How Effective is Personalization in Persuasive Interventions for Reducing Sedentary Behav- ior and Promoting Physical Activity: A Sys- tematic Review Noora Aldenaini1, 2, Rita Orji1, and Srinivas Sampalli1 1 Faculty of Computer Science, Dalhousie University, Halifax, NS, Canada 2 Department of Computer Science, Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia nr412864@dal.ca, rita.orji@dal.ca, srini@cs.dal.ca Abstract. The use of persuasive systems and devices to reduce sedentary lifestyles and encour- age individuals to be more physically active is progressively gaining interest. This paper presents a 13 years review (from 2006 to 2019) of personalized persuasive technologies (PTs) for promot- ing physical activity (PA) and discouraging sedentary behavior (SB). Moreover, we decon- structed the various implementations and operationalizations of the personalization in these PTs and compared their effectiveness. Specifically, this paper aims to (1) shed light on the multiple ways of implementing personalization in these PTs to promote PA and reducing SB, (2) evaluate the effectiveness of personalized PTs for promoting PA and decreasing SB, (3) highlight trends and offer suggestions for research in the area of personalizing PTs. Keywords: Persuasive technology, Physical activity, Sedentary behavior, Personalization, Goal recommendations, Activity recommendations, Motivation, Health 1 Introduction Persuasive technologies (PTs) are interactive computing systems, apps, or devices that are purposely developed to influence users to adopt healthy behaviors and attitudes and prevent risky ones without using coercion or deception [15], [41]. PTs are mainly im- plemented to promote healthy behavior and prevent disease or to manage diseases or other health conditions [38], [40], but also have been used in other domains. Persuasive 2020, Adjunct proceedings of the 15th Int. conference on Persuasive Technology. Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). 2 Sedentary behavior (SB) is generally characterized as a long time sitting behavior – when an individual expends lower or equal to 1.5 metabolic equivalent (£ 1.5 MET) [57], [44]. MET “is defined as the amount of oxygen consumed while sitting at rest and is equal to 3.5 ml O2 per kg body weight x min” [25]. Accordingly, SB and an insuffi- cient amount of PA are global health concerns, as they lead to obesity and morbidity, and they are the fourth leading reason for mortality globally, with an estimated 3.2 million deaths worldwide [60]. Therefore, increasing or maintaining a suitable level of moderate-intensity PA is essential for avoiding or mitigating different diseases and health complications such as obesity, cancer, diabetes, and cardiovascular diseases [12], [19]. Persuasive Technology (PT) interventions are considered powerful tools for helping people to adopt healthy behaviors such as maintaining or increasing PA levels and re- ducing sedentary lifestyles. PTs are often deployed using various technological plat- forms (e.g. fitness activity trackers and sensors, smartphones, websites, social network- ing sites (SNSs), games, desktop computers, and ambient displays) [1], [41]. Activity trackers, for instance, are often used to track and monitor personal data (e.g. the user’s PA level, step count, heart rate, and time spent sedentary), and these activity tracker devices are used mostly with other technology platforms (e.g. smartphones, web-based apps) to provide users feedback about their PA progress, personalized feedback, and personalized recommendations [1], [32], [26]. Research has suggested that personalizing PTs increases their effectiveness [39]. Thus, personalizing PT intervention is crucial because people are different with respect to their personal preferences, activity levels, objectives, requirements, lifestyles, and health conditions, and even personality [42], [43], [46]. As a result, there is an increas- ing demand for PTs to be personalized to increase their effectiveness. The personaliza- tion strategy requires a system to provide personalized content or services to increase relevance, motivation, and persuasion effects [22]. There is an increasing number of reviews of PTs for health and wellness. However, most of these reviews focused on the general area of health and wellness and their gen- eral application of various persuasive strategies, for example, see Orji and Moffat [41], and Aldenaini et al. [1]. Research has highlighted the importance of personalizing PT, and the personalization strategy is one of the most frequently employed persuasive strategies [3] used in PT designs. Hence, there is a need for in-depth research into ana- lyzing various implementations of personalization in persuasive strategies and the ef- fectiveness of personalized PTs. Therefore, this paper presents a 13-years review (from 2006 to 2019) of personalized PTs for promoting PA and discouraging SB. Moreover, we deconstructed the various implementations and operationalization of the personalization in these PTs to evaluate and compare their effectiveness. Specifically, this paper aims to (1) shed light on the various implementations a personalization in PT interventions for increasing PA and reducing SB, (2) evaluate the effectiveness of personalized PTs for promoting PA and decreasing SB, (3) highlight trends and offer suggestions for research in the area of personalizing persuasive technologies. 3 2 Related Works An increasing number of systematic reviews is being conducted to determine the effec- tiveness of PTs in various domains. Ghanvatkar et al. [19] provided a scoping review of personalization strategies employed in PA interventions. They included 49 eligible studies in their review paper. They examined personalization strategies in the form of feedback or recommendations. Furthermore, they identified six types of a personaliza- tion strategy based on different forms of implementation in their reviewed studies. These personalization types are summarized as shown in Table 1. There are other interesting review papers on PT interventions in the area of PA and/or SB in general – not focus specifically on personalization. For example, Almutari and Orji [2] provided a systematic review of articles that focused on PT for promoting PA. They analyzed the effectiveness of PT that employed social influence strategies such as comparison, cooperation, and competition only. Their findings revealed that PTs employing social support strategies to promote PA are promising in motivating users to be physically active. Similarly, Aldenaini et al. [1] conducted a 16-years systematic review (from 2003 to 2019) of PTs and their effectiveness for promoting PA and discouraging SB. They high- lighted trends in their outcomes such as research methods, behavioral theories, persua- sive strategies and different ways of implementing each strategy, system design, and employed technologies. Their findings revealed that employing PTs were effective and promising in promoting a desirable behavior change among different populations when employed with a suitable persuasive strategy. They also provided a list of interesting recommendations for advancing PTs’ future research. Furthermore, Wang et al. [58] conducted a systematic review of studies targeted at re- ducing SB in the work environment. They used the persuasive system design (PSD) model [22] to evaluate the effectiveness and utilization of PT in discouraging prolonged SB among office works. Their findings showed that a reminder was the most frequently employed PSD strategy. They also found that coupling a reminder strategy with educa- tion sessions was more promising than hourly reminders alone. Our systematic review paper included studies that employed a personalization strategy in their PT design to promote PA and/or reduce SB. In contrast, many existing studies tend to focus only on either promoting PA or reducing SB, and hardly on both. We also aimed at examining various ways of implementing personalization in different PTs. Again, our review specifically focused on PTs employing a personalization strategy in the area of SB