Move2Play: An Innovative Approach to Encouraging People to Be More Physically Active Pavol Bielik pbielik@acm.org Michal Tomlein michal.tomlein@gmail.com Peter Krátky kratky.peto@gmail.com Štefan Mitrík stevo.mit@gmail.com Michal Barla barla@fiit.stuba.sk Mária Bieliková bielik@fiit.stuba.sk Institute of Informatics and Software Engineering Faculty of Informatics and Information Technologies Slovak University of Technology in Bratislava Ilkovicova ˇ 3, 842 16 Bratislava, Slovakia ABSTRACT Throughout the last decade, there has been an alarming de- crease in daily physical activity among both children and adults. Medical experts agree that physical activity is crit- ical to maintaining fitness, reducing weight and improving health. Yet so many people have difficulty increasing and maintaining physical activity in everyday life. We have cre- ated a solution called Move2Play, which encourages a health- ier lifestyle and motivates to participate in regular physical activity. We have integrated four essential parts that form the basis for long-term progress and sustainability — Activ- ity Recommendation, Tracking, Evaluation and Motivation. In order to recognize and assess physical activity, we have developed a system for mobile phones that collects data from various sensors, such as accelerometer, GPS and GSM. We provide personalization, i.e. we have proposed and realized activity recommendation for an individual user to ensure that the users engage in regular exercise, as opposed to oc- casional outbursts of activity, which are unhealthy and even harmful. We discuss our proposed mechanisms of activity recommendation and the concept of motivation, which rep- resent a key element for any system that fights the sedentary lifestyle of the modern generation. Categories and Subject Descriptors J.3 [Computer Applications]: Life and Medical Sciences— Health I.2.1 [Artificial Intelligence]: Applications and Expert Systems—Medicine and science H.1.m [Information Systems]: Models and Principles—Miscellaneous General Terms Design, Experimentation Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. IHI’12, January 28–30, 2012, Miami, Florida, USA. Copyright 2012 ACM 978-1-4503-0781-9/12/01 ...$10.00. Keywords Physical Activity, Personalized Activity Recommendation, Activity Tracking, Activity Evaluation, User Model, Health Awareness, Ubiquitous Computing, Pervasive Technology 1. INTRODUCTION Information technology continues its rapid development and brings us new opportunities for solving problems of our society. One of the biggest problems nowadays is the lack of physical activity and the so-called sedentary lifestyle. Mil- lions of people all over the world suffer from illnesses such as cardiovascular disease, high blood pressure or diabetes, where the lack of physical activity is one of the major risk factors. According to the World Health Organization, the main cause of obesity and overweight is an energy imbal- ance between consumed and expended calories, which is also caused by the lack of physical activity and an improper diet. Furthermore, regular physical activity affects our mental health, reduces feelings of anxiety, depression and promotes a sense of well-being. From another point of view, physical inactivity and obe- sity not only lower the quality of life of individuals but also bring a financial burden on public health systems and na- tional governments. Their estimated cost grew from 507 billion in 2006 to more than 708 billion in 2008 in the USA alone. Thanks to information campaigns we all know that peo- ple should be more active, we all know that people need more physical exercise. However, these campaigns fail to bring the required impact, fail to change the sedentary lives of the majority. Here information technologies can help ac- tively promote a healthier lifestyle. We propose a compre- hensive process of physical activity management, driven by various kinds of motivational factors supporting its users in achieving the required amount of physical activity per day. The process consists of recommendation of the appropriate amount of physical activity in the form of a personalized daily plan and automatic tracking and subsequent evalua- tion of performed physical activity. An important part of our proposal is the integration of different motivational factors, both intrinsic and extrinsic, which ensure the long-term sustainability of the whole pro- cess. The motivation comes hand in hand with education, 61

