Gamification of physical activity: A systematic literature review of comparison studies Jonna Koivisto1 [0000-0002-6631-2571] and Juho Hamari1,2 [0000-0002-6573-588X] 1 Gamification Group, Faculty of Information Technology and Communication Sciences, Tampere University, Tampere, Finland 2 Gamification Group, Faculty of Humanities, University of Turku, Turku, Finland jonna.koivisto@tuni.fi juho.hamari@tuni.fi Abstract. Gamification is commonly implemented with the goal of transforming activities, systems and services to afford similar experiences and motivational support as games do. In health and exercise contexts, the motivational support drawn from games is considered to encourage performing these activities that commonly lack motivation. However, an empirically rigorous body of literature examining the effects of gamification has been lacking. This is especially prob- lematic in health contexts where unfounded claims can have detrimental effects. This systematic literature review of 16 comparison studies on gamification of physical activity examines what kinds of gamification have been studied in the pursuit of which outcomes, and what results the studies have attained. The results show that gamification of physical activity has provided positively oriented re- sults; however, with more rigorous study designs the results are less optimistic. Research is focused on measuring performed physical activity, but mostly relies on self-reported data instead of objective measurement. Keywords: gamification, physical activity, literature review. 1 Introduction and background Health and exercise are among the most common contexts for gamification ventures, both in research and in practice [1][18][21]. Gamification refers to transforming activ- ities, systems and services towards affording similar experiences as games are consid- ered to afford [17]. As motivational benefits are perceived to be at the core of games [13][28][35], gamification is commonly employed in contexts where people commonly lack motivation such as education, work and healthcare [21][24][30][43][41][27]. Gamification presents an especially interesting technology in the area of physical activity as games are sometimes perceived to encourage a sedentary lifestyle (see e.g. [29][32][38]). However, there has been an in-flux of location-based games (such as Pokémon Go) [22][25] as well as exergames [32] that further have made gaming rele- vant in terms of physical health. Beyond physical activity becoming a way to play games, intentional gamification further attempts to adopt the motivational facets of gaming and implementing them into pursuits with direct health outcomes in mind. GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 106 However, while gamification has been popularly and academically predicted to be a powerful technology for engagement and behavioral change (e.g. [11]), the field has still lacked an empirically rigorous body of literature examining its effects [21][31][34]. This has especially been the case for the health sciences where the thresholds for sci- entific rigor in terms of research design can be considered higher than in the transdis- ciplinary mother fields of game studies [1][18][21][33]. Therefore, this study presents a systematic literature review of existing comparison studies (16 studies) examining the effectiveness of gamification on physical activity- related outcomes. The focus of this review is on studies conducted with adult partici- pants; that include a gameful intervention and a comparative study setting meaning that the intervention results are contrasted to parallel conditions or a baseline measurement; and that report subjective or objective outcomes related to physical activity. The review investigates how gamification of physical activity has been implemented, what out- comes have been addressed with the gameful interventions, and what results have been attained regarding these outcomes. 2 Review procedure The literature search was conducted in 11/2018 in Scopus, Web of Science, and Pub- Med databases. The searches were conducted using search terms covering the termi- nology presented in Table 1. The search strategies presented by Schoeppe et al. [36] were used as reference for physical activity –related search terminology. The specific search strings were formulated according to the search logic of each database, but con- taining the same terminology. Table 1 also reports the number of records retrieved from each database. Table 1. Search terms used and amount records received from databases. Search terms gamif* AND health* AND "physical activity" OR walk* OR "physical fitness" OR "physical health" OR "leisure activity" OR "motor activity" OR exercis* OR sport* OR sedentary OR sitting OR inactiv* OR step* OR pedomet* OR acceleromet* Database