.05 (58.17) Experimental 91.07 (26.79) 116.67 (52.31) group and a total score of 4.30 (SD 0.50) for the intervention Participants’ Satisfaction group. The intervention group answered 16 questions, with 4 The satisfaction questionnaire was completed by 13 (81%) of specific questions related to the techniques used during the the 16 participants. Overall, participants expressed high game. Table 3 provides an extended description of the items satisfaction with the FitLab Game, reporting positive and their scores for each group. experiences, with a total score of 3.87 (SD 0.42) for the control Table 3. Items and scores of the ad hoc game’s satisfaction questionnaire for the control group (n=5) and intervention group (n=8)a. Items Control group, mean (SD) Intervention group, mean (SD) 1 “I can learn self-control techniques better and faster with games like this.” N/Ab 4 (0.53) 2 “Using games like this would be easy for me.” 4.2 (0.45) 4 (1.07) 3 “Playing to learn techniques is more fun than following a pacer.” 4.6 (0.55) 4.87 (0.35) 4 “I would be able to play this game even if nobody was there to teach me.” N/A 3 (0.76) 5 “In the future, I would play if I wanted to learn how to control my heart rate.” 4.2 (0.84) 4 (0.76) 6 “The aim of the game was clear the whole time I was playing.” 4 (1.22) 5 (0) 7 “The instructions for functionality were clear.” N/A 5 (0) 8 “While playing, I understood how to achieve points and how to end the game 4.2 (1.30) 4.6 (0.52) successfully.” 9 “While playing, I was only thinking about the game.” 3.6 (1.14) 4.75 (0.46) 10 “While playing, I forgot everything else around me.” 4 (1) 4.5 (0.53) 11 “While playing, I didn’t realise how much time was passing.” 3.2 (0.84) 4.13 (0.83) 12 “When I did something in the game, I knew if it was right or wrong.” 3.4 (1.82) 4.25 (1.04) 13 “By playing I learned how the techniques affect my heart rate, and how I can N/A 4.13 (0.64) modify it.” 14 “The game format increased my motivation to continue playing and achieve more 3.4 (0.89) 4.63 (0.52) bonus points.” 15 “A game that applied different breathing techniques would increase my interest 4 (0.71) 4.63 (0.52) in this topic.” 16 “The levels are fun and esthetically pleasing.” 4.2 (0.45) 4.38 (0.74) a Total score: control group—mean 3.87 (SD 0.42); intervention group—mean 4.30 (SD 0.50). b N/A: not applicable. https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 8 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al Change in HRV Parameters which was different from the increase observed in the control When analyzing the changes in HRV before and after the game, group (P=.02; Figure 7). In contrast, the experimental group the results showed some differences between the experimental showed a decline in RMSSD from baseline to test, which was and control groups (Table 4). In particular, the experimental more pronounced than that in the control group (P=.04). No group showed a decrease in RR mean from baseline to test, other HRV parameters presented a significant or tendential change. Table 4. Multivariate analysis of variance of heart rate variability variables in baseline and test conditions in the control and experimental groups. Variables Baseline, mean (SD) Test, mean (SD) P value RR meana .02 Control 800.33 (109.02) 816.89 (106.59) Experimental 1004.52 (148.41) 930.90 (132.45) SDRRb .56 Control 62.55 (24.12) 65.64 (20.16) Experimental 111.35 (33.97) 109.81 (32.44) RMSSDc .04 Control 43.10 (19.96) 41.60 (20.16) Experimental 93.80 (40.71) 80.07 (31.73) pNN50d .36 Control 23.83 (17.53) 20.77 (17.25) Experimental 42.90 (18.63) 34.64 (17.52) LFe .39 Control 3063.58 (2789.61) 2595.49 (2190.80) Experimental 6612.13 (3438.65) 6972.52 (3760.22) f .10 HF Control 661.70 (458.82) 739.14 (547.80) Experimental 2985.17 (2311.75) 2046.96 (1594.17) a RR mean: average RR intervals. b SDRR: SD of RR intervals. c RMSSD: root mean square of differences between adjacent RR intervals. d pNN50: percentage of consecutive RR intervals that differ by >50 ms between them. e LF: low frequency. f HF: high frequency. https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 9 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al Figure 7. Mean, 95% CI, and distribution of average RR intervals (RR mean) and root mean square of differences between adjacent RR intervals (RMSSD), when comparing baseline and test situations in the experimental and control groups. the baseline), compared with the control group (increase of 1.31 Discussion Bonus points in the test; Table 2). The same trend was observed Overview in the total playing time, where the experimental group kept playing 7.6 seconds longer than the control group, but these The aim of this study was 2-fold: to develop a serious game results were not statistically significant (Table 2). All these that teaches the user to self-regulate iHR using HRVB and to results are consistent with the idea that the application of specific carry out a pilot study to test the game with participants. A total strategies in the experimental group could have improved the of 16 participants took part in the pilot study and were divided self-regulation of heart variability, more than simply practicing into a control group and an experimental group. The results the game (as shown with the control group). However, these were analyzed in 2 parts: one focused on