which can be delivered in a non-intrusive manner, taking into account the current context of the user. We realized this process in a solution called Move2Play, taking advantage of the capabilities of widely available smart- phones to measure, assess and recommend physical activity. Albeit the proposed concept is generic enough and suitable for any age group, we focused in our realization and exper- iments more closely on children and developed a children- specific implementation of the system called Move2PlayKids, putting even more stress on motivation suitable for children and involving parents in the process. We conducted several experiments evaluating the individual parts of the system as well as the whole concept in cooperation with doctors and real users, which yielded encouraging results. The paper is structured as follows: In section 2, we sum- marize design requirements and propose the essential parts of a system that encourages physical activity. In section 3, we provide an overview of existing solutions. In section 4, we present an overview of Move2Play. In section 5, we give a detailed description of personalized recommendation for a healthier lifestyle. In section 6, we present the realization and evaluation of the Move2PlayKids system. Finally, in section 7, we conclude the paper and discuss future work. 2. DESIGN REQUIREMENTS TO ENCOURAGE PHYSICAL ACTIVITY There have been many solutions encouraging physical ac- tivity proposed in recent years. Unfortunately, most of them were not adopted enough to fulfill the intended purpose. To find out why, we analyzed the existing solutions to identify design flaws that greatly hindered their adoption in prac- tice. Based upon literature review [1, 6, 21] we identified key design requirements that represent a base for a success- ful solution for encouraging physical activity. 1. Give the user proper credit for activity. Physical activ- ity measurements should be made with sufficient accu- racy as measuring less activity is demotivating. It is preferable to calibrate the measurement device so that more activity is measured rather than less. 2. Provide personal awareness of activity level. Self-as- sessment and monitoring are very effective and one of the most widely used motivation techniques. A sys- tem which tracks user’s physical activity should pro- vide this kind of information in a clear and easily ac- cessible way. 3. Ensure fair play. When users compete, it is important that the competition is fair and that every user has an equal opportunity to win. In case the users feel they have a certain kind of disadvantage, they become frustrated and demotivated. 4. Provide a variety of motivational tools. It is recom- mended to provide different types and methods of mo- tivation to target a wider spectrum of population. Also, for a single user it is more engaging when there are multiple types of motivation throughout the day. 5. Provide feedback on activities done, as users themselves generally do not know whether the physical exercise they have performed is sufficient and appropriate. 6. Consider the practical constraints of users’ lifestyles. Activity measurement devices should be practical for everyday use and not intrusive. With an external mea- surement device, users often forget to take the device with them or forget to charge the batteries. Also, man- ual input of data into the system is often required. 7. Provide both short-term and long-term motivation. One of the issues that current solutions experience is that after a few weeks, users continually drop out and stop using the system [11]. It is therefore important to pro- vide both short-term and long-term motivation to com- bat this issue. 8. Support social influence. Social relations are an impor- tant part of everyone’s life. People are interconnected, and so their health habits are interconnected as well. If a friend starts exercising or eating healthier, it directly increases the probability that we will do the same in the future. A recent study [7] shows that this influ- ence is very significant and reaches as far as to social distance1 of 4. 9. Provide possibility of integration with existing solutions. There already are many great projects, especially games that we can take advantage of instead of developing new ones from scratch. We need a design that is open to third-party games, tools or research projects which seek to add value to the overall solution. 10. Protect users’ privacy. It is important to protect users’ data concerning their physical activity. Developers need to address implementation issues, such as com- munication and database security and functional re- quirements such as the ability for users to determine which data are visible to their friends. Although some of the requirements may seem obvious, they are often forgotten or not paid adequate attention to. We believe that these ten design requirements need to be thoroughly considered in any system that encourages phys- ical activity, as they summarize the most important user needs and expectations. We propose to divide the aforementioned requirements into four essential parts, which we believe form the basis for long-term progress and sustainability, as shown in Fig. 1. • Activity Recommendation. Firstly, we need to provide personalized activity recommendation, so the user knows how much and what type of activity is ap- propriate. As many users are discouraged even before beginning activity, believing that they can never exer- cise enough to actually change anything, the purpose is to show the user that her goal is achievable and she does not need to completely and radically change her lifestyle to reach it. • Activity Tracking. In the next step, we need to measure activity automatically as accurately as pos- sible and with the least possible extra burden on the user. Requirements to manually fill complicated charts describing the performed physical activity or to wear 1Social distance of 1 refers to a friend, a distance of 2 to a friend of a friend, etc. 62