and PA. Table 1: Types of Personalization [19]. Type of Personalization Meaning/Definition Goal Recommendations Quantified goals such as step count, floor count, duration of exercise, calorie burn rate. 4 Activity Recommendations Recommending a specific type of PA or behavior such as standing, walking, running, cycling. Fitness Partner Recommendations Matching a user of a system to other users who are similar and have the same target goals for motivat- ing them in maintaining or increasing their PA lev- els. Educational Content Increasing users' knowledge and awareness by sending personalized feedback about the health benefits of PA or some techniques and tips for im- proving PA. Motivational Content Motivating users to improve their PA by sending personalized motivational feedback and reinforce- ment messages. Intervention Timing Finding the right and suitable time to send a rec- ommendation or feedback to the users such as sending a notification reminder to a user at a suita- ble time and opportune moments 3 Methods and Coding Scheme As this paper aims to evaluate and analyze PT interventions (e.g. systems, apps, or websites) that implemented a personalization strategy, we used quantitative content analysis to classify data into different categories and compare between different out- comes [48]. We used Google Scholar, ACM Digital Library, Springer, PubMed, Elsevier Scopus, and EBSCOHost databases to search for and select relevant studies. We searched the terms “Persuasive Technology and Physical Activity”, “Persuasive Technology and Sedentary Behavior”, “Technology and Physical Activity Interventions”, “Technology and Sedentary Behavior Interventions”, “Fitness and Persuasive Technology”, “Pro- longed Sitting and Sedentary Lifestyles”, “Physical Activity Apps or Applications”, and “Sedentary Behavior Apps or Applications”. We also searched through the relevant references from the reviewed studies. We used Mendeley, a reference manager, to or- ganize the obtained articles. After searching through different databases, we identified 534 unique titles, of which 315 articles were excluded by title, while 219 articles were considered eligible based on title examinations. We investigated each title to check whether it falls within the scope of the review or not. We excluded those titles that targeted other health domains – domains not related to PA and/or SB (e.g. dental health, smoking cessation, unwanted pregnancy, eating habits, risky sexual behavior, alcohol drinking, etc.). After examin- ing the abstract of the 219 remaining articles, we excluded 181 articles, and we included a total of 38 articles. The included articles: were published in English between 2006 and 2019; implemented any form of a personalization strategy, included personalized 5 recommendations or feedback to promote PA and/or reducing SB; and were targeted at PA or SB, or both. We summarized the search and exclusion process, as shown in Fig- ure 1. We coded the 38 relevant articles by adapting the coding scheme created by Orji and Moffat [41] and validated by many other researchers, including [1], [3], [2]. As shown in Table 2, the coding sheet contains the study author(s), year of publication, domains, technology platforms, duration of a study, evaluation methodology, persuasive strate- gies, theories, targeted outcomes, audience age demographic, number of participants, results, and country of a study. We added a new identification to the coding sheet, which is the type of personalization as we adapted from Ghanvatkar et al. [19]. The types of personalization refer to the different ways of implementing a personalization strategy throughout the PT interventions. Appendix 1 provides a comprehensive overview of our coding sheet. Fig. 1. The Studies Selection Workflow. Table 2: Coding Scheme Analysis- Adapted from Orji and Moffat [41]. S/N Identification Example 1 Author(s)/Year Name of author(s) and year of publication. 2 Domain PA, SB, Mental Health, etc. 3 Technology Mobile, Web, Computer applications, etc. 4 Evaluation Methods Quantitative, Qualitative, and Mixed. 6 5 Types of Personalization Goal, activity, or fitness partner recommendations, educa- (Ways of implementa- tional or motivational content feedback, and intervention tim- tions) ing. 6 Persuasive Strategies Motivational strategies employed. 7 Duration of Evaluation Weeks, months, years, etc. 8 Theories Theories implemented either on the system design of a PT or on the evaluation of a study. 9 Targeted Outcomes Behavior, Motivation, Attitudes, Awareness, etc. 10 Audience Age De- Children, Adults, Old People, etc. mographics 11 Number of Participants How many participants involved in the study assessment? 12 Results Successful or not. 13 Country Country of a study where conducted. 4 Results Our analysis of existing PT interventions that applied personalization in their system designs for PA promotion and/or SB reduction revealed interesting outcomes. This sec- tion presents our findings from the reviewed studies in detail. 4.1 Persuasive Technology for Physical Activity and Sedentary Behavior by Year and Country Based on our inclusion and exclusion criteria, Figure 2 shows that most of the articles were published between 2016 and 2019. Three articles (8%) were published in each of 2010, 2012, 2013, and 2014. One article (3%) was published in each of 2006, 2007, 2008, and 2011. 7 Fig. 2. Persuasive Technologies for Physical Activity Promotion and Sedentary Behavior Re- duction Trends by Year. Figure 3 shows that the reviewed studies were conducted in 16 countries. The USA had the largest number of studies with a total of 17 (45%). The UK followed the USA with a total of 4 studies (11%). Netherlands and Canada were in third place with a total of 3 studies (8%) each. Fig. 3. Persuasive Technologies for Physical Activity Promotion and Sedentary Behavior Re- duction Trends by Country. 8 4.2 Targeted Domains All the studies included in this review were targeted at either promoting PA and/or discouraging SB. We categorized the targeted health domains into three groups: PA, SB, and mixed (the studies that focused on both PA promotion and SB reduction). Thirty-two studies (84%) were targeted at maintaining or increasing PA, while four studies (11%) focused on decreasing SB. Only two studies (5%) targeted both the PA and SB domains. Figure 4 presents the targeted health domains covered in this review paper. Fig. 4. Targeted Health Domains. 4.3 Evaluation Methodologies Used for Promoting Physical Activity and Reducing Sedentary Behavior Figure 5 shows the percentage of the total number of studies employing in each evalu- ation methodology. The evaluation methodologies covered in the reviewed studies are categorized into three main methods: qualitative, quantitative, and mixed (methods that include both qualitative and quantitative methodologies). The evaluation approach most commonly used in the studies was mixed methods with a total of 17 studies (45%). The qualitative methodology ranked second with a total of eight studies (21%). This is followed by the quantitative methodology with a total of five studies (13%). Eight stud- ies (21%) did not evaluate their PT designs. 9 Fig. 5. Evaluation Methods Used in the Reviewed Studies. 4.4 Effectiveness of Personalized PT based on Evaluation Methods Figure 6 shows that out of the 17 studies that employed a mixed-methods approach, ten studies (59%) reported fully successful outcomes, and seven studies (41%) reported partially successful outcomes. Fully successful outcomes mean those studies that re- ported all positive outcomes with respect to achieving their design objectives, as re- ported by the authors in their papers. By partially successful outcomes, we mean that the results of the PT evaluation show a mixture of both positive and negative outcomes respect to achieving their design objectives. The negative outcome means studies that were totally unsuccessful at achieving their design objectives. Out of the eight studies that employed a qualitative methodology, five studies (63%) were fully successful, two studies (25%) were partially successful, and only one study (12%) did not specify its outcomes. All five studies that employed only a quantitative methodology to evaluate their PT designs reported fully successful outcomes. 