Personalized and Trusted Behaviour Change Technologies

这项拟议的研究将推动计算机系统领域的最先进水平，支持人们为了自己的利益和社区的利益而改变自己的行为。它将开发新的个性化说服技术和激励机制，这些技术和激励机制得到用户的信任，不仅让他们，而且让他们的社交圈(朋友、邻居、家人)参与支持他们实现行为改变目标。我将从经验(通过用户研究和观察)、理论(通过设计信任和激励机制并通过理论分析和模拟对其进行评估)和实践(通过设计和评估真实系统中的激励机制)来探索有效的激励机制。我将使用社会心理学中关于动机和认知偏差的精选理论作为起点，来指导说服策略和激励机制的设计，以激励用户从事有益的行为。应用领域包括网上购物、自我完善(学习)、健康的生活方式、在工作和休闲时实践安全行为，以及增加孤立(老年人或受损)个人的积极社会关系。 虽然在过去五年中，在设计说服和行为改变软件方面进行了积极的研究，但拟议的研究在三个方面是新颖的： 1)大多数现有的方法仍然是“一刀切”的，至多提供针对用户类型的离线定制策略。拟议的研究将在新的用户建模基础设施的基础上开发个性化的说服性干预措施，该基础设施允许收集来自不同来源的不同数据，并考虑到信息来源和消费者的可信度。 2)用户对说服性技术的信任还没有得到充分的研究，因此无法知道他们是否因为缺乏信任而失败。这项拟议的研究将探索用户信任与说服性技术有效性之间是否存在关联，以及是否考虑其他社会因素，如关系和文化，是否会影响用户信任和说服性技术有效性。 3)拟议的研究将在重要领域产生几个行为改变系统。他们将分享理论思想和方法，但每个人都将探索在特定领域和激励目标的背景下纳入动态个性化、信任和机制设计的问题。拥有多个系统将提供一个更高层次的视角，了解拟议的框架、不同方法和战略是如何概括的。

The proposed research will advance the state of the art in the area of computer systems supporting people to change their behaviour for their own benefit and for the benefit of the community. It will develop novel personalized persuasive technologies and incentive mechanisms that are trusted by users and that involve not just them, but their social circles (friends, neighbours, family members) in supporting them to achieve their behaviour change goals. I will explore effective motivational mechanisms empirically (through user studies and observations), theoretically (by designing trust and incentive mechanisms and evaluating them through theoretical analysis and simulations) and in practice (by designing and evaluating motivational mechanisms in real systems). I will use as a starting point selected theories of motivation and cognitive bias from social psychology to guide the design of persuasive strategies and incentive mechanisms that motivate users to engage in beneficial behaviours. The application areas include online shopping, self-improvement (learning), living a healthy lifestyle, practicing safe behaviours at work and leisure, and increasing the positive social connections of isolated (elderly or impaired) individuals. While there has been active research in the area of designing software for persuasion and behaviour change over the last five years, there are three aspects in which the proposed research is novel: 1) Most existing approaches are still "one size fits all", providing at most offline tailoring of strategies to user types. The proposed research will develop personalized persuasive interventions, based on a new user modeling infrastructure that allows to aggregate diverse data from various sources, and for different purposes, considering how trustworthy the sources and the consumers of the information are. 2) User trust in persuasive technologies has not been researched sufficiently, so there is no way to know if they are failing due to a lack of consideration of trust. The proposed research will explore if there is a correlation between user trust and the effectiveness of persuasive technologies and if considering other social factors, such as relationships and culture can influence both user trust and persuasive technology effectiveness. 3) The proposed research will result in several behaviour change systems in important areas. They will share theoretical ideas and methodological approaches, but each will explore the problems of incorporating dynamic personalization, trust and mechanism design in the context of the particular domain and motivational purpose. Having multiple systems will provide a higher-level perspective on how the proposed frameworks, different methods and strategies generalize.