Putting actual behavior back into the social sciences: A Plea for video-based interaction ethology

To a growing extent, the social sciences have become uncoupled from the study of real-life behavior. This unfortunate situation is linked to the default methods used—including self-reported accounts and experimental simulations—which offer a coarse-grained and often ecologically invalid picture of human behavior in its natural context. To break this methodological impasse, we suggest that digital video data provides a much-needed reality check of the social science conceptualizations of human behavior. Specifically, we argue that the means to this is an interactional ethological approach, fueled by the recent availability of high-quality video recordings of public place behavior. We illustrate this prospect through an outline of recent video-observational research on bystander intervention behavior, and discuss further avenues for the advancement of a video-driven interaction ethology.