changes in the results should be interpreted with caution because of the small performance of the game and the other on fluctuations on HRV sample size and the high values of SDs shown in Table 2. parameters. The results related to the usefulness of the game are discussed in the Game and App Performance section. In our study, a few gamification techniques were introduced to increase players’ desire to engage with the tasks. General Game and App Performance techniques include having a goal to aim toward, a set of rules Analyzed globally, the gaming performance of all participants to follow, a challenge to complete, and real-time feedback to was in line with the results of another gaming study [39], in tell the player how he is doing and how close he is to completing which the participants played a biofeedback game 6 times and the goal. More specific gamification techniques included Bonus both the control group that received simulated feedback and the points, increasing challenges, and performance graphs. Reynard experimental group that received physiological feedback et al [40], in a systematic review and meta-analysis of digital improved their gaming performance. These results indicate that interventions for emotion regulation, found that digital game playing a given game, in itself, leads to improvement in its final interventions consistently provided stronger evidence compared score through repeated training or habituation. That is, both with other types of digital interventions. In this review, 64% of groups got to experiment and learn how to move the game’s digital game studies provided feasibility for engagement, character. Regarding the differences in performance between implementation, adherence, expectations, and transference to groups, the results showed that the experimental group had a real life. However, they did not explain which elements of the greater improvement in all indicators compared with the control specific game design were responsible for this feasibility. For group, even though the differences were not significant. The their part, Maher et al [41] concluded that gamified apps were experimental group showed a greater improvement in Bonus used substantially longer than nongamified apps. Other studies score (increase of 4.25 Bonus points in the test with respect to highlight specific game elements that impact health behavior https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 10 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al and achieve adherence, similar to digital rewards, such as points, compared with the control group (score >3). Regarding the badges, or rewards that we used in our game [42]. Similarly, in experience of flow during the game, although both groups scored the design of our game, we complied with five gamification >3, the intervention group scored >4 across the 3 items. principles relating to the motivation of the user [43,44]: (1) Similarly, in terms of motivation, the intervention group scored meaningful purpose, as the goal of the users participating in the >4 and the control group scored >3. Finally, both groups found study was to learn a technique to improve their health; (2) the screens visually appealing and balanced in terms of colors, meaningful choice, as users could control how they achieved with scores >4. their goal of eating the apples on the screen; (3) supporting As shown in Table 3, the techniques provided to the intervention player archetypes, as in our game, the initial iHR level is group were found to be useful in establishing control over iHR individualized based on the user characteristics; (4) feedback, and were considered important for playing successfully. Thus, as the user’s actions to control the movement on the screen the neutral score in the personal ability factor indicates that the based on their own cardiac variability affect progress, and it is instructions based on different techniques for controlling iHR clearly communicated; and (5) visibility, as the amount of should be considered necessary for success in the game. The progress made in each level (apples, points, and Bonuses gameplay was understood by both the groups, and the game achieved) is visible at all times. Most of these principles are was generally perceived as easy and enjoyable. With regard to reaffirmed by the users’ responses to the subsequent flow and motivation, the intervention group reported a more questionnaire, as discussed in Participants’Satisfaction section. favorable experience with the game. This can be explained by Thus, the different techniques applied in the present serious the use of techniques that enable participants to exert voluntary game could contribute to behavioral changes, such as increasing control over the characters, thereby enhancing the game players’ engagement with the tasks and adhering to the use of experience through the interactive component of the game. the game. For example, the instant feedback provided by the These findings are in line with those of previous studies, game may have helped produce high engagement and indicating that a gamified visualization component results in enjoyability for the participants, as noted by Siriaraya et al [45]. higher intrinsic motivation [48]. It is worth noting that the At the same time, the fact that the levels of the game are adapted participants