10 Fig. 6. Comparative Effectiveness of PTs Based on Evaluation Methods. 4.5 Target Audience of Personalized PTs The number of participants in the evaluation of the PT interventions for promoting PA and discouraging SB varies significantly across the reviewed studies. The number of participants ranged from 4 to 129,010. For the reviewed studies that had multiple phases, we combined the number of participants from all phases. As represented in Figure 7, sixteen studies (42%) were targeted at adults (31–54 years old), while six studies (16%) were targeted at young adults (18–30 years old). Children and elderly people were targeted in five studies (13%) for each. Four studies (11%) targeted teenagers, and only six studies (16%) did not specify their target population. Fig. 7. Target Audience of Personalized PTs. 11 4.6 Effectiveness of Personalized PTs Across Various Age Groups Figure 8 shows the effectiveness of personalized PTs based on the population age de- mographic. Out of the 16 studies targeted at adults, six studies (38%) reported fully successful outcomes, seven studies (44%) reported partially successful outcomes, two studies (12%) did not evaluate their PT designs, and only one study (6%) did not specify its outcomes. All six studies that targeted at young adults reported fully successful outcomes. For children, out of the five studies targeted at them, three studies (60%) showed fully suc- cessful outcomes, one study (20%) was partially successful, and one study (20%) did not evaluate its PT design. Out of the five studies targeted at elderly people, four studies (80%) were fully successful, and only one study (20%) did not provide an evaluation for its design. Three studies (75%) targeted at teenagers reported fully successful out- comes, and only one study (25%) targeted at teenagers was partially successful. Out of the six studies that did not specify their target population, two studies (33%) were fully successful, and four studies (67%) had no evaluation. Fig. 2. Effectiveness of Personalized PTs Across Various Age Groups. 4.7 Personalization Implementations in PT for PA and SB Most reviewed PTs implemented a personalization strategy as personalized feedback and personalized recommendations. With respect to feedback, the personalization strat- egy was mostly employed as personalized educational or personalized motivational content. However, for the recommendations, personalization was employed as person- alized activity recommendations, personalized goal recommendations, or personalized fitness partner recommendations in line with Ghanvatkar et al. [19]. According to Ghanvatkar et al. [19], there are six types of a personalization strategy, as shown in Table 1. Thus, we adapted their categorizations of personalization and used it 12 in analyzing the personalization approaches employed in the reviewed studies. The comprehensive details about each type/category of personalization and their definitions can be found in Ghanvatkar et al. [19]. Figure 9 shows that out of the 38 reviewed studies, twenty-three studies (61%) imple- mented personalization strategy as personalized motivational feedback to users. Nine- teen (50%) of the total studies implemented the personalization strategy as personalized goal recommendations. Personalized activity recommendations and personalized inter- vention timing ranked third with a total of ten studies (26%) for each. Just five studies (13%) implemented a personalization strategy as personalized educational feedback. Intervention timing is a type of personalization that uses the context of the feedback or recommendation to find the right and suitable time to deliver it to the user [19]. Only one study (3%) implemented a personalization strategy as a personalized fitness partner recommendation. Fig.3. Personalization Implementations in Persuasive Technologies for PA and SB. Besides the above personalization approaches, other implementations of a personaliza- tion strategy were identified in the reviewed studies. For instance, Dantzing et al.[10] implemented personalization in the form of context-aware coaching, as users got a daily personalized step goal, and they were coached through a custom-design smartphone application to achieve their targeted goal by receiving personalized messages in real- time. Thus, the work of Dantzing et al.[10] provided an example of three types of per- sonalization: goal recommendations, educational content, and motivational content. Another example implementation of the personalization as goal recommendations and motivational content was found in Bounce [31], a smartphone app for breast cancer 13 survivors that encourages them to engage in more PA. The Bounce app reminds users of their PA goals. It also provides personalized and customizable reinforcements and virtual rewards such as badges and trophies once users achieve progress towards ac- complishing their goals and sends pop-up messages to congratulate users on their PA progress. Schafer et al. [50] delivered personalization as personalized motivational content. They provided personalized gamified feedback through a smartphone application. This feed- back was delivered by providing a simple visualization of the activity level through a personalized animated avatar and represented them based on the gender of children. Francillette et al. [17] showed an example of implementing personalization as person- alized goal recommendations and intervention timing. They designed a smartphone ex- ergame app to motivate individuals with severe mental health conditions to integrate PA into their daily lives. The app enables players to plan and set their PA goals based on the players’ profiles, which allows the system to generate different PA choices with different difficulty levels according to the players’ predetermined choices as personal- ized goal recommendations. These PA choices were delivered to the players at suitable times as personalized intervention timing. The PRO-Fit application [11] showed an example of implementing personalization as personalized fitness partner recommendations, goal recommendations, activity recom- mendations, and intervention timing. PRO-Fit is a personalized smartphone fitness ap- plication that recommends personalized PA sessions based on the user’s calendar and availability, fitness goals, and preferences. Furthermore, the app integrates with the user’s social networks, and, based on them, the app suggests “fitness buddies/partners” that have similar fitness goals, preferences, and availability. Table 3 provides a summary of different ways of implementing a personalization strat- egy in our reviewed studies based on the types of personalization that have been clas- sified, defined, and validated by Ghanvatkar et al. [19] in their scoping review study. Table 3: Implementation of A Personalization Strategy. Personalization Strategy Implementation Categories Personalized Goal Recom- Sending personalized messages and notifications to re- mendations mind users about their target quantified goals (e.g. per- sonalized context-aware coaching, generating different PA choices and sessions based on the users’ goals and preferences). For example of implementations, see [31], [10], [11], and [17], etc. Personalized Activity Rec- Sending personalized suggestions of suitable type of ommendations physical activities to users (e.g. biking, running, aero- bic, yoga, cycling, stretching, walking). For example, see [11], and [23], etc. 14 Personalized Fitness Partner Matching users of a PT system to a similar partner who Recommendations have the same target goals to increase their motivations (e.g. suggesting “fitness buddies/partners” that have similar fitness goals, preferences, and availability). For example, see [11]. Personalized Motivational Sending personalized feedback or messages that aim to Content motivate users to change their behavior of engaging in more PA (e.g. context-aware coaching that