TU Darmstadt / ULB / TUbiblio

Pervasive Behavior Interventions - Using Mobile Devices for Overcoming Barriers for Physical Activity

Dutz, Tim (2017)
Pervasive Behavior Interventions - Using Mobile Devices for Overcoming Barriers for Physical Activity.
Technische Universität Darmstadt
Dissertation, Erstveröffentlichung

Kurzbeschreibung (Abstract)

Extensive cohort studies show that physical inactivity is likely to have negative consequences for one’s health. The World Health Organization thus recommends a minimum of thirty minutes of medium-intensity physical activity per day, an amount that can easily be reached by doing some brisk walking or leisure cycling. Recently, a Taiwanese-American team of scientists was able to prove that even less effort is required for positive health effects and that as little as fifteen minutes of physical activity per day will increase one’s life expectancy by up to three years on the average. However, simply spreading this knowledge is not sufficient. Roughly one in three Europeans and US-Americans does not even meet the minimum recommendations for physical activity, although the majority of these people is aware of the damage that their behavior may do to their health. And this ‘willful wrongdoing’ does not only concern individuals: Due to the large number of inactive people, the problem of sedentary behavior affects societies as a whole, not the least by increasing public health costs.

But if it is not a lack of knowledge that causes this problem, what is? And what can be done to stimulate leisure-time physical activity? The Fogg Behavior Model (FBM), developed by psychologist and Stanford-lecturer B.J. Fogg, explains the factors that determine whether or not a given person will show a desired behavior. The core components of the FBM include a trigger that can be perceived by the target person and that she associates with the desired behavior, as well as her ability and motivation for this behavior at the time when the trigger reaches her. If the combined amount of ability and motivation exceeds a lower limit, the so-called activation threshold, then the triggered person will behave in the desired way; otherwise, she will not. Based on the understanding of human behavior that the FBM conveys, this thesis focuses on the question of how mobile devices can assist people in reaching the minimum amount of daily physical activity that is required for health benefits.

An in-depth analysis of the problem reveals that of the three possible strategies – trying to increase a user’s ability for leisure-time physical activity, trying to increase her motivation for the same, and trying to increase her short-term awareness for its necessity and feasibility through triggers – the creation of adaptive triggers is the most promising approach. This task in turn consists of several sub-problems, such as the problem of how to recognize the user’s current contextual situation, the problem of how to decide, whether or not the recognized situation is suited for an activation attempt, and the problem of interacting with the user in those cases in which an activation attempt seems worthwhile. Learning from the user’s behavior and understanding her preferences and constraints is the key factor in the creation of accurate and reliable intervention mechanisms. To this end, smartphone sensors, wearables, and Web services are utilized for collecting information about the state of the user and her environment. This data is then analyzed by a supervised learning machine which, based on prior experience, estimates the probability for a successful activation attempt in the current situation. Ideally, the learner will identify a kairotic moment: A situation, in which a trigger is bound to initiate the desired behavior. If it does, it reaches out to the user.

Multiple types of such triggering mechanisms were embedded into the mobile exergame ‘Twostone’, an application that requires brisk walking or easy running from its users. During a field study with thirty participants, the performances of these different approaches were compared against one another. The study revealed a surprising result: Not the most-knowledgeable intervention mechanism emerged as a winner, but it was rather the triggering variant that relied on a reduced number of contextual information to achieve both the highest triggering success rates and the best user acceptance. The study also showed that intervention mechanisms can indeed increase the prevalence of a desired behavior, but only if the user has a positive attitude towards the respective activity. As such, both the conceptual model for technology-based interventive measures and the evaluation results that are presented in this thesis offer valuable insights for developers of devices and applications that aim to foster desired behaviors in general and increased levels of daily physical activity in particular.

Typ des Eintrags: Dissertation
Erschienen: 2017
Autor(en): Dutz, Tim
Art des Eintrags: Erstveröffentlichung
Titel: Pervasive Behavior Interventions - Using Mobile Devices for Overcoming Barriers for Physical Activity
Sprache: Englisch
Referenten: Steinmetz, Prof. Dr. Ralf ; Malaka, Prof. Dr. Rainer
Publikationsjahr: 2017
Ort: Darmstadt
Datum der mündlichen Prüfung: 28 November 2016
URL / URN: http://tuprints.ulb.tu-darmstadt.de/6127
Kurzbeschreibung (Abstract):

Extensive cohort studies show that physical inactivity is likely to have negative consequences for one’s health. The World Health Organization thus recommends a minimum of thirty minutes of medium-intensity physical activity per day, an amount that can easily be reached by doing some brisk walking or leisure cycling. Recently, a Taiwanese-American team of scientists was able to prove that even less effort is required for positive health effects and that as little as fifteen minutes of physical activity per day will increase one’s life expectancy by up to three years on the average. However, simply spreading this knowledge is not sufficient. Roughly one in three Europeans and US-Americans does not even meet the minimum recommendations for physical activity, although the majority of these people is aware of the damage that their behavior may do to their health. And this ‘willful wrongdoing’ does not only concern individuals: Due to the large number of inactive people, the problem of sedentary behavior affects societies as a whole, not the least by increasing public health costs.

