This dissertation investigates social reactions to prosocial norm violation, or violating a social norm to help others. I investigated how prosocial norm violation (e.g., wearing inappropriate clothing to advocate for an important cause) separately affects how powerful that person appears to others and how much power people are willing to give to that person.

Across four main studies and five supplementary studies, when a target performed a prosocial action, they were seen as slightly less powerful, but were much less likely to be given a powerful role, when that action was a norm violation (versus norm consistent). When the positive impact of the prosocial action was heightened, the norm violator was seen as just as powerful as the norm follower, but was still less likely to be given power. Mediation analyses suggest that prosocial norm violators were perceived as more agentic than prosocial norm followers, which increased power perception. However, the violators were also perceived as less communal, which decreased both power perception and power conferral.

In conclusion, prosocial norm violation leads to mixed social outcomes: while prosocial violators were perceived as more agentic, they were also perceived as less communal and less powerful, and were given less power. Prosocial norm violation appears to not be a clear pathway to power.

This dissertation investigated whether individuals lose influence by saying no to others’ helping requests. Considering both dominance and prestige as pathways to influence, I found that the answer varied depending on whether the help was costly to provide. When someone was asked for help, saying no had two opposing effects on their actual and perceived influence by increasing their dominance, but decreasing their prestige. The cost of providing help moderated these effects. Overall, refusing to help decreased a person’s influence when helping cost little time, effort, or money, compared to both agreeing to help and a control condition. This effect was eliminated or reversed with a higher cost of helping. This is because individuals who refused to provide low-cost help were perceived as less prestigious and influential than those who refused to provide high-cost help, but individuals who agreed to help were perceived similarly regardless of helping’s cost. Participants were insensitive to the cost of helping when evaluating individuals who agreed to help, because they failed to make spontaneous comparisons between different costs of helping. In fact, when participants had a chance to see both high-cost and low-cost helping requests, they perceived individuals who agreed to provide high-cost help to be more prestigious and influential than those who agreed to provide low-cost help. In this case, increasing the cost of helping did not reduce the negative effect of refusing to help on perceived prestige and influence. In conclusion, this research highlights the importance of both prestige and dominance effects for a person’s influence, and the cost of helping as a major contextual factor for helping’s consequences. The findings imply that employees who want to obtain influence should say yes to helping requests when providing help is not very costly, but should feel free to say no when providing help is truly costly.