Short abstract:

Video technology is ubiquitous in HRI. However, despite its widespread use, there is no consensus on methodological approaches. As an epistemic tool to make the social visible in social robotics, we advocate videographic analysis as a qualitative methodological insight from STS.

Long abstract:

Video technology is as ubiquitous in experimental HRI settings as it is in robotics in general. Videos are used as communication tools for the general public or funding agencies. Although sometimes deceptive, videos serve as heuristic devices in studies, as proof of concept, or as tools for documentation. Despite the widespread use of video 'in the field', there is little consensus on methodological approaches to using this data.

Drawing on a forthcoming paper, we propose the use of videographic analysis as an epistemic tool in the research process to make the social in social robotics visible - literally. While HRI is technology-driven, it has integrated experimental quantitative methods of psychology in the past. With qualitative videography, we want to add methodological STS knowledge to the epistemic process.

In our talk, we will show why video analysis, with its own set of methodological considerations, is particularly suited to the complex interactions with current robots, which are often characterized by ruptures, expectation violations, and other ambivalences, which are 'repaired' or ignored by the human counterpart.

To illustrate the advantages of the method and to give insights into its application, we draw on examples from our research. We follow the complexity of the situation analyzed from the single experiment in the laboratory to experiments 'in the field'. In addition, we discuss the importance of ethnography for videographic work in HRI in terms of interpreting and making sense of the data recorded and the conception of the video recordings.