Research On Design Strategies Of Social Robots Based On User Privacy Concerns

The social robot is a robot technology that interacts with human users naturally and intuitively by using the same social norms as humans.During interaction with users,the social robot collects a large amount of user data to provide personalized services;it is also equipped with highly sensitive microphones,and cameras,and always stays awake,which raises privacy concerns for users.Most of the current research is focused on privacy protection technology and engineering development.However,even with the application of privacy-preserving technologies,users still have privacy concerns due to the anthropomorphic design features of social robots or the lack of intuitive privacy statements.Therefore,the key question is: What factors drive users’ privacy concerns? How do these factors affect users’ privacy concerns?Based on the S-O-R(stimulus-organism-response)theoretical framework,communication privacy management theory,and media equation theory,this thesis takes the design characteristics of social robot and incorporates design-related factors such as anthropomorphism,warmth,competence,and transparency of interactive information into the research model through theoretical studies.Perceived privacy risk and perceived privacy control are also used as mediating variables in this thesis’ s research model to better understand the privacy concerns during user interactions with the social robot in home scenarios.An online questionnaire was administered to 223 social robot users in China,and198 valid questionnaires were obtained and empirically analyzed using PLS-SEM(partial least squares structural equations)as well as fsQCA(fuzzy set qualitative comparative analysis).PLS-SEM shows that anthropomorphism,warmth,competence,and information transparency are key factors influencing privacy concerns,that perceived social power negatively moderates the relationship between anthropomorphism and privacy concerns,and that perceived privacy risk mediates the relationship between anthropomorphism,warmth,information transparency,and privacy concerns.The fsQCA results further validate the findings of PLS-SEM and reveal five combinatorial configurations of factors that lead to higher levels of user privacy concerns.Finally,based on the above findings,this thesis constructs a guiding framework for social robot privacy design and uses it for design practice.The main innovative points of this thesis are(1)empirically demonstrating the factors influencing social robot users’ privacy concerns from a psychological cognitive perspective,and(2)integrating the design-related factors of the social robot to fill the gap of how design factors comprehensively affect users’ privacy concerns,and(3)using a combination of PLS-SEM and fsQCA to overcome the problem of oversimplification of hypothesis testing in traditional regression analysis.Overall,this study broadens our understanding of the privacy concerns of social robot users and provides further theoretical and practical insights for subsequent scholars and design developers.