The Social Cognition of Robots: Human–Robot Interaction in Social Life Across Groups and Contexts

For decades, the idea that social robots would soon become part of everyday life has circulated widely within the Human–Robot Interaction research community, and also in media narratives or public discourse. Yet, despite remarkable technological progress, we still do not see robots around us. This gap between expectation and reality invites a deeper question: rather than asking when robots will become widespread, we should ask why they are not already present. Beyond technical limitations, high costs, or infrastructural barriers, one often overlooked reason lies in our limited understanding of how robots are perceived and integrated within the social groups and communities that define human life. Humans are inherently social beings: they act within groups that shape norms, goals, and identities. If robots are to truly become part of society, they must be able to engage with these group dynamics. The main objective of this thesis is to explore how HRI changes when moving from dyadic encounters (one human and one robot) to group and community contexts. The central research question guiding the work is: How does HRI shift across different social groups and social settings? To address this challenge, I focused on two complementary sub-questions: (1) How does the introduction of a robot into a group shape social dynamics?; and (2) How do different stakeholders accept and make sense of HRI within their specific social environments? To answer these questions, I have adopted a multidisciplinary methodology combining social psychology and robotics. In my work I have conducted several experiments involving social robots and human participants. Quantitative and qualitative approaches have been integrated, including behavioral experiments, self-report measures, and field studies. Different robotics platforms have been tested: iCub, NAO, Alter-Ego, and UF850 6-DOF robotic arm. The use of diverse systems, varying in embodiment and autonomy, allowed me to explore how robot morphology and social affordances influence human perception and interaction across contexts. I developed and implemented new experimental paradigms inspired by social psychology, such as the Cyberball Social Exclusion paradigm, to investigate how social norms and group membership shape people’s judgments of robots. In parallel, I created novel versions of methodological tools like the Social Bench Tool, designed to assess implicit attitudes toward robots among children. Finally, I extended my research beyond the laboratory, conducting in-the-wild studies in cultural institutions and creative environments. Across these studies, results consistently demonstrated that group membership and social context strongly influence how humans perceive and interact with robots. In controlled settings, participants tended to judge robots more harshly than human agents during social exclusion scenarios, revealing less tolerance against robots that violate social norms. However, children meeting the iCub robot showed increasing physical and psychological proximity after live encounters, highlighting early-stage mechanisms of social acceptance. In real-world contexts, museum visitors perceived the autonomous robot Alter-Ego as engaging and enjoyable, though its acceptance depends on the individual and contextual factors. Finally, artists collaborating with a robotic painting system described the experience as reflective and creatively stimulating, suggesting that robots can become co-creative partners rather than mere tools. By systematically comparing HRI across multiple social groups and contexts, this work contributes to a deeper understanding of how different users comprehend and accept robots. The results underscore the importance of considering social belonging and contextual framing when designing, deploying, and evaluating social robots. Future developments of this research may inform the design of adaptive, socially aware robots capable of participating in group activities, education, cultural mediation, and artistic co-creation. More broadly, the findings advocate for a shift in HRI research—from studying isolated dyadic interactions toward embracing the complexity of collective and community-based human–robot relations.