encourages users to continue maintaining a good levels of practic- ing PA, personalized gamified and visual feedback of the user’s activity level through a personalized ani- mated avatar). For example, see [10], [31], and [50], etc. Personalized Educational A personalized feedback that targeted at increasing the Content awareness and knowledge of the users by providing them with different tips and instructions (e.g. context- aware coaching that educate users in how to specific type of PA). For example, see [10], and [14], etc. Personalized Intervention Intervention timing is a type of personalization that Timing uses the context of the feedback or recommendation and finds the right and suitable time to deliver it to the user (e.g. recommending PA choices PA to be deliv- ered to the players at their suitable times and availabil- ity). For instance, see [11], and [17], etc. 4.8 Effectiveness of Personalization in Physical Activity and Sedentary Behavior Domains Figure 10 summarizes the effectiveness of the reviewed studies, all of which employed some form of personalization in their persuasive intervention design. Out of the total 38 reviewed studies, twenty studies (52%) reported fully successful outcomes, nine studies (24%) were partially successful, eight studies (21%) did not evaluate their PT designs, and only one study (3%) did not specify its evaluation outcomes. Overall, forty-seven (76%) of the total reviewed studies reported successful outcomes, whether fully or partially successful. Figure 11 shows the effectiveness of various implementations of a personalization strat- egy. Out of the 23 studies that employed motivational feedback, 16 studies (69%) re- ported fully successful outcomes, five studies (22%) reported partially successful out- comes, and only two studies (9%) had no evaluation of their PT designs. Out of the 19 studies that implemented a personalization strategy as goal recommendations, eight studies (42%) were fully successful, four studies (21%) were partially successful, six studies (32%) did not evaluate their system, and only one study (5%) did not specify its outcomes. 15 Out of the ten studies that applied activity recommendations, six studies (60%) showed fully successful outcomes, two studies (20%) reported partially successful outcomes, and two studies (20%) had no evaluation. Of the ten studies that employed the personalized intervention timing as an implemen- tation of persuasive strategy, five studies (50%) reported fully successful outcomes, three studies (30%) were partially successful, and just two studies (20%) were not eval- uated. Finally, out of the five studies that implemented educational feedback as a form of personalization, three studies (60%) were fully successful, one study (20%) was par- tially successful, and one study (20%) had no evaluation. Only one study employed a fitness partner recommendation as a form of personalization, and this study did not evaluate its PT design. Fig.4. Overall Effectiveness of Personalized PT for PA and SB. 16 Fig. 5. Effectiveness of Personalized PT Based on How a Personalization Strategy was Imple- mented. 5 Discussion We found that many of the reviewed studies (45%) employed a mixed-methods evalu- ation, which is a combination of qualitative and quantitative evaluation methodologies. We highly recommend researchers to apply the mixed evaluation methodology to fully reveal the comprehensive impact of their PT design and its effectiveness rather than employing either the qualitative method alone or a quantitative method alone. The majority of the reviewed studies targeted adults and young adults with a total of 22 studies (58%). Thus, we recommend that researchers design more PT for promoting PA and discouraging SB targeting other populations such as the elderly, teenagers, and children. Our findings showed that the studies targeted at adults and young adults re- ported the largest number of successful outcomes, whether fully or partially successful. This is perhaps because adults and young adults are at an active age, and they can prac- tice moderate-intensity PA easily more than older populations. Similarly, adults and young adults are more aware of the benefits of PA and the consequences of a sedentary lifestyle than teenagers and children [1], [2]. Based on the reviewed studies, we found that personalizing PT for promoting PA and discourage SB among users is an effective and promising approach of increasing the effectiveness of PTs, with a total of 29 (76%) of the entire reviewed studies (38) report- ing successful outcomes, both fully and partially successful outcomes. 17 We included studies that implemented any form of personalization in the PT design for PA and SB domains. The results of our analysis revealed that personalization was de- livered and implemented differently from one study to another based on the user’s pro- file, preferences, predetermined goals, and availability. We found that the most com- mon and effective way of implementing a personalization strategy based on the re- viewed studies emerged to be personalized motivational content. We believe this is because motivation is one of the most crucial factors for persuading users to achieve a particular objective [37], such as, in our case, increasing PA levels and reducing SB. Thus, personalized feedback that aims to motivate users to change their behavior of engaging in more PA and reducing SB plays an essential role in improving users’ mo- tivation and commitment. The second most frequently employed and effective type of personalization implemen- tation is the goal recommendations. Personalized goal recommendations were delivered to users based on the PA goals they set on the PT system. It is important to mention that goal recommendations describe quantified goals such as step counts, calories burned, and duration of activity [19]. Users appreciated systems that reminded them of their personal goals and objectives. Based on our findings, personalized activity recommendations and intervention timing ranked as the third most commonly employed and effective implementation of person- alization in PTs for promoting PA and decreasing SB. Activity recommendation entails prescribing different types of activities for achieving a set goal such as running, walk- ing, or cycling. Intervention timing focuses on sending feedback or recommendations to users at a suitable and available time for them because users may ignore or forget feedback sent when they are busy with other primary tasks that take greater priority [19]. Although educational content and fitness partner recommendations seem to be the least commonly employed types of personalization implementation, they are considered ef- fective tools to increase the awareness of the users about the benefits of PA for health and general well-being and the consequences of SB and to provide external motivations through matching users with partners who have similar goals and objectives to assist them in maintaining their PA level [19]. Therefore, we recommend researchers to em- ploy a personalization strategy and deliver it to users using different types of imple- mentation based on the users’ needs and combine it with other complementary persua- sive strategies to achieve a desired behavior change objective. Since most of the reviewed studies employ more than one persuasive strategy, this makes it challenging to know which of the employed persuasive strategies contributed to the effectiveness of a PT and lead to the observed behavior change, such as increasing PA levels and reducing SB. Therefore, we cannot attribute the observed results on the effectiveness of the PTs to their use of a personalization strategy only. 18 6 Conclusion This paper presented some interesting trends on various implementations of a person- alization strategy and provided a general overview of personalized PT interventions for promoting PA and discouraging SB. We uncovered various ways of implementing per- sonalization in different PTs and compared their effectiveness, including personalized motivational and educational content feedback, personalized goal and activity recom- mendations, personalized intervention timing, and personalized fitness partner recom- mendations. Finally, our findings revealed that PT interventions for promoting PA and decreasing SB are effective and promising when they are personalized. REFERENCES 1. Noora Aldenaini, Felwah Alqahtani, Rita Orji, and Sampalli Srinivas. 