Number of records retrieved Scopus (search limited to Title-Abstract-Keywords) 198 Web of Science (searches conducted as Topic searches) 88 Pubmed 70 TOTAL 356 The literature searches resulted in a total of 356 records. After removal of duplicates and records not containing a study (e.g. proceedings books), the remaining 243 records were screened for inclusion based on the predetermined PICOS-criteria (see Table 2). The full literature identification, screening and eligibility evaluation process is reported in Figure 1. GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 107 Table 2. Review questions and inclusion criteria (PICOS-criteria). Review questions Inclusion criteria Population Adults (≥18 years); participant mean age ≥18 years Intervention Game or gamification intervention targeting physical activity Comparisons No intervention, an active control intervention, baseline measurement, or standard treatment/rehabilitation Outcomes Quantitatively measured user-related subjective and objective out- comes Study design Quantitative comparison study written in English 356 records identified through database searches Identification 243 records after removal of duplicates and removal of conference proceeding books 147 records excluded based on: Screening 243 records screened - Title (N=26) - Abstract (N=121) 81 full-text articles excluded due to reasons: - No controlled study setting (N=54) Eligibility - Not empirical research; study protocols; work-in-progress (N=9) - Based on same intervention as in another included study by same 96 full-text articles assessed research team (N=6) for eligibility - No actual physical activity related Included measurements reported (N=4) - Full paper not in English (N=3) 1 study identified - Study population mean age < 18 years through the full-text 16 articles included in the (N=2) assessment of final synthesis - Not enough details on how articles measurement conducted (N=2) - Study population mostly children (N=1) Fig. 1. A flowchart describing the study selection procedure. After the rigorous screening and eligibility evaluation of titles, abstracts, and finally, full-text articles, 15 studies were identified as eligible for the final synthesis. In addi- tion, one record was identified based on a reference in another full-text article, evalu- ated as eligible for inclusion, and thus, included to the review. Therefore, the final num- ber of studies included in the synthesis is 16. The literature selection process was carefully documented using Refworks reference management tool and Microsoft Excel for transparency. The analyses were conducted using the guidelines provided by Webster and Watson [42].The literature selection pro- cess and analyses were conducted by the main author. GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 108 3 Analysis A clear majority of the reviewed studies were published in journals (12/16) instead of conferences (4/16). The gameful interventions for increasing physical activity have mainly relied on the same affordances as the gamification field in general [21][37] (see Table 3). Point- based mechanics and activity goals were identified in half of the reviewed studies, ac- companied often with performance rankings and visualizations of one’s performance. Interestingly, the health gamification field has also included collaboration-based me- chanics in their interventions as indicated by the high frequency of teams as a gameful affordance. Noteworthy, however, are also the less common but innovative affordances included in the studies, for example, real-world activity, e.g. steps, being transformed into a game currency in a virtual world [40], and social contracts and duel competitions between individuals for engaging individuals in physical activity [39]. Table 3. Affordances included in the reviewed studies . Affordance Studies including Freq. Points, score [2][5][7][8][16][23][39][44] 8 Goals [7][8][9][16][23][26][39][44] 8 Leaderboard [2][7][14][40][44] 5 Progress visualization [5][7][16][26][39] 5 Teams, leagues [8][14][16][23][26] 5 Virtual rewards [2][14][26][39] 4 Full game, thus affordances not identified [3][4][12][19] 4 Badges [2][5][7] 3 Messaging with users/team/clinician [5][8][26] 3 Levels [7][23][39] 3 Team progress visualization [8][16][26] 3 Quizzes [2][7] 2 Real-world rewards [2][16] 2 Challenges [7][14] 2 Personalized feedback/messages [14][39] 2 Virtual losses [14] 1 Virtual reality (VR) environment [9] 1 Shadowing (comparing current to earlier performance) [9] 1 Virtual environment with city building/management [40] 1 Virtual tracking based on real-world action [14] 1 Duel-type competitions [39] 1 Game currency based on steps [40] 1 Social contracts [39] 1 GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 109 Table 4. Outcome measures studied in the reviewed studies . Subjective/ Outcome measure Studies including Freq. objective Physical activity* S [2][3][7][16][19][26][39][40] 8 Duration of usage, active time etc.