Can socially assistive robots be used to help people develop healthy habits?

社交辅助机器人（SAR）描述了通过社交而不是物理交互为人类用户提供帮助的机器人[1]。SAR包括HRI，社会，康复和服务机器人[2]。SAR的应用可以包括帮助激励参与者参与并遵守教育，福利和健康等任务[3]-[5]。然而，这种机器人在这些领域的有效性在很大程度上取决于机器人说服参与者的能力。其中说服是一种活动，涉及说服者（机器人）试图改变被说服者（人类）的精神状态[6]。以前的工作集中在调查SAR通过提供鼓励来激励参与者在任务期间保持参与的能力[5]。虽然，在训练过程中积极参与是必不可少的，但在使用机器人形成长期习惯方面还没有做太多的工作。习惯是一个人对经常遇到的环境的自动反应[7]。健康的习惯包括早餐时喝一杯水或午餐后散步。此外，在受控环境之外，长期暴露于有说服力的SAR的影响的纵向研究并不多。一项为期两周的使用机器人教日本儿童英语的实地研究显示，在第二周，与机器人的互动迅速下降[8]。虽然在小学进行的不同的为期四个月的用户体验研究表明，当他们收到机器人的正确答案时，他们会与机器人进行更多的互动，例如点亮眼睛或鼓掌。因此，确定维持长期关系的机制是需要进一步研究的HRI的一个关键领域。这将包括确定机器人是否需要在数周或数月内调整/进化其反应以保持高参与度。从人类行为研究中发现了许多在说服中发挥作用的语言和非语言即时性，这些包括但不限于凝视，手势和面部表情[10]。在博士学位的初始阶段，需要对说服的人类心理学模型进行调查，以确定将使用哪些说服策略，以及是否需要为个人量身定制说服策略。

Socially assistive robots (SAR) describes robots that provide assistance to human users through social rather than physical interaction [1]. SAR encompasses HRI, social, rehabilitation and service robotics [2]. The applications of SAR can include helping motivate participants to engage and comply in tasks such as education, well-being and health [3]-[5]. However, the effectiveness of such robots in these areas is very much dependent on the robot's ability to persuade the participant. Where persuasion is an activity that involves the persuader (robot) trying to change the mental state of the persuadee (human) [6].Previous work done has focused on investigating a SAR's ability to motivate participants to stay engaged during a task by providing encouragement [5]. Although, active participation during a training session is essential there hasn't been much work done in using robots to form longlasting habits. Where habits are a person's automatic response to a frequently encountered context [7]. Healthy habits can include drinking a glass of water at breakfast or going for a walk after lunch.Additionally, there haven't been many longitudinal studies on the effect of long exposure to persuasive SAR outside of a controlled setting. A two-week field study to teach Japanese children English using a robot showed a rapid decline in engagement with the robot on the second week [8]. While different four month user experience study in primary school to teach language to children showed there was more engagement with the robot when they received a reward from the robot for correct answers such as lighting up eyes or clapping hands [9]. As a result, identifying the mechanisms for maintaining a long-term relationship is a crucial area of for HRI that needs further study. This will include identifying whether the robot needs to adapt/evolve its responses over weeks or months to maintain high engagement.There are many verbal and non-verbal immediacies identified from human behaviour research that play a part in persuasion, these include but are not limited to gaze, gesture and facial expression [10]. During the initial stages of the PhD an investigation into human psychology models of persuasion will need to be done to identify which persuasive strategies will be usedand whether persuasive strategies need to be tailored for individuals.