What Can we Learn About Human Nature from Interacting with Strangers? Relationship Type Determines Behavior in the Dictator Game

Behavioral decision-making research has been exceptionally useful in the quest of the social sciences to understand human nature. A frequent assumption of this research is that using strangers as anonymous interaction partners allows for the clearest demonstration of <i>basic</i> human nature. But a diverse array of literature - from social and clinical psychology to ethology - suggests that a stranger is far from a "baseline partner." We argue against the overreliance on strangers in economic games and that instead of one baseline partner, typical relationships should fall into <i>basic types</i> of partners, all eliciting different behaviors. Two high-powered experiments (<i>N</i>s = 848 and 2400) in which participants played a hypothetical dictator game with one of sixteen partners (e.g., mother, friend, stranger) found particular clusters of interaction partners in which the possible partners were grouped into different and intuitively meaningful relationship types (i.e., loved ones, intimate partners, companions, contractual partners, infrahumanized others). The clusters suggest a typology of basic human relationships and predict behavior even when controlling for relationship distance. The findings help to calibrate the outcomes of past dictator games utilizing strangers and offer an interpretative context with a system of relationship types.