Attitudes towards social robots and how they are affected by direct and indirect contact with a humanoid social robot

The aim of this thesis is to investigate people’s attitudes towards social robots and the factors that affect these attitudes. It uses a wide range of methodologies and aim to offer valuable and comprehensive information about how attitudes towards social robots work and how they can be manipulated. The thesis starts with a literature review which defines concepts, gives a theoretical framework, and explains the evolution of robotics and the field of Human Robot Interaction (HRI). One of the main questions that this thesis aims to answer is what are people’s attitudes towards robots and what factors affect these attitudes. To address this question, a systematic review was carried out by extracting information and analysing the outcomes of studies which examined people’s attitudes, acceptance, anxiety and trust towards social robots. The results of this review indicate that people typically have slightly positive attitudes towards robots, with the type of exposure to robots (e.g., direct contact, indirect contact or no contact) being the main factor affecting this outcome. This thesis then presents a set of empirical studies that investigate whether and how attitudes towards social robots are affected by direct contact and a particular type of indirect contact (namely, extended contact). As explicit and implicit attitudes toward robots may diverge, these experiments measured both implicit and explicit attitudes. The findings suggested that direct contact affects both explicit and implicit attitudes toward social robots, while extended contact only affects implicit attitudes. Some possible explanations include supraliminal priming, a defensive reaction to a possible persuasion, or the idea that implicit attitudes are more unstable than explicit attitudes. This thesis enables new ways of thinking and investigating human-robot relationships. Specifically, it shows that social psychology techniques (usually used to study how people interact with each other) can be applied to study how people interact with social robots. Another implication of these findings is that intergroup contact may actually affect attitudes towards robots, which could be used to create realistic opinions about social robots.