expressed a strong interest in continuing to play to each participant’s cardiac recording means that each person the FitLab Game in the future, reflecting their engagement and plays in a personalized environment, which is thought to willingness to play to learn about cardiac variability control. increase motivation toward the game [46]. Finally, regarding the app’s functionality, we highlight the possibility of using On the basis of the results presented, we propose an extensive either Bluetooth devices or photoplethysmography sensors [47], use of the game with the main goal of helping individuals learn allowing participants to play in different contexts and with how to self-regulate their cardiac variability to achieve different resources. emotional self-control. This could be applied in different contexts, including clinical (eg, in anxiety or stress disorders) The serious game created for this study can be used in multiple and nonclinical populations. This is in line with the results of situations (in the long term). The game provides flexibility to a systematic review where consistent evidence was found for play in new scenarios, depending on the specific training or reduced negative emotional experience provided by digital situation goals set by the player. Furthermore, there is a menu games [40]. Ideally, these games should be used under the for practice in which the players can play to test different supervision of health professionals to facilitate learning-specific techniques. Therefore, this game supports repetitive practice as techniques for controlling cardiac variability. This is reaffirmed an effective training method. In addition, the game uses the because the only response <4 in the specific items for the Bonus and time measurements to self-track the learning intervention group is for the item: “4 I would be able to play progression. As the players enhance their level of expertise, the this game even if nobody was there to teach me” (rating of 3 Bonus points and time increase accordingly. This mechanism out of 5). Nonetheless, as shown in Table 3, participants reported ensures that as the player improves, they are rewarded with that the game was easy to understand and play (item 2), higher Bonuses and more time in the level, motivating them. indicating that both the control and experimental groups Participants’ Satisfaction understood the game well. Participants expressed an overall positive satisfaction with the HRV Analysis and Biofeedback FitLab Game, with a greater total score in the intervention group Regarding HRV, the results of the pilot study showed some (4.30 out of 5) than in the control group (3.87 out of 5). As differences in the cardiac behavior of the experimental and Table 3 shows, this higher satisfaction in the intervention group control groups (Table 4; Figure 7). This suggests that the is also supported by the fact that all the items that both groups strategies used by the experimental group to compete in the had in common showed a higher score of 4 (out of 5) only for game improved their control of cardiac variability, which in the intervention group. Both groups considered the game to be turn may have enhanced their performance. Owing to its pilot easy and fun (punctuations >4 for both groups). The goal and nature and limited number of participants, the precise reasons strategic approach to achieve Bonuses and continue playing underlying the relationship between control of cardiac variability were clear for both groups, with scores >4. Furthermore, both and performance cannot be conclusively determined. groups expressed positive intention (score >4) to continue Nonetheless, it is possible that the results reflect the activation playing the FitLab Game and learn how to control their HR in and the control of attention mechanisms in the experimental the future. The participants in the intervention group rated the group. If we consider that the game requires a certain level of feedback provided by the game more positively (score >4) https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 11 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al concentration and attention to play, especially if the participants articles used set breathing paces, usually between 4.5 and 6.5 are remembering and applying the techniques that were just times per minute, as HRVB to induce relaxation [60,61]. For learnt, some arousal level is needed, reflected in a reduction in example, on the topic of gaming, Al Osman et al [62] used the activity of the parasympathetic nervous system, as shown game-based biofeedback to control the activity of the ANS and by a reduction, although not significant, in the experimental showed that HRV increased following a controlled breathing group (HF parameter) compared with a slight increase in the rate. In contrast, the goal of this study was not to increase, control group. This result is consistent with the increase, but maintain, or decrease HRV but to teach participants to control not significant, in the LF values in the experimental group its rises and falls through various techniques, including breathing compared with the decrease in the control group. There is control, but not at a set pace. Thus, our study is not comparable evidence that higher LF values are related to a greater activation with most other studies, given