* O [4][12][14][44] 4 Physical activity/performance* O [5][14][23][44] 4 Engagement/adherence with app/solu- O [8][26][44] 3 tion Knowledge related to condition, S [2][7] 2 health, physical activity etc. Energy expenditure* O [4][12] 2 Enjoyment of physical activity S [12][40] 2 Healthcare utilization S [2] 1 Medication overuse S [2] 1 Empowerment S [2] 1 Perceived benefits of game on health S [3] 1 Perceived exertion* S [4] 1 Affect S [4] 1 Duration of activity* S [19] 1 Messaging within service O [5] 1 Self-efficacy S [7] 1 Glycated hemoglobin (HbA1c) levels* O [8] 1 Expired gases* O [9] 1 Oxygen uptake capacity* O [9] 1 Intrinsic motivation S [9] 1 Subjective vitality S [9] 1 Future exercise intentions S [9] 1 Heart rate* O [12] 1 Perception of game S [14] 1 Outdoor time S [19] 1 Weight data* O [23] 1 Quality of life S [26] 1 Gaming motivation S [19] 1 *Measurement related to actual physical activity or actual physical outcomes Most of the reviewed studies focused on physical activity or the duration of the ac- tivity as the outcome measure of the intervention (see Table 4). Half of the reviewed studies were mainly based on subjectively and half on objectively measured data. The most commonly used self-reported measurement instrument in the reviewed literature GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 110 was the International Physical Activity Questionnaire (IPAQ) [6]. Objectively meas- ured data was in most cases retrieved from an app or an activity tracker tool, e.g. FitBit, used in the intervention. In total, only 11 outcome measurements of the total 28 were related to the actual physical activity or actual physical outcomes. Rest of the outcomes were either other health-related behavioral or psychological measurements, or behavioral or perception measurements related to the solution used in the intervention. A clear majority of the outcomes have been studied only in one study each. Details of the study designs are reported in Table 5. 10 out of the 16 studies were randomized controlled trials (RCT), 3 studies reported a partially randomized study design, and 3 were categorized as comparison studies without an actual experimental setting. Only one of the RCT studies was reported to be fully blinded, while 4 studies reported single-blinded or partially blinded designs. The remaining RCT studies pro- vided no information about blinding procedures with the exception of one study stating that the setting was not blinded. 8 of the 16 studies reported either fully or partially positive results from the gameful interventions related to the physical health outcomes. However, 7 of the 16 studies re- ported that the gameful intervention did not show statistically significant improvements compared to the comparison conditions or report equally positive and negative results regarding the effects of the gameful intervention on physical activity –related outcomes. Table 5 reports also the study designs regarding the comparison. The analysis sug- gests that the studies with more rigorous study designs, i.e. fully controlled settings, have less positive results regarding the physical activity –related outcomes than study settings with baseline measurements as comparisons. Furthermore, in four studies the intervention was a full commercial game and thus individual affordances were not iden- tified or studied. 4 Discussion This systematic review focused on examining how gamification has been implemented for the goal of increasing physical activity, what outcomes this body of literature has examined, and finally, what kinds of results the comparison studies on gamification physical activity have attained. Only 16 comparison studies were identified for the re- view, which is surprising given the prevalence of gamification in physical activity [21]. The affordances implemented in the gameful interventions for increasing physical activity have followed the common patterns identified in gamification literature in gen- eral [21][37]. Points and leaderboards are the most common elements implemented alongside goals and progress visualization tools. Goals concretize the target behavior, progress visualizations provide support and indicators of progression toward the health goal, and points act as a virtual reward for the target behavior. Interestingly, for exam- ple, collaborative affordances were quite often implemented within the body of litera- ture [8][14][16][23][26][39], which is not such a common approach within the gamifica- tion research field in general [21]. GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 111 Table 5. Study details. Intervention Study