2020. Trends in Persuasive Technologies for Physical Activity and Sedentary Behavior: A Systematic Review. Frontiers in Artificial Intelligence Journal for Human Learning and Behavior Change, 21 Febraury 2020: 85. 2. Najla Almutari and Rita Orji. 2019. How Effective Are Social Influence Strategies in Persuasive Apps for Promoting Physical Activity?: A Systematic Review. 167–172. https://doi.org/10.1145/3314183.3323855 3. Felwah Alqahtani, Ghazayil Al Khalifah, Oladapo Oyebode, and Rita Orji. 2019. Apps for Mental Health: An Evaluation of Behavior Change Strategies and Recommendations for Future Development. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2019.00030 4. Maximilian Altmeyer, Pascal Lessel, Tobias Sander, and Antonio Krüger. 2018. Extending a Gamified Mobile App with a Public Display to Encourage Walking. https://doi.org/10.1145/3275116.3275135 5. Sonia M. Arteaga, Mo Kudeki, Adrienne Woodworth, and Sri Kurniawan. 2010. Mobile system to motivate teenagers’ physical activity. 1. https://doi.org/10.1145/1810543.1810545 6. Dale S. Bond, J. Graham Thomas, Hollie A. Raynor, Jon Moon, Jared Sieling, Jennifer Trautvetter, Tiffany Leblond, and Rena R. Wing. 2014. B-MOBILE - A smartphone- based intervention to reduce sedentary time in overweight/obese individuals: A within- subjects experimental trial. PLoS ONE 9, 6. https://doi.org/10.1371/journal.pone.0100821 7. Scott A Cambo, Daniel Avrahami, and Matthew L Lee. 2017. BreakSense: Combining Physiological and Location Sensing to Promote Mobility during Work-Breaks. Proceedings of the 2017 Acm Sigchi Conference on Human Factors in Computing Systems (Chi’17). https://doi.org/10.1016/j.engappai.2016.10.012 8. Luca Chittaro and Riccardo Sioni. 2012. Turning the classic snake mobile game into a location-based exergame that encourages walking. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 7284 LNCS: 43–54. https://doi.org/10.1007/978-3-642-31037-9_4 9. Fabio Ciravegna, Jie Gao, Neil Ireson, Robert Copeland, Joe Walsh, and Vitaveska Lanfranchi. 2019. Active 10- Brisk Walking to Support Regular Physical Activity. Proceedings Paper in PervasiveHealth2019. https://doi.org/10.1145/3329189.3329208 10. Saskia van Dantzig, Murtaza Bulut, Martijn Krans, Anouk van der Lans, and Boris de Ruyter. 2018. Enhancing physical activity through context-aware coaching. https://doi.org/10.1145/3240925.3240928 11. Saumil Dharia, Vijesh Jain, Jvalant Patel, Jainikkumar Vora, Rizen Yamauchi, 19 Magdalini Eirinaki, and Iraklis Varlamis. 2016. PRO-Fit: Exercise with friends. Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2016: 1430–1433. https://doi.org/10.1109/ASONAM.2016.7752437 12. David W. Dunstan, Alicia A. Thorp, and Genevieve N. Healy. 2011. Prolonged sitting: Is it a distinct coronary heart disease risk factor? Current Opinion in Cardiology. https://doi.org/10.1097/HCO.0b013e3283496605 13. Daphne Economou, Miriam Dwek, Claire Roberston, Bradley Elliott, Thanos Kounenis, Tayebeh Azimi, Mohammad Ramezanian, and Nathan Bell. 2017. PhytoCloud: A Gamified Mobile Web Application to Modulate Diet and Physical Activity of Women with Breast Cancer. In Proceedings - IEEE Symposium on Computer-Based Medical Systems. https://doi.org/10.1109/CBMS.2017.164 14. Muhammad Fahim, Thar Baker, Asad Masood Khattak, and Omar Alfandi. 2017. Alert me: Enhancing active lifestyle via observing sedentary behavior using mobile sensing systems. 2017 IEEE 19th International Conference on e-Health Networking, Applications and Services, Healthcom 2017 2017-Decem: 1–4. https://doi.org/10.1109/HealthCom.2017.8210838 15. B J Fogg. 1998. Persuasive Computers: Perspectives and Research Directions. Retrieved March 14, 2019 from www.captology.org 16. Derek Foster, Conor Linehan, Ben Kirman, Shaun Lawson, and Gary James. 2010. Motivating physical activity at work: Using persuasive social media for competitive step counting. Proceedings of the 14th International Academic MindTrek Conference on Envisioning Future Media Environments - MindTrek ’10: 111. https://doi.org/10.1145/1930488.1930510 17. Yannick Francillette, Bruno Bouchard, Eric Boucher, Sébastien Gaboury, Paquito Bernard, Ahmed Jérome Romain, and Kévin Bouchard. 2018. Development of an Exergame on Mobile Phones to Increase Physical Activity for Adults with Severe Mental Illness. https://doi.org/10.1145/3197768.3201521 18. Eva Geurts, Fanny Van Geel, Peter Feys, and Karin Coninx. 2019. WalkWithMe- Personalized Goal Setting and Coaching for Walking in People with Multiple Sclerosis. In Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization - UMAP ’19, 51–60. https://doi.org/10.1145/3320435.3320459 19. Suparna Ghanvatkar, Atreyi Kankanhalli, and Vaibhav Rajan. 2019. User models for personalized physical activity interventions: Scoping review. Journal of Medical Internet Research. https://doi.org/10.2196/11098 20. Liam G. Glynn, Patrick S. Hayes, Monica Casey, Fergus Glynn, Alberto Alvarez- Iglesias, John Newell, Gearóid ÓLaighin, David Heaney, and Andrew W. Murphy. 2013. SMART MOVE - a smartphone-based intervention to promote physical activity in primary care: Study protocol for a randomized controlled trial. Trials. https://doi.org/10.1186/1745-6215-14-157 21. Md Sanaul Haque, Wali Mohammad Abdullah, Sadiqur Rahaman, Maarit Kangas, and Timo Jämsä. 2016. Persuasive health and wellbeing application: A theory-driven design in promoting physical activity. 1st International Conference on Medical Engineering, Health Informatics and Technology, MediTec 2016. https://doi.org/10.1109/MEDITEC.2016.7835369 22. Oinas-Kukkonen Harri and Harjumaa Marja. 2009. Persuasive Systems Design: Key Issues, Process Model, and System Features. Communications of the Association for Information Systems 24, 1: 96. https://doi.org/10.17705/1CAIS.02428 23. Qian He and Emmanuel Agu. 2014. On11: An Activity Recommendation Application to Mitigate Sedentary Lifestyle. Proceedings of the 2014 Workshop on Physical Analytics: 3–8. https://doi.org/10.1145/2611264.2611268 24. Yan Hong, Deborah Vollmer Dahlke, Marcia Ory, Angela Hochhalter, Jana Reynolds, 20 Ninfa Pena Purcell, Divya Talwar, and Nola Eugene. 2013. Designing icanfit: A mobile- enabled web application to promote physical activity for older cancer survivors. Journal of Medical Internet Research. https://doi.org/10.2196/resprot.2440 25. M. Jetté, K. Sidney, and G. Blümchen. 1990. Metabolic equivalents (METS) in exercise testing, exercise prescription, and evaluation of functional capacity. Clinical Cardiology 13, 8: 555–565. https://doi.org/10.1002/clc.4960130809 26. Michel C.A. Klein, Adnan Manzoor, and Julia S. Mollee. 2017. Active2Gether: A personalized m-health intervention to encourage physical activity. Sensors (Switzerland) 17, 6: 1–16. https://doi.org/10.3390/s17061436 27. Nicholas D. Lane, Mu Lin, Mashfiqui Mohammod, Xiaochao Yang, Hong Lu, Giuseppe Cardone, Shahid Ali, Afsaneh Doryab, Ethan Berke, Andrew T. Campbell, and Tanzeem Choudhury. 2014. BeWell: Sensing sleep, physical activities and social interactions to promote wellbeing. Mobile Networks and Applications. https://doi.org/10.1007/s11036-013-0484-5 28. Daehyoung Lee, Georgia Frey, Alison Cheng, and Patrick C. Shih. 2018. Puzzle walk: A gamified mobile app to increase physical activity in adults with autism spectrum disorder. In 2018 10th International Conference on Virtual Worlds and Games for Serious Applications, VS-Games 2018 - Proceedings. https://doi.org/10.1109/VS- Games.2018.8493439 29. Yuzhong Lin, Joran Jessurun, Bauke de Vries, and Harry Timmermans. 2011. Motivate: towards Context-Aware Recommendation Mobile System for Healthy Living. 250–253. https://doi.org/10.4108/icst.pervasivehealth.2011.246030 30. Chiraphruet Mansart, Siriluck Sukitphittayanon, Panitan Pantongkhum, and Supphachai Thaicharoen. 