that we did not focus on breathing of the sympathetic nervous system [49]. At the same time, a paces but rather on an array of techniques to teach the significant reduction in cardiac variability was observed in the participants how to increase or decrease their HRV. These experimental group, as shown by the RR mean and RMSSD techniques include the voluntary production of apneas to parameters. The set of previous results could explain the suddenly increase cardiac variability [35], the performance of improvement in the control of arousal or activation in the muscle movements to achieve a reduction in HRV (in line with experimental group. In line with this idea, Fuentes-García et al Guidi et al [37]), or voluntary swallowing. This last strategy [50] found that in chess games, the player’s level and the can produce changes in some HRV parameters in healthy people, difficulty of the game determined the activity of the ANS of such as SD of RR intervals, LF, and HF power [63], and effortful the individuals, with higher demands in attentional focus swallowing increases LF power and the LF/HF ratio [64]. On resulting in lower HRV. Thus, in this study, the experimental a practical level, in our case, these strategies allowed the player group may have shown reduced vagal activity and cardiac to move the character up and down the game screen because variability because they were given techniques to control their they provide rapid changes in cardiac variability. HRV, requiring greater attention when playing [51]. In that sense, HRV is an indicator that the experimental group was Limitations and Future Directions active and applying what had been taught. This study had 3 main limitations. The first limitation is the sample size of the pilot study. Even if the results show An array of scientific papers highlights the potential of significance in the mean iHR (proportional to the RR mean) incorporating biofeedback into different types of digital game and RMSSD, they should still be considered preliminary. The interventions, including virtual and augmented reality, internet second limitation is that participants were not monitored before therapy, biofeedback and neurofeedback, digital games, and and after the game but only during it. This means that the HRV web-based programs. In a systematic review by Reynard et al indicators were collected as the game was being played, thus [40], 39 interventions were identified, 10% of which were based in an active state and not in a resting situation. Our interest was on biofeedback. The effectiveness and impact of incorporating to test whether there were changes in the HRV parameters and biofeedback into digital games have been tested in a range of game performance at all, as a pilot study, but in future studies, topics, from stress reduction to physical rehabilitation [52]. For it would be interesting to observe the influence of our instance, Knox et al [53] and Wenck et al [54] studied intervention on basal HRV. biofeedback-driven digital game interventions for stress levels before the emergence of smartphones and concluded that Finally, most previous studies using techniques such as HRVB symptoms of anxiety and depression were significantly included a sample of clinical patients, unlike this study, which alleviated in children. Similarly, Jerčić and Sundstedt [55] was conducted in healthy university students. For instance, showed that visual and gameplay biofeedback in serious games Siepman et al [65] compared the effects of HRVB in both supported emotional regulation skills. Moreover, aside from healthy people and individuals classified as depressed, showing the impact on the psychophysiology of individuals, digital that the latter experienced a significantly greater improvement interventions appear to be accessible and attractive compared in HRV, compared with healthy volunteers. Similarly, Pyne et with one-to-one clinical therapies [40], particularly for the al [66] reported that HRVB training reduces symptoms of younger population [56]. Lüddecke and Felnhofer [57] posttraumatic stress in older people only. These studies highlight demonstrated that biofeedback interventions increased the that HRV parameters seem to be difficult to alter with HRVB motivation and long-term participation of users in health-related in healthy and younger people than in older or clinical interventions. Finally, games that incorporate biofeedback, populations, potentially explaining the results of this study, with compared with other digital interventions, seem to provide the few to no significant changes in some HRV parameters. In the most evidence for emotional skills transference to real life [40]. future, a larger sample size of healthy university students or In conclusion, there is a growing body of research that testing the FitLab Game in a clinical population could provide emphasizes the utility of incorporating biofeedback into digital more consistent results. Finally, it would also be interesting to game interventions for its attractiveness and potential to enhance provide new strategies for the voluntary control of HRV, perhaps emotional regulation, among other benefits. further refining the ones we have already developed in this study. It would also be necessary to deepen the swallowing Returning to our study, our results seem to contradict most