When Do People Learn More from Others’ Prosocial Behavior? A Meta-Analysis of Prosocial Modeling Effect

Prosocial behavior is contagious, such that prior work has shown people (across all ages) rely on social examples –whether and what others contribute – as guides of their own prosocial decisions. Specifically, both children and adults are more likely to cooperate (Doescher & Sugawara, 1992; Braver & Barnett, 1976), donate or share resources (Poulos & Liebert, 1972; Wagner & Wheeler, 1969) and care for others (Gray & Pirot, 1984; Bryan & Test, 1967) when they learn that others have done the same. The prototypical prosocial modeling situation involves a prosocial model and someone who observes the model’s behavior (referred to as observer in this paper). Specifically, an observer is exposed to a prosocial model who helps someone in need. Then, the observer is given the opportunity to also help someone in need. The critical outcome of interest is whether the observer imitates the model’s behavior. Despite the widespread use of modeling in encouraging helping behavior, research has suggested that the effects of prosocial modeling events vary as a function of the social context in which they occur (Masor, Hornstein, & Tobin, 1972). Specifically, some findings suggest that prosocial modeling has negligible effect when helping is costly (Wagner & Wheeler, 1969), when the intensity of modeling behavior is low (Rosenbaum, 1956) and often doesn’t generalize to other behaviors than those that are modeled (Braver & Barnett, 1976). Thus, our first goal of this meta-analysis was to examine the robustness of the prosocial modeling effect.