This is a theoretical analysis and laboratory study of social exchange, based on an expected value model of social power. In this model a power structure indicates opportunities for exchange relations and a sample space of exchange networks. The approach incorporates a micro-model of bargaining that predicts exchange outcomes in each of these possible exchange networks. When drawing on observed relative frequencies of exchange networks (such as arise during repetitive trials of an experiment), the expected value model currently provides the most accurate account of the distribution of benefits among actors in exchange networks. However, the model does not predict the relative frequency of exchange networks that may arise during the exchange process and, in one instance, it fails to account for the observed distribution of exchange payoffs. The model now has been elaborated to include an account of the incidence of exchange networks, and it addresses the anomalous finding reported in the literature. The present project will assess the elaborated model both analytically and empirically, in comparison with a competing graph-theoretic model, employing a set of networks that have not been previously examined in the literature, for which the two approaches make markedly contrasting predictions. %%% This is fundamental research on exchange, bargaining and power in social networks, an area of research that is rapidly coming to a clear consensus about the mathematical laws that describe a range of important phenomena. Rigorous understanding of the dynamic and structural factors that influence the concentration of social capital in real-world networks of exchange depends upon working out the abstract relationships in theory and laboratory experiments such as conducted in this project.