strategy, as Yildiz and Doma [38] and Yıldız [67] concluded studies using the HRVB technique, which conclude that this that spontaneous saliva swallowing can change some short-term technique helps improve HRV [58,59] with the general aim of HRV parameters significantly even in healthy people. achieving relaxation. However, this was not the case. These https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 12 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al Conclusions improvement in the management of arousal or activation This study developed a serious game, called the FitLab Game, compared with the control group. Thus, the application of that teaches the user to self-regulate the iHR using cardiac specific strategies in the experimental group may have improved variability biofeedback. The game was also tested with the the self-regulation of HRV more than simply practicing the participants in a pilot study. Overall, the FitLab Game appears game. to be a promising serious game because of its ease of use, high The usefulness of HRV as a health indicator is well known, and engagement, and enjoyability provided by the instantaneous HRV biofeedback is considered an established intervention in feedback. The strategies learned by the experimental group to the clinical context. Thus, learning simple and voluntary compete in the game, which included the voluntary production strategies through the developed game can help improve of apneas, muscle movements, and voluntary swallowing, self-control and arousal management. Ultimately, with sufficient improved the self-control of cardiac variability and, in turn, training, the game can become a relaxation technique for the may have enhanced performance. A significant reduction in general population or clinical populations, or, for example, an cardiac variability observed in the experimental group, based activation technique for athletes in certain situations. on the RR mean and RMSSD parameters, could explain the Acknowledgments This work was supported by grants PID2019-107473RB-C21 and PID2019-107473RB-C22 funded by MCIN/AEI/10.13039/501100011033 by the Spanish Government and 2021SGR-00806 funded by the Catalan Government. Conflicts of Interest None declared. Multimedia Appendix 1 Gameplay of the FitLab Game. [MP4 File (MP4 Video), 20497 KB-Multimedia Appendix 1] References 1. Rodas G, Pedret C, Ramos J, Capdevila L. Heart rate variability: definition, measurement and clinical relation aspects (part II). Arch Medicina Deporte 2008 Mar 1;25(124):119-127 [FREE Full text] 2. Cygankiewicz I, Zareba W. Heart rate variability. Handb Clin Neurol 2013;117:379-393 [doi: 10.1016/B978-0-444-53491-0.00031-6] [Medline: 24095141] 3. Heiss S, Vaschillo B, Vaschillo EG, Timko CA, Hormes JM. Heart rate variability as a biobehavioral marker of diverse psychopathologies: a review and argument for an "ideal range". Neurosci Biobehav Rev 2021 Feb;121:144-155 [doi: 10.1016/j.neubiorev.2020.12.004] [Medline: 33309905] 4. Tiwari R, Kumar R, Malik S, Raj T, Kumar P. Analysis of heart rate variability and implication of different factors on heart rate variability. Curr Cardiol Rev 2021 Sep;17(5):e160721189770 [doi: 10.2174/1573403x16999201231203854] 5. Sharma M, Rajput JS, Tan RS, Acharya UR. Automated detection of hypertension using physiological signals: a review. Int J Environ Res Public Health 2021 May 29;18(11):5838 [FREE Full text] [doi: 10.3390/ijerph18115838] [Medline: 34072304] 6. Goessl VC, Curtiss JE, Hofmann SG. The effect of heart rate variability biofeedback training on stress and anxiety: a meta-analysis. Psychol Med 2017 Nov;47(15):2578-2586 [doi: 10.1017/S0033291717001003] [Medline: 28478782] 7. Forte G, Favieri F, Casagrande M. Heart rate variability and cognitive function: a systematic review. Front Neurosci 2019 Jul 9;13:710 [FREE Full text] [doi: 10.3389/fnins.2019.00710] [Medline: 31354419] 8. Souza HC, Philbois SV, Veiga AC, Aguilar BA. Heart rate variability and cardiovascular fitness: what we know so far. Vasc Health Risk Manag 2021 Nov;Volume 17:701-711 [doi: 10.2147/vhrm.s279322] 9. Lehrer P, Kaur K, Sharma A, Shah K, Huseby R, Bhavsar J, et al. Heart rate variability biofeedback improves emotional and physical health and performance: a systematic review and meta analysis. Appl Psychophysiol Biofeedback 2020 Sep 08;45(3):109-129 [doi: 10.1007/s10484-020-09466-z] [Medline: 32385728] 10. Pizzoli SF, Marzorati C, Gatti D, Monzani D, Mazzocco K, Pravettoni G. A meta-analysis on heart rate variability biofeedback and depressive symptoms. Sci Rep 2021 Mar 23;11(1):6650 [FREE Full text] [doi: 10.1038/s41598-021-86149-7] [Medline: 33758260] 11. Blase K, Vermetten E, Lehrer P, Gevirtz R. Neurophysiological approach by self-control of your stress-related autonomic nervous system with depression, stress and anxiety patients. Int J Environ Res Public Health 2021 Mar 24;18(7):3329 [FREE Full text] [doi: 10.3390/ijerph18073329] [Medline: 33804817] 12. Lehrer PM, Vaschillo E, Vaschillo B. Resonant frequency biofeedback training to increase cardiac variability: rationale and manual for training. Appl Psychophysiol Biofeedback 2000 Sep;25(3):177-191 [doi: 10.1023/a:1009554825745] [Medline: 10999236] https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 13 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al 13. Lalanza JF, Lorente