participants N1 Timeframe Study design Intervention compared to Results2 Developed web-based inter- Rheumatoid arthritis patients Randomized controlled trial, sin- Control group without [2] vention for rheumatoid arthri- Mean age 57.95, SD 12.29 155 4 months gle-blinded study design access to intervention Fully + tis patients 54.2% male, 45.8% female Players of Pokémon Go Player mean age 26.8, SD 461 Null or Comparison study with a repeated Non-players of Pokémon [3] Pokémon Go 8.2 (base- 3 months measures design Go equal 193 males, 265 females, 3 line) +/- transgenders Students and university staff Control group with same Dance Central game for Xbox Randomized controlled trial; no in- [4] Kinect Mean age 26.5, SD 7.1 44 1 session formation on blinding intervention, primed as Partial - 56.8% male non-game Developed mobile fitness ap- Students and university staff Randomized controlled repeated [5] plication that connects to Fit- Age mainly 20-30 years 36 10 days measures study; no information on Baseline measurement Fully + Bit 15 males, 21 females blinding Healthy employees Developed app for promotion Age ≤ 35 30%; 36 to 45 Randomized controlled trial, no in- Control group without Partial [7] of physical activity connected 30%; ≥ 46 40% in IG 144 6 weeks formation on blinding access to intervention + to FitBit 61.3% male in IG Veterans with type 2 diabe- tes Developed mHealth tool con- Randomized controlled trial, not Standard care for type 2 Partial [8] nected to FitBit Mean age 67.56, SD 5.81 27 13 weeks blinded diabetes patients + 26 males, 3 females (of ini- tial 29 participants) Sedentary or recreationally Partially randomized cross-over Non-gamified VR er- VR-solution for a HIIT cy- active adults Partial [9] cling exercise Mean age 22, SD 4 16 4 sessions study (order of sets randomized), gometry as control meas- + no information on blinding urement 8 males, 8 females Experimental games for a sta- Fitness center customers Partially randomized cross-over Control measurement Partial [12] tionary bike and a rowing ma- Mean age 31.5 24 1 session study (order of exercises random- without games + chine 9 males, 15 females ized), no information on blinding GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 112 Sedentary office workers Null or Developed mobile health Mean age 40,6. SD 11,7 in Randomized controlled trial, at Control group using Fit- [14] game used with FitBit IG 144 10 weeks least partially blinded Bit only equal +/- 79.2% female in IG Developed location-based Town residents game with an online platform Baseline measurement [16] for self-reporting physical ac- 62,6% over 18 years 329 7 days Comparison study before intervention Fully + 38.0% of participants male tivity Pokémon Go players [19] Pokémon Go Mean age 23.4, SD 5.88 444 6 weeks Comparison study Baseline measurement Fully + 50.75% male, 49.3% female Developed research platform Overweight adults Null or Randomized, controlled trial, fully Control group without [23] used alongside Withings scale Mean age 41.4 196 36 weeks blinded study design. access to intervention equal and Withings app. 85.7% of participants female +/- Insufficiently active adults Control group having Developed physical activity Age 18 to <25 23.6%; 25 to Null or Randomized controlled trial, sin- teams and health moni- [26] intervention delivered via a <35 29.1%; 35 to <45 110 20 weeks gle-blinded study design toring, no access to inter- equal Facebook app 26.4%; 45 to 65 17.3% +/- vention 70.9% female Healthy adults Null or Developed online, interactive Randomized controlled trial, sin- Control group without [39] physical activity tool Mean age 55.3, SD 11.2; 21 3 months gle-blinded study design access to intervention equal 11 males, 10 females +/- Adults Partially randomized repeated Null or Developed web-based social Mean age 37.7, SD 10.18 measures study (order of condi- Control group using Fit- [40] game connected to Fitbit 17 male, 44 females (of ini- 50 30 days tions randomized), no information Bit only equal +/- tial survey respondents) on blinding Undergraduate communica- Control using quantified Null or Developed prototype promot- tions students Randomized controlled trial, no in- version of the app with- [44] ing active walking Mean age 23.39, SD 1.40 59 10 days formation on blinding out gamification equal +/- 44 females, 15 males, 1 Both intervention and control groups included in N, if the study was controlled. IG = intervention group, CG = control group. 