But if it is not a lack of knowledge that causes this problem, what is? And what can be done to stimulate leisure-time physical activity? The Fogg Behavior Model (FBM), developed by psychologist and Stanford-lecturer B.J. Fogg, explains the factors that determine whether or not a given person will show a desired behavior. The core components of the FBM include a trigger that can be perceived by the target person and that she associates with the desired behavior, as well as her ability and motivation for this behavior at the time when the trigger reaches her. If the combined amount of ability and motivation exceeds a lower limit, the so-called activation threshold, then the triggered person will behave in the desired way; otherwise, she will not. Based on the understanding of human behavior that the FBM conveys, this thesis focuses on the question of how mobile devices can assist people in reaching the minimum amount of daily physical activity that is required for health benefits.

An in-depth analysis of the problem reveals that of the three possible strategies – trying to increase a user’s ability for leisure-time physical activity, trying to increase her motivation for the same, and trying to increase her short-term awareness for its necessity and feasibility through triggers – the creation of adaptive triggers is the most promising approach. This task in turn consists of several sub-problems, such as the problem of how to recognize the user’s current contextual situation, the problem of how to decide, whether or not the recognized situation is suited for an activation attempt, and the problem of interacting with the user in those cases in which an activation attempt seems worthwhile. Learning from the user’s behavior and understanding her preferences and constraints is the key factor in the creation of accurate and reliable intervention mechanisms. To this end, smartphone sensors, wearables, and Web services are utilized for collecting information about the state of the user and her environment. This data is then analyzed by a supervised learning machine which, based on prior experience, estimates the probability for a successful activation attempt in the current situation. Ideally, the learner will identify a kairotic moment: A situation, in which a trigger is bound to initiate the desired behavior. If it does, it reaches out to the user.

Multiple types of such triggering mechanisms were embedded into the mobile exergame ‘Twostone’, an application that requires brisk walking or easy running from its users. During a field study with thirty participants, the performances of these different approaches were compared against one another. The study revealed a surprising result: Not the most-knowledgeable intervention mechanism emerged as a winner, but it was rather the triggering variant that relied on a reduced number of contextual information to achieve both the highest triggering success rates and the best user acceptance. The study also showed that intervention mechanisms can indeed increase the prevalence of a desired behavior, but only if the user has a positive attitude towards the respective activity. As such, both the conceptual model for technology-based interventive measures and the evaluation results that are presented in this thesis offer valuable insights for developers of devices and applications that aim to foster desired behaviors in general and increased levels of daily physical activity in particular.

Alternatives oder übersetztes Abstract:
Alternatives AbstractSprache

Die gesundheitlichen Folgen körperlicher Inaktivität sind durch umfangreiche epidemiologische Kohortenstudien hinreichend belegt. Entsprechend empfiehlt die Weltgesundheitsorganisation jedem Erwachsenen ein Mindestmaß von täglich dreißig Minuten körperlicher Aktivität mittlerer Intensität, also beispielsweise zügiges Gehen oder gemütliches Radfahren. Einem taiwanesisch-amerikanischen Forscherteam gelang der Nachweis, dass bereits die Hälfte dieser Menge, also nur fünfzehn Minuten, genügen, um die eigene Lebenserwartung um bis zu drei Jahre zu erhöhen. Allerdings führt das Wissen über den Zusammenhang zwischen Bewegung und Gesundheit nicht zwingend zu einer Erhöhung des Aktivitätslevels. Etwa ein Drittel der europäischen und US-amerikanischen Bevölkerung leistet nicht einmal das erforderliche Mindestmaß, obwohl das Wissen um die möglichen Folgen dieses Verhaltens meist vorhanden ist. In der Häufung hat dieses bewusste Fehlverhalten Einzelner auch Auswirkungen auf die Gesamtgesellschaft, insbesondere durch einen Anstieg staatlicher Gesundheitsausgaben.