2015. Go Run Go: An Android Game-Story Application for Aiding Motivation to Exercise. Proceedings - 2015 IEEE International Symposium on Multimedia, ISM 2015: 407–410. https://doi.org/10.1109/ISM.2015.49 31. Gabriela Marcu, Anjali Misra, Karina Caro, Meghan Plank, Amy Leader, and Andrea Barsevick. 2018. Bounce: Designing a Physical Activity Intervention for Breast Cancer Survivors. https://doi.org/10.1145/nnnnnnn.nnnnnnn 32. Corby K Martin, L Anne Gilmore, John W Apolzan, Candice A Myers, Diana M Thomas, and Leanne M Redman. 2016. Smartloss: A Personalized Mobile Health Intervention for Weight Management and Health Promotion. JMIR mHealth and uHealth. https://doi.org/10.2196/mhealth.5027 33. Siobhan McMahon, Mithra Vankipuram, and Julie Fleury. 2013. Mobile Computer Application for Promoting Physical Activity. Journal of Gerontological Nursing. https://doi.org/10.3928/00989134-20130226-01 34. Hazwani Mohd Mohadis and Nazlena Mohamad Ali. 2016. Designing persuasive application to encourage physical activity at workplace among older workers. In 2016 6th International Conference on Digital Information and Communication Technology and Its Applications, DICTAP 2016. https://doi.org/10.1109/DICTAP.2016.7544013 35. Adity Mutsuddi and Kay Connelly. 2012. Text Messages for Encouraging Physical Activity Are they effective after the novelty effect wears off? 2012 6th International Conference on Pervasive Computing Technologies for Healthcare (PervasiveHealth) and Workshops: 33–40. https://doi.org/10.4108/icst.pervasivehealth.2012.248715 36. T. Oliveira, D. Leite, and G. Marreiros. 2016. PersonalFit - Fitness App with intelligent plan generator. In ACM International Conference Proceeding Series. https://doi.org/10.1145/2948992.2949014 37. Rita Orji. 2017. Why are persuasive strategies effective? Exploring the strengths and weaknesssses of socially-oriented persuasive strategies. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). https://doi.org/10.1007/978-3-319-55134-0_20 38. Rita Orji and Regan L. Mandryk. 2014. Developing culturally relevant design guidelines 21 for encouraging healthy eating behavior. International Journal of Human Computer Studies. https://doi.org/10.1016/j.ijhcs.2013.08.012 39. Rita Orji, Regan L. Mandryk, and Julita Vassileva. 2017. Improving the efficacy of games for change using personalization models. ACM Transactions on Computer- Human Interaction. https://doi.org/10.1145/3119929 40. Rita Orji, Regan L. Mandryk, Julita Vassileva, and Kathrin M. Gerling. 2013. Tailoring persuasive health games to gamer type. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems - CHI ’13. https://doi.org/10.1145/2470654.2481341 41. Rita Orji and Karyn Moffatt. 2018. Persuasive technology for health and wellness: State- of-the-art and emerging trends. Health Informatics Journal 24, 1: 66–91. https://doi.org/10.1177/1460458216650979 42. Rita Orji, Lennart E. Nacke, and Chrysanne Di Marco. 2017. Towards personality- driven persuasive health games and gamified systems. In Conference on Human Factors in Computing Systems - Proceedings. https://doi.org/10.1145/3025453.3025577 43. Rita Orji, Gustavo F. Tondello, and Lennart E. Nacke. 2018. Personalizing persuasive strategies in gameful systems to gamification user types. In Conference on Human Factors in Computing Systems - Proceedings. https://doi.org/10.1145/3173574.3174009 44. N. Owen, A. Bauman, and W. Brown. 2009. Too much sitting: A novel and important predictor of chronic disease risk? British Journal of Sports Medicine. https://doi.org/10.1136/bjsm.2008.055269 45. Kiemute Oyibo, Abdul-Hammid Olagunju, Babatunde Olabenjo, Ifeoma Adaji, Ralph Deters, and Julita Vassileva. 2019. BEN’FIT: Design, Implementation and Evaluation of a Culture-Tailored Fitness App. https://doi.org/10.1145/3314183.3323854 46. Kiemute Oyibo, Rita Orji, and Julita Vassileva. 2017. The influence of culture in the effect of age and gender on social influence in persuasive technology. In UMAP 2017 - Adjunct Publication of the 25th Conference on User Modeling, Adaptation and Personalization. https://doi.org/10.1145/3099023.3099071 47. Nithya Ramanathan, Faisal Alquaddoomi, Hossein Falaki, Dony George, Cheng-Kang Hsieh, John Jenkins, Cameron Ketcham, Brent Longstaff, Jeroen Ooms, Joshua Selsky, Hongsuda Tangmunarunkit, and Deborah Estrin. 2012. ohmage: An open Mobile System for Activity and Experience Sampling. https://doi.org/10.4108/icst.pervasivehealth.2012.248705 48. Daniel Riffe, Stephen Lacy, and Frederick G. Fico. 1997. Analyzing Media Messages: Using Quantitative Content Analysis in Research. https://doi.org/10.2307/358412 49. Ankita Samariya, Anud Sharma, Margiawan Fitriani, Tucker Ferguson, Jerry Alan Fails, and Jerry Alan. 2019. KidLED: A colorful approach to children’s activity awareness. https://doi.org/10.1145/3311927.3326594 50. Hanna Schäfer, Joachim Bachner, Sebastian Pretscher, Georg Groh, and Yolanda Demetriou. 2018. Study on Motivating Physical Activity in Children with Personalized Gamified Feedback. https://doi.org/10.1145/3213586.3225227 51. Sarah Justine Guy Skriloff, Dario C Gonzalez, Kurtis C Christensen, Logan J Bentley, and Cody V Mortensen. 2016. FitPlay Games: Increasing Exercise Motivation Through Asynchronous Social Gaming. Proceedings of the 2016 CHI Conference Extended Abstracts on Human Factors in Computing Systems. https://doi.org/10.1145/2851581.2890367 52. M Sohn and Jeunwoo Lee. 2007. UP health: Ubiquitously Persuasive Health Promotion with an Instant Messaging System. 25th SIGCHI Conference on Human Factors in Computing Systems 2007, CHI 2007: 2663–2668. https://doi.org/10.1145/1240866.1241059 53. P. Spiesberger, F. Jungwirth, C. Wöss, S. Bachl, J. Harms, and T. Grechenig. 2015. 22 Woody: A location-based smartphone game to increase children’s outdoor activities in urban environments. In ACM International Conference Proceeding Series. https://doi.org/10.1145/2836041.2841210 54. K G Stanley, I Livingston, A Bandurka, R Kapiszka, and R L Mandryk. 2010. PiNiZoRo: A GPS-based exercise game for families. International Academic Conference on the Future of Game Design and Technology, Future Play 2010: 243– 246. https://doi.org/10.1145/1920778.1920817 55. Tammy Toscos, Anne Faber, Shunying An, and Mona Praful Gandhi. 2006. Chick Clique- Persuasive Technology to Motivate Teenage Girls to Exercise. https://doi.org/10.1145/1125451.1125805 56. Tammy Toscos, Anne Faber, Kay Connelly, and Adity Mutsuddi Upoma. 2008. Encouraging physical activity in teens can technology help reduce barriers to physical activity in adolescent girls? In Proceedings of the 2nd International Conference on Pervasive Computing Technologies for Healthcare 2008, PervasiveHealth. https://doi.org/10.1109/PCTHEALTH.2008.4571073 57. Mark S. Tremblay, Salomé Aubert, Joel D. Barnes, Travis J. Saunders, Valerie Carson, Amy E. Latimer-Cheung, Sebastien F.M. Chastin, Teatske M. Altenburg, and Mai J.M. Chinapaw. 2017. Sedentary Behavior Research Network (SBRN) – Terminology Consensus Project process and outcome. International Journal of Behavioral Nutrition and Physical Activity 14, 1: 75. https://doi.org/10.1186/s12966-017-0525-8 58. Yunlong Wang, Lingdan Wu, Jan Philipp Lange, Ahmed Fadhil, and Harald Reiterer. 2018. Persuasive technology in reducing prolonged sedentary behavior at work: A systematic review. Smart Health. https://doi.org/10.1016/j.smhl.2018.05.002 59. Jingwen Zhang and John B. Jemmott III. 2019. Mobile App-Based Small-Group Physical Activity Intervention for Young African American Women: a Pilot Randomized Controlled Trial. Prevention Science: 1–10. https://doi.org/10.1007/s11121-019-01006-4 60. Physical activity. Retrieved February 7, 2020 from https://www.who.int/health- topics/physical-activity APPENDIX 1 Appendix 1. A Comprehensive Overview of PT for Physical Activity and/or Sedentary Behavior # Authors Do- Technol- Applica- Persuasive Strategies Theo- Evalua- Types of Per- Duration Targeted Audience No. of Results Country of Arti- main ogy tion / Pro- /Affordances ries tion sonalization Outcomes Age Partici- of Study cles, ject Name Method (Ways of imple- Group pants Year, mentation) Refer- ence 1 Foster et PA Smartphon StepMatron Tracking, Personalization, none Mixed motivational 3 Weeks Behavior Adults 10 Fully Suc- UK al.