S, Bullich R, García C, Losilla JM, Capdevila L. Methods for Heart Rate Variability Biofeedback (HRVB): a systematic review and guidelines. Appl Psychophysiol Biofeedback 2023 Mar 14 (forthcoming). [doi: 10.1007/s10484-023-09582-6] [Medline: 36917418] 14. Free C, Phillips G, Felix L, Galli L, Patel V, Edwards P. The effectiveness of M-health technologies for improving health and health services: a systematic review protocol. BMC Res Notes 2010 Oct 06;3:250 [FREE Full text] [doi: 10.1186/1756-0500-3-250] [Medline: 20925916] 15. Harrer M, Adam SH, Fleischmann RJ, Baumeister H, Auerbach R, Bruffaerts R, et al. Effectiveness of an internet- and app-based intervention for college students with elevated stress: randomized controlled trial. J Med Internet Res 2018 Apr 23;20(4):e136 [FREE Full text] [doi: 10.2196/jmir.9293] [Medline: 29685870] 16. Gummidela VN, Silva DR, Gutierrez-Osuna R. Evaluating the role of breathing guidance on game-based interventions for relaxation training. Front Digit Health 2021 Dec 9;3:760268 [FREE Full text] [doi: 10.3389/fdgth.2021.760268] [Medline: 34957462] 17. Perlwitz P, Stemmann J. Flow and self-efficacy in a serious game for STEM education. In: Söbke H, Spangenberger P, Müller P, Göbel S, editors. Serious Games. Cham, Switzerland: Springer; Sep 15, 2022. 18. Perttula A, Kiili K, Lindstedt A, Tuomi P. Flow experience in game based learning – a systematic literature review. Int J Serious Games 2017 Mar 22;4(1):57-72 [doi: 10.17083/ijsg.v4i1.151] 19. Lei H, Chiu MM, Wang D, Wang C, Xie T. Effects of game-based learning on students’ achievement in science: a meta-analysis. J Educ Comput Res 2022 Feb 08;60(6):1373-1398 [doi: 10.1177/07356331211064543] 20. Su CH, Cheng CH. A mobile gamification learning system for improving the learning motivation and achievements. J Comput Assist Learn 2014 Nov 06;31(3):268-286 [doi: 10.1111/jcal.12088] 21. Michael DR, Chen S. Serious Games: Games that Educate, Train and Inform. Boston, MA: Thomson Course Technology; 2006. 22. Fleming TM, Bavin L, Stasiak K, Hermansson-Webb E, Merry SN, Cheek C, et al. Serious games and gamification for mental health: current status and promising directions. Front Psychiatry 2017 Jan 10;7:215 [FREE Full text] [doi: 10.3389/fpsyt.2016.00215] [Medline: 28119636] 23. Mubin O, Alnajjar F, Jishtu N, Alsinglawi B, Al Mahmud A. Exoskeletons with virtual reality, augmented reality, and gamification for stroke patients' rehabilitation: systematic review. JMIR Rehabil Assist Technol 2019 Sep 08;6(2):e12010 [FREE Full text] [doi: 10.2196/12010] [Medline: 31586360] 24. Tong T, Chignell M, Tierney MC, Lee J. A serious game for clinical assessment of cognitive status: validation study. JMIR Serious Games 2016 May 27;4(1):e7 [FREE Full text] [doi: 10.2196/games.5006] [Medline: 27234145] 25. Djaouti D, Alvarez J, Jessel JP. Classifying serious games: the G/P/S model. In: Handbook of Research on Improving Learning and Motivation through Educational Games: Multidisciplinary Approaches. Hershey, PA: IGI Global; 2011. 26. Oliveira W, Hamari J, Shi L, Toda AM, Rodrigues L, Palomino PT, et al. Tailored gamification in education: a literature review and future agenda. Educ Inf Technol 2022 Jun 29;28(1):373-406 [doi: 10.1007/s10639-022-11122-4] 27. Quer G, Gouda P, Galarnyk M, Topol EJ, Steinhubl SR. Inter- and intraindividual variability in daily resting heart rate and its associations with age, sex, sleep, BMI, and time of year: retrospective, longitudinal cohort study of 92,457 adults. PLoS One 2020 Feb 5;15(2):e0227709 [FREE Full text] [doi: 10.1371/journal.pone.0227709] [Medline: 32023264] 28. Gilgen-Ammann R, Schweizer T, Wyss T. RR interval signal quality of a heart rate monitor and an ECG Holter at rest and during exercise. Eur J Appl Physiol 2019 Jul 19;119(7):1525-1532 [doi: 10.1007/s00421-019-04142-5] [Medline: 31004219] 29. Hawkins GE, Rae B, Nesbitt KV, Brown SD. Gamelike features might not improve data. Behav Res Methods 2013 Jun;45(2):301-318 [doi: 10.3758/s13428-012-0264-3] [Medline: 23055169] 30. Högberg J, Hamari J, Wästlund E. Gameful Experience Questionnaire (GAMEFULQUEST): an instrument for measuring the perceived gamefulness of system use. User Model User Adap Inter 2019 Feb 28;29(3):619-660 [doi: 10.1007/s11257-019-09223-w] 31. Iten N, Petko D. Learning with serious games: is fun playing the game a predictor of learning success? Br J Educ Technol 2014 Nov 11;47(1):151-163 [doi: 10.1111/bjet.12226] 32. Li C, Chang Q, Zhang J, Chai W. Effects of slow breathing rate on heart rate variability and arterial baroreflex sensitivity in essential hypertension. Medicine (Baltimore) 2018 May;97(18):e0639 [FREE Full text] [doi: 10.1097/MD.0000000000010639] [Medline: 29718876] 33. Van Diest I, Verstappen K, Aubert AE, Widjaja D, Vansteenwegen D, Vlemincx E. Inhalation/exhalation ratio modulates the effect of slow breathing on heart rate variability and relaxation. Appl Psychophysiol Biofeedback 2014 Dec;39(3-4):171-180 [doi: 10.1007/s10484-014-9253-x] [Medline: 25156003] 34. Zaccaro A, Piarulli A, Laurino M, Garbella E, Menicucci D, Neri B, et al. How breath-control can change your life: a systematic review on psycho-physiological correlates of slow breathing. Front Hum Neurosci 2018 Sep 7;12:353 [FREE Full text] [doi: 10.3389/fnhum.2018.00353] [Medline: 30245619] 35. Bouten J, Bourgois JG, Boone