2 Fully + = fully positive results, Partial + = partially positive results, Null or equal +/- = no effects or an equal amount of positive and negative results reported, Partial - = partially negative results, Fully - = fully negative results GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 113 The gameful interventions studied in the current body of literature mainly included several affordances simultaneously rather than having investigated the effects of singu- lar gamification affordances individually. Thus most of the studies were not able to identify which affordances were more effective than others regarding the physical health –related outcomes. Furthermore, a few studies examined the effects of a full commercial game on physical activity behavior [3][4][12][19]. For example, Broom and Flint [3] and Kaczmarek et al. [19], focused on the effects of Pokémon Go on the physical activity of the users. When the gameful intervention is based on a full game with various features, it is similarly impossible to identify which aspects of the game have lead to the detected results. Thus, studying isolated game elements is encouraged [21]. Similarly, Schoeppe et al. [36] has suggested that the research on health and fit- ness apps should seek to study effects of singular features in order to identify the effec- tive app features from the ineffective ones. The analysis of the current body of literature indicates that most of the studies meas- ured performed physical activity as the outcome measurement. However, many of the studies also relied on subjective self-reported data instead of objective measurement. The most commonly implemented self-report measurement instrument for physical ac- tivity was the International Physical Activity Questionnaire (IPAQ) [3][7][16][19][40]. While gathering self-reported data is often a more cost-efficient way of data gathering compared to, for example, collecting sensor data, more reliable results would be gained with triangulation of data combining both, subjective and objective measurement. Previous research on gamification has also identified novelty effects to impact the outcomes of the gamification solutions [20][10][15]. The results presented by Gremaud et al. [14] provide further evidence of the decline of the effects with time. Data trian- gulation could provide important insights in future research also about reasons for the declining effects. As shown by the analysis of the outcomes measured in the reviewed literature, the variety of different outcomes is large and many outcomes are examined only in one isolated study. Therefore, the current body of literature still lacks in cumulation of re- search on the same outcome measures and replication. Previous literature reviews on gamification have suggested that the research field would benefit from seeking to use validated measurement instruments in order to accumulate the knowledge regarding specific outcomes [21]. The results of the reviewed literature provide support for prior findings, that on a general level, the results regarding the effectiveness of gamification on the outcome variables are positively oriented [21]. However, when scrutinizing the results and con- trasting them with the study designs in the reviewed papers, the more rigorous study settings seem to provide more neutral results. The fully positive findings have been reported mainly in studies with baseline measurements as comparisons [5][16][19] in- stead of study settings including a randomized design with control conditions. These findings suggest that gamification of physical activity provides promising results of its effectiveness, but more research with controlled study settings would be needed to sub- stantiate the promises. Study designs with full randomization and control conditions are especially im- portant in the context of health-related activities to indicate whether the interventions GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 114 can provide health benefits. The requirements for the study designs are even stricter when examining the benefits of a gameful solution as part of clinical healthcare. For this review, a full quality assessment of the studies was not conducted, mainly because not all of the studies were RCTs. The analysis indicated, however, that the studies lack in e.g. blinding procedures. In order to be able to provide convincing results to justify use of gameful interventions, for example, as part of healthcare practices, the quality of the study setting and designs would benefit from improving. Similar suggestions for study designs have been provided by Schoeppe et al. [36] for research on health and fitness apps. As noted in literature discussing the development of the research on gamification [24][31][21], it has taken some years for the research field to develop the methodolog- ical approaches to include more comparative research designs. Thus, the fact that a literature review focusing solely on comparison studies on gamification of a specific topic can today be conducted can be considered a sign of the maturation of the field. Acknowledgements This work has been