(2010), e mobile, Goal Setting, Self-Moni- content and goal cessful [16] Computer, toring, Social Support (So- recommendation Pedometer cial Learning, Compari- sons, Competition, Rank- ings Recognition, Giving Comments) 2 He and SB Smartphon On11 Tracking, Reduction, Tun- none Qualita- activity recom- 2 Weeks Behavior, Adults 8 Partially USA Agu e mobile neling, Tailoring, Person- tive mendations, and Awareness Successful (2014), alization, Goal Setting, intervention tim- [23] Self-Monitoring (Self-Re- ing flection), Reminder, Sug- gestion, Liking 3 Fahim et SB Smartphon Alert Me Tracking, Personalization, none Quantita- educational con- Unspeci- Behavior, Unspeci- 0 Fully Suc- Russia al.(2017) e mobile Self-Monitoring, Re- tive tent, activity fied Awareness fied cessful , [14] minder recommenda- tion, interven- tion timing 4 Mohadis PA Smartphon WargaFit Tracking, Reduction, Tun- none Mixed goal recommen- Unspeci- Behavior Elderly 8 Fully Suc- Malaysia and Ali e mobile neling, Tailoring, Person- dations, and ac- fied cessful (2016), alization, Self-Monitoring, tivity recom- [34] Simulation, Rehearsal, mendation Praise, Reminders, Sug- gestions, Similarity, Ex- pertise, Real world feel, Third-Party Endorsement, Verifiability, Social Sup- port (Social Learning, So- cial Comparison, Norma- tive influence, Social fa- cilitation, Competition, Recognition) Persuasive 2020, Adjunct proceedings of the 15th Int. conference on Persuasive Technology. Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). 2 Appendix 1. (continued) # Au- Do- Technology Applica- Persuasive Strategies /Af- Theo- Evalua- Types of Per- Duration Targeted Audi- No. of Results Country thors of main tion / fordances ries tion sonalization Outcomes ence Partici- of Study Arti- Project Method (Ways of imple- Age pants cles, Name mentations) Group Year, Refer- ence 5 Cambo SB Smartphone BreakSe Tracking, Personalization, none Mixed intervention tim- 8 Days Behavior Adults 6 Partially USA et mobile, nse Self-Monitoring, Rewards, ing, and activity Successful al.(2017 Smartwatch Reminder recommendation ), [7] 6 Mansart PA Smartphone- Go Run Tracking, Tunneling, Per- none Mixed motivational Unspecified Behavior Adults 10 Fully Suc- Thailand et and based mobile Go sonalization, Self-Monitor- content cessful al.(2015 SB exergame ing, Rewards, Social Sup- ), [30] port (Sharing) 7 Lin et PA Smartphone Motivate Tracking, Reduction, Per- none Mixed motivational 5 Weeks Behavior, Adults 6 Fully Suc- Nether- al.(2011 mobile appli- sonalization, Feedback content, inter- Awareness cessful lands ), [29] cation, web from users (Self-Report), vention timing, application Reminder, Suggestion activity recom- mendations 8 Dharia PA Smartphone PRO-Fit Tracking, Reduction, Per- none none goal recommen- none Behavior Un- 0 none USA et mobile appli- sonalization, Self-Monitor- dations, activity speci- al.(2016 cations ing, Reminder, Suggestion, recommenda- fied ), [11] Authorization, Social Sup- tions, fitness port partner recom- mendations, in- tervention tim- ing 9 Klein et PA Smartphone Ac- Tracking, Reduction, Tai- MBR, Qualita- motivational 3 Months Behavior, Young 100 Fully Suc- Nether- al.(2017 mobile appli- tive2Get loring, Personalization, TTM, tive content, inter- Awareness Adults cessful lands ), [26] cation, Web her Goal Setting, Self-Monitor- DCM, vention timing, page, Face- ing, Simulation, Reminder, SCT, activity recom- book, Weara- Suggestion, Liking, Social SRT, mendations ble activity Role, Surface Credibility, HAPA tracker (Fit- Social Support (Social bit) Comparison) 3 Appendix 1. (continued) # Authors Do- Technol- Application / Persuasive Theories Evaluation Types of Per- Duration Targeted Audience No. of Results Country of Arti- main ogy Project Strategies /Af- Method sonalization Outcomes Age Group Partici- of Study cles, Name fordances (Ways of imple- pants Year, mentations) Refer- ence 10 Arteaga PA Smartph Mobile App Tracking, Re- TPB, Qualitative activity recom- 1 Month Behavior Teenagers 5 Fully Suc- USA et one mo- duction, Per- TMB, mendations, mo- Motiva- cessful al.(2010) bile sonalization, PRT tivational con- tion , [5] game ap- Self-Monitor- tent plication ing, Reward, Social Support (Competition) 11 Luca PA Smartph LocoSnake Tracking, Re- none Mixed motivational 5 Minutes Attitude Young 15 Fully Suc- Italy Chittaro one mo- game duction, Per- content Adults cessful and Ric- bile ex- sonalization, cardo ergame Self-monitor- Sioni ing, Simula- (2012), tion, Rewards [8] 12 Haque et PA Mobile iGO Personaliza- SDT Qualitative motivational 1 Week Behavior Young 26 Fully Suc- Finland al.(2016) to web tion, Self-Mon- content Adults cessful , [21] applica- itoring, Re- tion (An- wards, Re- droid) minder, Social Support (Com- petition, Recognition) 13 Bond et SB Smartph B-MOBILE Tracking, Per- none Quantita- goal recommen- 16 Behavior, dults 30 Fully Suc- USA al.(2014) one ap- sonalization, tive dations, inter- Months Motiva- cessful , [6] plication, Goal Setting, vention timing tion Weara- Self-Monitor- ble sen- ing, Praise, Re- sor wards, Re- minder 4 Appendix 1. (continued) # Au- Do- Technol- Application Persuasive Strategies Theories Evalua- Types of Per- Dura- Targeted Audience No. of Results Coun- thors of main ogy / Project /Affordances tion sonalization tion Outcomes Age Group Partici- try of Arti- Name Method (Ways of imple- pants Study cles, mentations) Year, Refer- ence 14 Skriloff PA Smartpho FitPlay Tracking, Personaliza- none none intervention tim- none Behavior, Unspecified 0 none USA et ne mobile Games plat- tion, Social Support (Co- ing, and activity Motivation al.(2016 applica- form operation, Competition) recommendation ), [51] tion (An- droid), Wearable activity tracker (Fitbit) 15 McMah PA Smartpho Ready~Stea Reduction, Personaliza- WMT, none goal recommen- none Behavior Elderly 0 none USA on et ne mobile dy tion, Goal Setting, Self- USS, dations al.(2013 applica- Monitoring, Simulation, TDP ), [33] tion Praise, Rewards, Social Role 16 Rama- PA and Mobile to ohmage Tracking, Personaliza- none none goal recommen- none Behavior Unspecified 0 none USA nathan Experi- web plat- tion, Self-Monitoring, dations, motiva- et ence form Feedback from users tional content al.(2012 Sam- (Self-Report), Praise, ), [47] pling Surface Credibility, So- cial Support 17 Stanley PA and Smartpho PiNiZoRo Tracking, Reduction, none Qualita- activity recom- Unspeci- Behavior, Children 4 Fully Canada et Obesity ne mobile Personalization, Simula- tive mendations fied Awareness Suc- al.(2010 game ap- tion cessful ), [54] plication 5 18 Toscos PA Mobile Mobile App Tracking, Personaliza- none Mixed motivational 3 Weeks Behavior Teenagers 8 Par- USA et phone ap- tion, Goal Setting, Self- content tially al.(2008 plication, Monitoring, Praise, Re- Suc- ), [56] Pedometer minder, Social Support cessful (Comparison, Competi- tion, Sharing) Appendix 1. (continued) # Au- Do- Technology Applica- Persuasive Strategies /Af- Theo- Evaluation Types of Person- Dura- Targeted Audience No. Results Coun- thors of main tion / Pro- fordances ries Method alization (Ways tion Outcomes Age of try of Arti- ject Name of implementa- Group Par- Study cles, tion) tici- Year, pant Refer- s ence 19 Mutsud PA Mobile text Mobile Tailoring, Personalization, Goal TTM Mixed goal recommen- 3 Behavior Young 30 Fully USA di and messaging phone text Setting, Self-Monitoring, Praise, dations, motiva- Months Adults Success- Con- app, Pedom- messaging Reward, Reminder, Suggestion, tional content ful nelly eter app Social Support (Sharing) (2012), [35] 20 Toscos Eating Cell-phone Chick Tracking, Reduction, Personali- none Mixed motivational con- 6 Days Behavior, Female 10 Fully USA et and PA application, Clique zation, Self-Monitoring, Praise, tent Aware- Teenagers Success- al.