A Rationale for a Gamified E-Coach Application to Decrease the Consumption of Sugar Sweetened Beverages

The design and implementation of applications for behavior change should be preceded by careful analysis of the behavior change process and the target population. We, therefore, present on the basis of a blended research approach a rationale, opportunities and basic requirements for an application that offers a program for reducing intake of sugar sweetened beverages (SSB) by adolescents. This paper discusses the role of e-coaching and gamification as two high-touch design patterns in the behavior change process. Both design patterns aim at supporting the individual in a transformational journey from a current state toward a desired state where the detrimental behavior should be replaced by healthy alternative behavior. First, an elementary behavior scheme is introduced that frames three empirical studies. In the first study (plenary focus groups; n = 13), participants advised to include system recommendations for alternative healthy behavior, stressed the need for personalization of the e-coach and showed strong appreciation for the inclusion of gamification elements. The second study (online survey; n = 249) showed that SSB-intake is highly contextual and that reasons for (limiting) consumption SSB varies greatly between individuals, which the e-coach application should take into account. In a final small-scale pilot study (n = 27), we observed the potential of the inclusion of gamification elements, such as challenges and rewards, to increase compliance to the self-monitoring process of SSB consumption. Building upon these insights and prior studies, we argue that an e-coach mimics the collaborative practice of the program; its main task is to enrich the interaction with cooperative conversational experiences, in particular with respect to the alignment between user and system, motivational encouragement, personalized advice, and feedback about the activities. In addition, we outline that gamification not only has the potential to increase self-monitoring of the target behavior, user engagement, and commitment with the intervention program, but also enables a designer to shift long-term negative outcome of excessive intake in real life to short-term consequences in a virtual environment. In future larger follow-up studies, we advise to integrate the two design patterns within a social network of virtual and human agents that play a variety of competitive, normative and supportive roles.