Key points are not available for this paper at this time.
Abstract Anthropomorphism represents a central theoretical term in social robotics and human robot interaction (HRI) research. However, the research into anthropomorphism displays several conceptual problems that translate into methodological shortcomings. Here we report the results of a scoping review, which we conducted in order to explore (i) how the notion of ‘anthropomorphism’ is understood in HRI and social robotics research, and (ii) which assessment tools are used to assess anthropomorphism. Three electronic databases were searched; two independent reviewers were involved in the screening and data extraction process; a total of 57 studies were included in the final review which encompassed 43 different robots and 2947 participants. Across studies, researchers used seven different definitions of anthropomorphism and most commonly assessed the phenomenon by use of amended versions of existing questionnaires ( n = 26 studies). Alternatively, idiosyncratic questionnaires were developed ( n = 17 studies) which, as a qualitative thematic analysis of the individual questionnaire items revealed, addressed nine distinct themes (such as attribution of shared intentionality, attribution of personality etc.). We discuss these results relative to common standards of methodological maturity and arrive at the conclusion that the scope and heterogeneity of definitions and assessment tools of anthropomorphism in HRI hinders cross-study comparisons, while the lack of validated assessment tools might also affect the quality of results. To nurture reflection on these methodological challenges and increase comparability within the field we conclude by offering a set of reporting guidelines for research on anthropomorphism, as a first constructive effort to facilitate a coherent theory of anthropomorphism.
Damholdt et al. (Tue,) studied this question.