A surprising number of the important decisions in modern society are undertaken by one person on behalf of other people. CEOs make decisions on behalf of employees in the firm; pastors on behalf of their congregations; generals on behalf of their troops; teachers on behalf of their students; gang leaders on behalf of their followers; parents on behalf of their family; members of corporate boards on behalf of shareholders; lawyers on behalf of their clients; and perhaps most obviously, elected representatives on behalf of the constituents they represent. Though students of decision-making have an impressive understanding of decisions made by those acting on their own behalf, an analysis has yet to be undertaken of the precise manner in which decisions taken on behalf of others differ from those occurring in a traditional dyadic context. In this research project, the investigators employ a series of experiments to isolate and thereby to ascertain the manner in which decisions are different when they are made on someone else's behalf. In an illustrative experiment, a subject is given $20 and told to divide it with a second player in any fashion. From existing research utilizing single-shot dictator and ultimatum games it is known that subjects typically give nearly half to their anonymous partner. But not known is the extent to which this generosity is curtailed (or enhanced) when subjects are acting on behalf of others. If the structure of the experiment is modified so that a subject acts on behalf of a group of five people (including the subject) and must decide how to allocate $100 (5 x $20) between his/her own group and a second (receiving) group, with all members within a group receiving the same amount, decisions are likely to be different. Identifying the precise nature of the difference in behavior between these two common and important decision arrangements is the central goal of this project. The intellectual merit of the project is that it provides a clear empirical test of the conditions under which leading theoretical models of decision-making apply. For example, the purest form of the rational choice model, drawn from microeconomics, hypothesizes that decision-makers will always attempt to maximize their short-term tangible gains, an expectation that often turns out to be false when people are acting on behalf of themselves. But a second theory, drawn from social and evolutionary psychology and utilized in the current project, emphasizes people's concern for the way others perceive them and therefore leads to the expectation that, somewhat ironically, decision-makers will be more protective of the resources of others than they are of their own. As such, the research holds considerable promise of determining the precise conditions under which leading models of decision-making apply. The research has a broader and more practical impact by helping citizens, reformers, and decision-makers themselves to understand the psychological pressures acting on decision-makers in various real-life contexts. Assuming the results of the experiments are as anticipated, if maximization of short-term gains is the desired goal, giving authority to representatives becomes the most fruitful approach. If, however, the desire is for decision-makers to share and to engage in altruistic punishment, the sensible thing to do is to let people make decisions for themselves.

National Science Foundation (NSF)
Division of Social and Economic Sciences (SES)
Standard Grant (Standard)
Application #
Program Officer
Brian D. Humes
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Rice University
United States
Zip Code