supported by Business Finland (376/31/2018 and 5479/31/2017) and participating partners, Satakunnan korkeakoulusäätiö and its collaborators, and Academy of Finland (Center of Excellence in Game Culture Studies). References [1] Alahäivälä, T., Oinas-Kukkonen, H.: Understanding persuasion contexts in health gamifica- tion: A systematic analysis of gamified health behavior change support systems literature. Inter- national Journal of Medical Informatics, 96, 62-70 (2016). [2] Allam, A., Kostova, Z., Nakamoto, K., Schulz, P. J.: The effect of social support features and gamification on a web-based intervention for rheumatoid arthritis patients: Randomized con- trolled trial. Journal of Medical Internet Research, 17(1), e14 (2015). [3] Broom, D. R., Flint, S. W.: Gotta catch 'em all: Impact of pokemon go on physical activity, sitting time, and perceptions of physical activity and health at baseline and three-month follow- up. Games for Health Journal, 7(6) (2018). [4] Chen, F. X., King, A. C., Hekler, E. B.: "Healthifying" exergames: Improving health out- comes through intentional priming. In: Proceedings of the Annual ACM Conference on Human Factors in Computing Systems, CHI 2014, 1855-1864 (2014). [5] Chen, Y., Pu, P.: HealthyTogether: Exploring social incentives for mobile fitness applica- tions. In: Proceedings of the Second International Symposium of Chinese CHI, 25-34 (2014). [6] Craig. C.L., Marshall, A.L., Sjöström, M. et al.: International physical activity questionnaire: 12-country reliability and validity. Medicine & Science in Sports & Exercise, 35(8), 1381-1395 (2003). [7] Dadaczynski, K., Schiemann, S., Backhaus, O.: Promoting physical activity in worksite set- tings: Results of a german pilot study of the online intervention healingo fit. BMC Public Health, 17(1), 6 (2017). [8] Dugas, M., Crowley, K., Gao, G. G., Xu, T., Agarwal, R., Kruglanski, A. W., Steinle, N.: Individual differences in regulatory mode moderate the effectiveness of a pilot mHealth trial for diabetes management among older veterans. PloS One, 13(3), e0192807 (2018). GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 115 [9] Farrow, M., Lutteroth, C., Rouse, P. C., Bilzon, J. L. J.: Virtual-reality exergaming improves performance during high-intensity interval training. European Journal of Sport Science, 1-9 (2018). [10] Farzan, R., DiMicco, J. M., Millen, D. R., Dugan, C., Geyer, W., Brownholtz, E. A.: Results from deploying a participation incentive mechanism within the enterprise. In: Proceedings of CHI, Florence, Italy, April 5–10, 2008, 563–572 (2008). [11] Gartner: Gartner says by 2015, more than 50 percent of organizations that manage innova- tion processes will gamify those processes (2011). < http://www.gart- ner.com/it/page.jsp?id=1629214 > (Accessed 19.12.2018) [12] Geelan, B., Zulkifly, A., Smith, A., Cauchi-Saunders, A., De Salas, K., Lewis, I.: Aug- mented exergaming: Increasing exercise duration in novices. In: Proceedings of the 28th Aus- tralian Computer-Human Interaction Conference, OzCHI 2016, 542-550 (2016). [13] Granic, I., Lobel, A., Engels, R. C.: The benefits of playing video games. American psy- chologist, 69(1), 66 (2014). [14] Gremaud, A. L., Carr, L. J., Simmering, J. E., Evans, N. J., Cremer, J. F., Segre, A. M., Polgreen, L. A., Polgreen, P. M.: Gamifying accelerometer use increases physical activity levels of sedentary office workers. Journal of the American Heart Association, 7(13), 10 (2018). [15] Hamari, J.: Transforming Homo Economicus into Homo Ludens: A Field Experiment on Gamification in a Utilitarian Peer-To-Peer Trading Service. Electronic Commerce Research and Applications, 12 (2013). [16] Harris, M. A.: Beat the street: A pilot evaluation of a community-wide gamification-based physical activity intervention. Games for Health Journal 7(3), 208-212 (2018a). [17] Huotari, K., Hamari, J.: A definition for gamification: anchoring gamification in the service marketing literature. Electronic Markets, 27(1), 21-31 (2017). [18] Johnson, D., Deterding, S., Kuhn, K. A., Staneva, A., Stoyanov, S., Hides, L.: Gamification for health and wellbeing: A systematic review of the literature. Internet interventions, 6, 89-106 (2016). [19] Kaczmarek, L. D., Misiak, M., Behnke, M., Dziekan, M., Guzik, P.: The pikachu effect: Social and health gaming motivations lead to greater benefits of Pokémon GO use. Computers in Human Behavior 75, 356-363 (2017). [20] Koivisto, J., Hamari, J.: Demographic differences in perceived benefits from gamification. Computers in Human Behavior, 35, 179-188 (2014). [21] Koivisto, J., Hamari, J.: The rise of motivational information systems: A review of gamifi- cation research. International Journal of Information Management 45, 191-210 (2019). [22] Koivisto, J., Malik, A., Gurkan B., Hamari, J.: Getting healthy