This project is based on recognition that the decision to enter social relationships is often a risky one; partners in potentially cooperative relationships might, in fact, cooperate, but they might not, acting exploitatively instead. The researchers study whether well-documented risk tolerance in the domain of losses and risk aversion in the domain of gains extends to the decision between entering and not entering particular risky social relationships, in particular, whether risk attitudes change as the degree of sociality involved increases.
As originally developed, prospect theory is based on the strong empirical regularity that the choice between a risky and a certain alternative is likely to return the risk over the certainty when framed as among losses, but to return the certainty over the risk when framed as among gains. While the extensive "why do people cooperate?" literature has focused on mechanisms that support cooperation, it is theoretically and empirically defendable that entering social relationships is a risky business; while others often will cooperate, they will also often defect, producing potentially quite substantial costs. From prospect theory, it is plausible to predict that people will be biased toward the (risky) alternative of entering potentially cooperative social relationships when confronting losses, but toward not entering them when confronting gains. If so, human sociality might be more a defensive response to the threat of loss than a constructive response to the possibility of gain.
Preliminary experiments provide strong support for this hypothesis, but: (1) Those experiments were conducted using student subjects who were responding to a small number of grade points as incentives, and thus they invite the attempt to replicate with more substantial incentives; and (2) The findings are sufficiently suggestive about the roots of human sociality that they invite elaboration. This research will (1) replicate these findings with more substantial incentives; (2) Test the robustness of the observed effect beyond the two-person relationships that were at stake in the original study, in particular, in n-person relationships; (3) Test explanations for the effect based on biased expectations of others' cooperation across the gains and losses domains; and (4) Most importantly, explore the possibility that the strength of the observed effect is a function of the degree to which subjects' choices are informed by social information. At one extreme, the researchers "subtract" all social content from the encounter, while at the other extreme the encounter is strongly social.
The broader impact of this project is a contribution to the building of conceptual and theoretical bridges across the various social sciences, and between those sciences and the life sciences. It also provides a more firm basis for the behavioral assumptions from which modeling in the social sciences proceeds. More particularly, the researchers' findings test whether prospect theory's well-known decision function is more strongly characteristic of social decision making than of decision making with no social content. The project also contributes to humans' understanding of their most basic attribute "their propensity to form cooperative groups and, more broadly, to enter social relationships with each other.