J. Hold your breath: peripheral and cerebral oxygenation during dry static apnea. Eur J Appl Physiol 2020 Oct 03;120(10):2213-2222 [doi: 10.1007/s00421-020-04445-y] [Medline: 32748010] https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 14 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al 36. Hawkins SM, Guensch DP, Friedrich MG, Vinco G, Nadeshalingham G, White M, et al. Hyperventilation-induced heart rate response as a potential marker for cardiovascular disease. Sci Rep 2019 Nov 29;9(1):17887 [FREE Full text] [doi: 10.1038/s41598-019-54375-9] [Medline: 31784617] 37. Guidi A, Greco A, Felici F, Leo A, Ricciardi E, Bianchi M, et al. Heart rate variability analysis during muscle fatigue due to prolonged isometric contraction. In: Proceedings of the 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). 2017 Presented at: 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC); July 11-15, 2017; Jeju, Korea (South) [doi: 10.1109/embc.2017.8037076] 38. Yildiz M, Doma S. Effect of spontaneous saliva swallowing on short-term heart rate variability (HRV) and reliability of HRV analysis. Clin Physiol Funct Imaging 2018 Jul 26;38(4):710-717 [doi: 10.1111/cpf.12475] [Medline: 28949087] 39. Raaijmakers SF, Steel FW, de Goede GM, van Wouwe NC, van Erp JB, Brouwer AM. Heart rate variability and skin conductance biofeedback: a triple-blind randomized controlled study. In: Proceedings of the Humaine Association Conference on Affective Computing and Intelligent Interaction. 2013 Presented at: Humaine Association Conference on Affective Computing and Intelligent Interaction; September 02-05, 2013; Geneva, Switzerland [doi: 10.1109/acii.2013.54] 40. Reynard S, Dias J, Mitic M, Schrank B, Woodcock KA. Digital interventions for emotion regulation in children and early adolescents: systematic review and meta-analysis. JMIR Serious Games 2022 Aug 19;10(3):e31456 [FREE Full text] [doi: 10.2196/31456] [Medline: 35984681] 41. Maher CA, Olds T, Vandelanotte C, Plotnikoff R, Edney SM, Ryan JC, et al. Gamification in a physical activity app: what gamification features are being used, by whom, and does it make a difference? Games Health J 2022 Jun 01;11(3):193-199 [doi: 10.1089/g4h.2021.0207] [Medline: 35501981] 42. Lister C, West JH, Cannon B, Sax T, Brodegard D. Just a fad? Gamification in health and fitness apps. JMIR Serious Games 2014 Aug 04;2(2):e9 [FREE Full text] [doi: 10.2196/games.3413] [Medline: 25654660] 43. Berglund A, Jaarsma T, Berglund E, Strömberg A, Klompstra L. Understanding and assessing gamification in digital healthcare interventions for patients with cardiovascular disease. Eur J Cardiovasc Nurs 2022 Aug 29;21(6):630-638 [FREE Full text] [doi: 10.1093/eurjcn/zvac048] [Medline: 35709297] 44. Floryan M, Chow PI, Schueller SM, Ritterband LM. The model of gamification principles for digital health interventions: evaluation of validity and potential utility. J Med Internet Res 2020 Jun 10;22(6):e16506 [FREE Full text] [doi: 10.2196/16506] [Medline: 32519965] 45. Siriaraya P, Visch V, Boffo M, Spijkerman R, Wiers R, Korrelboom K, et al. Game design in mental health care: case study-based framework for integrating game design into therapeutic content. JMIR Serious Games 2021 Dec 01;9(4):e27953 [FREE Full text] [doi: 10.2196/27953] [Medline: 34855611] 46. Sardi L, Idri A, Fernández-Alemán JL. A systematic review of gamification in e-Health. J Biomed Inform 2017 Dec;71:31-48 [FREE Full text] [doi: 10.1016/j.jbi.2017.05.011] [Medline: 28536062] 47. Plews DJ, Scott B, Altini M, Wood M, Kilding AE, Laursen PB. Comparison of heart-rate-variability recording with smartphone photoplethysmography, polar H7 chest strap, and electrocardiography. Int J Sports Physiol Perform 2017 Nov 01;12(10):1324-1328 [doi: 10.1123/ijspp.2016-0668] [Medline: 28290720] 48. Lukic YX, Klein SS, Brügger V, Keller OC, Fleisch E, Kowatsch T. The impact of a gameful breathing training visualization on intrinsic experiential value, perceived effectiveness, and engagement intentions: between-subject online experiment. JMIR Serious Games 2021 Sep 14;9(3):e22803 [FREE Full text] [doi: 10.2196/22803] [Medline: 34519662] 49. Borovac JA, D'Amario D, Bozic J, Glavas D. Sympathetic nervous system activation and heart failure: current state of evidence and the pathophysiology in the light of novel biomarkers. World J Cardiol 2020 Aug 26;12(8):373-408 [FREE Full text] [doi: 10.4330/wjc.v12.i8.373] [Medline: 32879702] 50. Fuentes-García JP, Villafaina S, Collado-Mateo D, de la Vega R, Olivares PR, Clemente-Suárez VJ. Differences between high vs. low performance chess players in heart rate variability during chess problems. Front Psychol 2019;10:409 [FREE Full text] [doi: 10.3389/fpsyg.2019.00409] [Medline: 30863351] 51. Vermeir JF, White MJ, Johnson D, Crombez G, Van Ryckeghem DM. The effects of gamification on computerized cognitive training: systematic review and meta-analysis. JMIR Serious Games 2020 Aug 10;8(3):e18644 [FREE Full text] [doi: 10.2196/18644] [Medline: 32773374] 52. Alarcón-Aldana AC, Callejas-Cuervo M, Bo AP. Upper limb physical rehabilitation using serious videogames and motion capture systems: a systematic review. Sensors (Basel) 2020 Oct 22;20(21):5989 [FREE Full text] [doi: 10.3390/s20215989] [Medline: 33105845] 53. Knox M, Lentini J, Cummings T, McGrady A, Whearty K, Sancrant L. Game-based