Bateson et al.'s (2006) Cues-of-Being-Watched Paradigm Revisited

Bateson, Nettle, and Roberts (2006) provided an intriguing experimental paradigm for investigating the effects of social cues on cooperative behavior in a real-world setting. By placing an image of a pair of eyes on a cupboard door above an "honesty box" for hot beverages, they induced substantially higher amounts of contributions. As this finding has a significant impact on the social sciences and assumptions concerning the meaning of social cues for human behavior, we systematically reanalyzed their procedure and statistical analyses and tried to replicate the results while taking personality factors into account. The overall results of our analysis and replication efforts do not unequivocally support conclusions about the effects of eyes as social cues for cooperative behavior. Problems start with the definition of cooperative behavior, underspecified methods, confounding variables, invalid statistical analyses, and a lack of insight into the factors - including personality factors - that modulate the expected effect. A follow-up experiment with 138 participants showed no effect of eyes on socially relevant concepts and attitudes. Furthermore, none of the personality factors that, according to Bateson et al.'s explanation of reputational concerns, might be a source of effects interacted with any of the measures we used to operationalize these socially relevant concepts and attitudes.