by catching them all: A study on the relationship between player orientations and health benefits in an augmented reality game. In: Proceedings of the 52nd Annual Hawaii International Conference on System Sciences (HICSS), Hawaii, USA, January 8-11, 2019 (2019). [23] Kurtzman, G. W., Day, S. C., Small, D. S., Lynch, M., Zhu, J., Wang, W., Rareshide, C. A. L., Patel, M. S.: Social incentives and gamification to promote weight loss: The LOSE IT ran- domized, controlled trial. Journal of General Internal Medicine 33(10), 1669-1675 (2018). [24] Landers, R. N., Auer, E. M., Collmus, A. B., Armstrong, M. B.: Gamification science, its history and future: Definitions and a research agenda. Simulation & Gaming, 49(3), 315-337 (2018). [25] Leorke, D.: Location-Based Gaming: Play in Public Space. Springer (2018). [26] Maher, C., Ferguson, M., Vandelanotte, C., Plotnikoff, R., De Bourdeaudhuij, I., Thomas, S., Nelson-Field, K., Olds, T.: A web-based, social networking physical activity intervention for insufficiently active adults delivered via Facebook app: randomized controlled trial. Journal of medical Internet research 17(7) (2015). [27] Majuri, J., Koivisto, J., Hamari, J.: Gamification of education and learning: A review of empirical literature. In: Proceedings of the 2nd International GamiFIN Conference, 11-19 (2018). GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 116 [28] Malone, T. W.: Toward a theory of intrinsically motivating instruction. Cognitive science, 5(4), 333-369 (1981). [29] Martínez-González, M. Á., Martinez, J. A., Hu, F. B., Gibney, M. J., Kearney, J.: Physical inactivity, sedentary lifestyle and obesity in the European Union. International Journal of Obe- sity, 23(11), 1192 (1999). [30] Morschheuser, B., Hamari, J., Koivisto, J., Maedche, A.: Gamified Crowdsourcing: Con- ceptualization, literature review, and future Agenda. International Journal of Human-Computer Studies, 106, 26-43 (2017). [31] Nacke, L. E., Deterding, C. S.: The maturing of gamification research. Computers in Human Behaviour, 450-454 (2017). [32] Peng, W., Lin, J. H., Crouse, J.: Is playing exergames really exercising? A meta-analysis of energy expenditure in active video games. Cyberpsychology, Behavior, and Social Networking, 14(11), 681-688 (2011). [33] Quandt, T., Van Looy, J., Vogelgesang, J., Elson, M., Ivory, J. D., Consalvo, M., Mäyrä, F.: Digital games research: A survey study on an emerging field and its prevalent debates. Journal of Communication 65(6), 975-996 (2015). [34] Rapp, A., Hopfgartner, F., Hamari, J., Linehan, C., Cena, F.: Strengthening gamification studies: Current trends and future opportunities of gamification research. International Journal of Human-Computer Studies (2018). [35] Ryan, R. M., Rigby, C. S., Przybylski, A.: The motivational pull of video games: A self- determination theory approach. Motivation and emotion, 30(4), 344-360 (2006). [36] Schoeppe, S., Alley, S., Van Lippevelde, W., Bray, N. A., Williams, S. L., Duncan, M. J., Vandelanotte, C.: Efficacy of interventions that use apps to improve diet, physical activity and sedentary behaviour: a systematic review. International Journal of Behavioral Nutrition and Physical Activity, 13(1), 127 (2016). [37] Seaborn, K., Fels, D. I.: Gamification in theory and action: A survey. International Journal of human-computer studies 74, 14-31 (2015). [38] Strum, R.: Childhood obesity—what we can learn from existing data on societal trends, part 1. Preventing Chronic Diseases 2:A12 (2005). [39] Thorsteinsen, K., Vittersø, J., Svendsen, G. B.: Increasing physical activity efficiently: An experimental pilot study of a website and mobile phone intervention. International Journal of Telemedicine and Applications (2014). [40] Walsh, G., Golbeck, J.: StepCity: A preliminary investigation of a personal informatics- based social game on behavior change. In: CHI'14 Extended Abstracts on Human Factors in Computing Systems 2371-2376 (2014). [41] Warmelink, H., Koivisto, J., Mayer, I., Vesa, M., Hamari, J.: Gamification of production and logistics operations: Status quo and future directions. Journal of Business Research (2018). [42] Webster, J., Watson, R. T.: Analyzing the Past to Prepare for the Future: Writing a Literature Review. MIS Quarterly, 26(2), xiii–xxiii (2002). [43] Xi, N., Hamari, J.: Does gamification satisfy needs? A study on the relationship between gamification features and intrinsic need satisfaction. International Journal of Information Man- agement, 46, 210-221 (2019). [44] Zuckerman, O., Gal-Oz, A.: Deconstructing gamification: Evaluating the effectiveness of continuous measurement, virtual rewards, and social comparison for promoting physical activity. Personal and Ubiquitous Computing, 18(7), 1705-1719 (2014). GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 117