biofeedback for paediatric anxiety and depression. Ment Health Fam Med 2011 Sep;8(3):195-203 [FREE Full text] [Medline: 22942901] 54. Wenck LS, Leu PW, D'Amato RC. Evaluating the efficacy of a biofeedback intervention to reduce children's anxiety. J Clin Psychol 1996 Jul;52(4):469-473 [doi: 10.1002/(sici)1097-4679(199607)52:4<469::aid-jclp13>3.0.co;2-e] 55. Jerčić P, Sundstedt V. Practicing emotion-regulation through biofeedback on the decision-making performance in the context of serious games: a systematic review. Entertain Comput 2019 Mar;29:75-86 [doi: 10.1016/j.entcom.2019.01.001] 56. Hollis C, Falconer CJ, Martin JL, Whittington C, Stockton S, Glazebrook C, et al. Annual Research Review: digital health interventions for children and young people with mental health problems - a systematic and meta-review. J Child Psychol Psychiatry 2017 Apr 10;58(4):474-503 [doi: 10.1111/jcpp.12663] [Medline: 27943285] https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 15 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al 57. Lüddecke R, Felnhofer A. Virtual reality biofeedback in health: a scoping review. Appl Psychophysiol Biofeedback 2022 Mar 03;47(1):1-15 [FREE Full text] [doi: 10.1007/s10484-021-09529-9] [Medline: 34860290] 58. Lin IM, Fan SY, Yen CF, Yeh YC, Tang TC, Huang MF, et al. Heart rate variability biofeedback increased autonomic activation and improved symptoms of depression and insomnia among patients with major depression disorder. Clin Psychopharmacol Neurosci 2019 May 31;17(2):222-232 [FREE Full text] [doi: 10.9758/cpn.2019.17.2.222] [Medline: 30905122] 59. Zucker TL, Samuelson KW, Muench F, Greenberg MA, Gevirtz RN. The effects of respiratory sinus arrhythmia biofeedback on heart rate variability and posttraumatic stress disorder symptoms: a pilot study. Appl Psychophysiol Biofeedback 2009 Jun;34(2):135-143 [doi: 10.1007/s10484-009-9085-2] [Medline: 19396540] 60. Lehrer P, Vaschillo B, Zucker T, Graves J, Katsamanis M, Aviles M, et al. Protocol for heart rate variability biofeedback training. Biofeedback 2013;41(3):98-109 [doi: 10.5298/1081-5937-41.3.08] 61. Lin IM, Tai LY, Fan SY. Breathing at a rate of 5.5 breaths per minute with equal inhalation-to-exhalation ratio increases heart rate variability. Int J Psychophysiol 2014 Mar;91(3):206-211 [doi: 10.1016/j.ijpsycho.2013.12.006] [Medline: 24380741] 62. Al Osman H, Dong H, El Saddik A. Ubiquitous biofeedback serious game for stress management. IEEE Access 2016 Mar 31;4:1274-1286 [doi: 10.1109/access.2016.2548980] 63. Gomes LM, Silva RG, Melo M, Silva NN, Vanderlei FM, Garner DM, et al. Effects of effortful swallow on cardiac autonomic regulation. Dysphagia 2016 Apr 9;31(2):188-194 [doi: 10.1007/s00455-015-9676-4] [Medline: 26650792] 64. Tanoue K, Sonoda M, Yamashita E, Tanaka H, Nuruki N. Swallowing-induced atrial tachyarrhythmia triggered by solid foods. Circulation 2014 Sep 23;130(13):e113-e115 [doi: 10.1161/circulationaha.114.010235] 65. Siepmann M, Aykac V, Unterdörfer J, Petrowski K, Mueck-Weymann M. A pilot study on the effects of heart rate variability biofeedback in patients with depression and in healthy subjects. Appl Psychophysiol Biofeedback 2008 Dec 19;33(4):195-201 [doi: 10.1007/s10484-008-9064-z] [Medline: 18807175] 66. Pyne JM, Constans JI, Nanney JT, Wiederhold MD, Gibson DP, Kimbrell T, et al. Heart rate variability and cognitive bias feedback interventions to prevent post-deployment PTSD: results from a randomized controlled trial. Mil Med 2019 Jan 01;184(1-2):e124-e132 [FREE Full text] [doi: 10.1093/milmed/usy171] [Medline: 30020511] 67. Yıldız M, Doma S. Issues related to reliability of HRV analysis and effect of spontaneous saliva swallowing on HRV. Anatol J Cardiol 2017 Jun;17(6):495-496 [FREE Full text] [doi: 10.14744/AnatolJCardiol.2017.7646] [Medline: 28617300] Abbreviations ANS: autonomic nervous system HF: high frequency HR: heart rate HRV: heart rate variability HRVB: heart rate variability biofeedback iHR: instantaneous heart rate LF: low frequency RMSSD: root mean square of differences between adjacent RR intervals RR mean: average RR intervals Edited by G Eysenbach; submitted 10.02.23; peer-reviewed by J Robertson, K Stawarz; comments to author 17.04.23; revised version received 12.06.23; accepted 18.07.23; published 24.08.23 Please cite as: Estrella T, Alfonso C, Ramos-Castro J, Alsina A, Capdevila L A Serious Game to Self-Regulate Heart Rate Variability as a Technique to Manage Arousal Level Through Cardiorespiratory Biofeedback: Development and Pilot Evaluation Study JMIR Serious Games 2023;11:e46351 URL: https://games.jmir.org/2023/1/e46351 doi: 10.2196/46351 PMID: ©Tony Estrella, Carla Alfonso, Juan Ramos-Castro, Aitor Alsina, Lluis Capdevila. Originally published in JMIR Serious Games (https://games.jmir.org), 24.08.2023. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Serious Games, is properly cited. The complete bibliographic https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 16 (page number not for citation purposes) XSL• FO RenderX JMIR SERIOUS GAMES Estrella et al information, a link to the original publication on https://games.jmir.org, as well as this copyright and license information must be included. https://games.jmir.org/2023/1/e46351 JMIR Serious Games 2023 | vol. 11 | e46351 | p. 17 (page number not for citation purposes) XSL• FO RenderX