Wenn Bewegungsmangel aber nicht auf mangelnde Informiertheit zurückzuführen ist, worauf dann? Und welche alternativen Maßnahmen können helfen? Das Foggsche Verhaltensmodell (FBM) des an der Stanford-Universität lehrenden Psychologen B.J. Fogg beschreibt die Faktoren, die darüber entscheiden, ob eine Person ein gewünschtes Zielverhalten zeigt. Die wesentlichen Komponenten des FBM sind ein wahrnehmbarer und gedanklich mit dem Zielverhalten verbundener Auslöser, der Trigger, sowie Befähigung und Motivation für das gewünschte Verhalten zum Zeitpunkt des Auftretens eines solchen Triggers. Sind Befähigung und Motivation in ausreichendem Maße vorhanden und überschreiten in Kombination die sogenannte Aktivierungsgrenze, so zeigt die ‚getriggerte‘ Person das Zielverhalten – andernfalls nicht. Basierend auf diesem Verständnis menschlichen Verhaltens befasst sich die vorliegende Arbeit mit der Frage, wie mobile Endgeräte dazu genutzt werden können, um Personen zuverlässig zum erforderlichen Mindestmaß an körperlicher Aktivität anzuregen.

Die Analyse der Problemstellung macht deutlich, dass von den drei möglichen Ansätzen – Steigerung der Befähigung für körperliche Aktivität, Steigerung der Motivation und Steigerung des Bewusstseins für Notwendigkeit und Machbarkeit durch den Einsatz von Triggern – die Entwicklung adaptiver Triggering-Mechanismen am vielversprechendsten ist. Diese Herausforderung lässt sich ihrerseits in mehrere Teilprobleme unterteilen, etwa das Problem des Erfassens der gegenwärtigen Situation, das Problem der Entscheidung, ob ein Interventionsversuch unternommen werden sollte, sowie das Problem der eigentlichen Nutzerinteraktion. Aus dem Nutzerverhalten zu lernen, um Vorlieben und Möglichkeiten des Nutzers richtig einzuschätzen, ist dabei die Schlüsselfähigkeit erfolgreicher Maßnahmen. Vor diesem Hintergrund wird von Smartphonesensorik, Wearables und Web Services Gebrauch gemacht, um Informationen über den Nutzer und die aktuelle Situation zu erhalten. Ein überwachter Lerner analysiert diese Daten und entschiedet auf Basis zurückliegender Erfahrungen, ob ein Aktivierungsversuch erfolgen sollte. Im Idealfall stellt der Lerner dabei fest, dass ein kairotischer Moment vorliegt: eine Situation, in der ein Trigger zuverlässig das gewünschte Verhalten auslöst.

Mehrere unterschiedliche Triggering-Mechanismen wurden in das mobile Fitnessspiel ‚Twostone‘ eingebettet, eine Anwendung, die vom Nutzer zügiges Gehen oder lockeres Laufen erfordert. In einer umfangreichen Feldstudie mit dreißig Teilnehmern wurden alle Varianten miteinander verglichen und ein unerwartetes Ergebnis erzielt: nicht derjenige Mechanismus mit der umfangreichsten Wissensbasis erreichte die höchsten Erfolgsraten und die beste Nutzerakzeptanz, sondern eine Variante mit einer reduzierten Menge an Kontextwissen. Zudem zeigt die Evaluation, dass intelligente Trigger tatsächlich die Häufigkeit eines gewünschten Verhaltens erhöhen können, aber nur, sofern beim Nutzer bereits eine positive Grundeinstellung gegenüber diesem Zielverhalten vorhanden ist. Mit diesen Erkenntnissen leistet die vorliegende Arbeit einen wichtigen Beitrag zur zukünftigen Entwicklung interventiver Maßnahmen und insbesondere diejenigen technischen Anwendungen, die Personen dabei unterstützen sollen, das für positive Gesundheitseffekte erforderliche Mindestmaß an körperlicher Aktivität zu erfüllen, können darauf aufbauen.

Deutsch
URN: urn:nbn:de:tuda-tuprints-61270
Sachgruppe der Dewey Dezimalklassifikatin (DDC): 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik
600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften und Maschinenbau
Fachbereich(e)/-gebiet(e): 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Datentechnik > Multimedia Kommunikation
18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Datentechnik
18 Fachbereich Elektrotechnik und Informationstechnik
Hinterlegungsdatum: 09 Apr 2017 19:55
Letzte Änderung: 09 Apr 2017 19:55
PPN:
Referenten: Steinmetz, Prof. Dr. Ralf ; Malaka, Prof. Dr. Rainer
Datum der mündlichen Prüfung / Verteidigung / mdl. Prüfung: 28 November 2016
Export:
Suche nach Titel in: TUfind oder in Google
Frage zum Eintrag Frage zum Eintrag

Optionen (nur für Redakteure)
Redaktionelle Details anzeigen Redaktionelle Details anzeigen