(2006 Pedometer Social Support (Cooperation, ness ful ), [55] Competition, Sharing) 21 Sohn PA and PDA text UP Health Tracking, Personalization, Goal none Qualitative goal recommen- 1 week Behavior, Adults 5 Partially South and Lee Smok- messaging, Setting, Self-Monitoring, Re- dations Aware- Success- Korea (2007), ing Instant Mes- ward or Punishment, Reminder, ness ful [52] saging (IM) Social Support (Cooperation, system, Mo- Competition) bile device 22 Glynn PA smartphone, Accupedo Tracking, Personalization, Goal none Qualitative goal recommen- 2 Behavior Adults 80 Unspeci- Ireland et Pedometer Setting, Self-Monitoring, Social dations Months fied al.(2013 Support (Sharing) ), [20] 6 23 Marcu PA Smartphone Bounce Reduction, Tunneling, Personali- TTM, Qualitative motivational con- 3 Weeks Behavior, Adults 4 Fully USA et al. mobile zation, Goal Setting, Self-Moni- SCT tent , goal recom- Attitude, Success- (2018), toring, Praise, Rewards, Re- mendations Aware- ful [31] minders, Social Role, Trustwor- ness, Mo- thiness, Expertise, Authority, tivation Social Support (Normative In- fluence, Cooperation, Social In- teraction) Appendix 1. (continued) # Au- Domain Technol- Applica- Persuasive Strategies Theories Evaluation Types of Per- Dura- Targeted Audience No. of Results Country thors of ogy tion / Pro- /Affordances Method sonalization tion Outcomes Age Partici- of Study Arti- ject Name (Ways of imple- Group pants cles, mentations) Year, Refer- ence 24 Zhang PA Mobile ap- PennFit Tracking, Personalization, SCT Quantitative motivational 3 Months Behavior, Young 91 Fully USA and plication, Self-Monitoring, Re- content, inter- Aware- Adults Suc- Jemmot Activity minder, Social Support vention timing ness cessful t Tracker (Comparison, Social Inter- (2019), (Fitbit) action ( messages with [59] chatting tool)) 25 Lee et PA and Smartphon Puzzle Tunneling, Personaliza- unspeci- none goal recommen- none Behavior, Adults 34 none USA al. SB e applica- Walk tion, Goal Setting, Self- fied dations Motiva- (2018), tion Monitoring, Praise, Re- tion [28] wards, Reminder, Liking 26 Lane et PA, Smartphon BeWell+ Tracking, Personalization, none Quantitative motivational 19 Days Behavior, Unspeci- 27 Fully UK al.(2014 Sleep, e applica- Self-Monitoring, Simula- content Social In- fied (Gen- Suc- ), [27] Social tion, and tion, Rewards, Liking, So- teraction eral) cessful Interac- ambient cial Support (Social Inter- tion display on action) the smartphone wallpaper 7 27 Hong et PA A mobile to iCanFit Tracking, Reduction, Tun- none Mixed goal recommen- 11 Behavior, Elderly 112 Fully USA al.( web appli- neling, Tailoring, Person- dations, motiva- months Motiva- Suc- 2013), cation alization, Goal Setting, tional content , tion cessful [24] (desktop Self-Monitoring, Sugges- educational con- version as a tion, Praise , Trustworthi- tent web, iPh- ness, Expertise, Surface one ver- Credibility, Real-world sion) Feel, Authority, Third- party Endorsements, Veri- fiability, Social Support ( Normative Influence, So- cial Interaction) Appendix 1. (continued) # Au- Domain Technol- Applica- Persuasive Strate- Theories Evaluation Types of Person- Duration Targeted Audience No. Results Country thors of ogy tion / Pro- gies /Affordances Method alization (Ways Outcomes Age Group of of Study Arti- ject Name of implementa- Par- cles, tions) tici- Year, pant Refer- s ence 28 Francil- PA A An Reduction, Tailoring, none Mixed goal recommen- 30 Behavior, Adults 15 Partially Canada lette et smartphone smartphone Personalization, Goal dations, interven- minutes Motivation Successful al.(2018 exergame exergame Setting, Self-Moni- tion timing ), [17] application app toring, Rewards, Re- minder, Liking 29 Dantzig PA A A digital Tracking, Personali- none Mixed goal recommen- 1 Month Behavior, Adults 70 Partially Nether- et smartphone smartphone zation, Self-monitor- dations, educa- Motivation Successful lands al.(2018 application coaching ing, Praise, Reminder, tional content, ), [10] , wearable system Suggestion motivational con- activity tent tracker de- vice 30 Alt- PA A gamified A gamified Tracking, Personali- SDT Mixed motivational con- 1 Month Behavior, Adults 12 Partially Germany meyer system in- mobile app zation, Self-Monitor- tent Motivation, Successful et cludes fit- ing, Rewards, Re- Usability al.(2018 ness minder, Similarity, ), [4] tracker, Social Support (Com- mobile app, parison, Normative website as a Influence) 8 public dis- play 31 Schafer PA A gamified A gamified Tracking, Personali- none Mixed motivational con- 1 Month Behavior, Children 61 Partially Germany et smartphone smartphone zation, Self-Monitor- tent Awareness, Successful al.(2018 app app ing, Praise, Rewards, Motivation, ), [50] Liking Acceptance, Attitude 32 Cirave- PA Mobile Active 10 Tracking, Reduction, unspeci- Quantita- goal recommen- 1 year and Behavior, Unspeci- 749,0 Fully Suc- UK gna et phone ap- Personalization, Self- fied tive dations and moti- 11 months Adherence fied 10 cessful al.(2019 plication monitoring, Goal-Set- vational content ), [9] ting, Praise, Rewards, Reminders, Expertise, Real-world feel Appendix 1. (continued) # Au- Do- Technology Applica- Persuasive Strategies Theories Evalua- Types of Duration Targeted Audience No. of Results Country thors of main tion / Pro- /Affordances tion Personaliza- Outcomes Age Partici- of Study Arti- ject Name Method tion (Ways Group pants cles, of imple- Year, mentations) Refer- ence 33 Oyibo PA Mobile BEN’FIT Tailoring, Personalization, SCT Mixed goal recom- 1 Months Behavior, Adults 120 Partially Canada, et phone appli- Goal-Setting, Self-Moni- mendations, Motivation Success- USA, al.(2019 cation toring, Rewards, Social motivational ful and Ni- ), [45] Support (Social Learning, content geria Social Comparison, Coop- eration) 34 Samar- PA Mobile ap- KidLED Tracking, Personalization, none none goal recom- none Motivation, Children none none USA iya et plication, mobile ap- Goal-Setting, Self-Moni- mendations Awareness al.(2019 Wearable plication toring, Social Support (So- ), [49] LED Color cial Learning, Social Com- Light Dis- parison) play, Activ- ity Tracker 9 35 Oliveira PA Mobile ap- PersonalFit Tracking, Reduction, Per- none none goal recom- none Self-man- Unspeci- none none Portugal et plication sonalization, Social Role mendations agement fied al.(2016 ), [36] 36 Econo- PA and Gamified PhytoCloud Tracking, Tailoring, Per- none none educational none Behavior Adults none none UK mou et Eating Mobile Web sonalization, Goal-Setting, content , mo- al.(2017 (Diet) App Self-Monitoring, Sugges- tivational ), [13] tions, Trustworthiness, Ex- content pertise, Surface Credibil- ity, Authority, Third-Party Endorsement, Social Sup- port (Social Learning, Nor- mative Influence, Recogni- tion (Ranking), Sharing ) Appendix 1. (continued) # Au- Domain Technol- Application Persuasive Theories Evaluation Types of Person- Duration Targeted Audience No. of Results Country thors of ogy / Project Strategies Method alization (Ways Outcomes Age Partici- of Study Arti- Name /Affordances of implementa- Group pants cles, tion) Year, Refer- ence 37 Geurts PA Mobile Walk- Tracking, GST Mixed goal recommen- 10 Weeks Behavior, Elderly 13 Fully Suc- Belgium et applica- WithMe Tunneling, dations, educa- Motivation cessful al.(2019 tion Tailoring, tional content, ), [18] Personaliza- motivational con- tion, Goal- tent Setting, Self- Monitoring, Praise, Exper- tise, Social Support (Sharing) 10 38 Spies- PA A gami- Woody Tracking, Per- none Mixed motivational con- 12 Days Behavior, Children 38 Fully Suc- Austria berger fied sonalization, tent Awareness, cessful et smartphon Simulation, Motivation al.(2015 e app Reminder, ), [53] Rewards, Lik- ing, Expertise ABBREVIATIONS MBR: Model-Based Reasoning TTM: Transtheoretical Model DCM: Dynamic Computational Model SCT: Social Cognitive Theory SRT: Self-Regulation Theory HAPA: Health Action Process Approach TPB: Theory of Planned Behavior TMB: Theory of Meaning Behavior PRT: Personality Theory SDT: Self-Determination Theory WMT: Wellness Motivation Theory USS: User-Specific Strategies TDP: Theoretical Design Principles GST: Goal-Setting Persuasive 2020, Adjunct proceedings of the 15th Int. conference on Persuasive Technology. Copyright © 2020 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).