In this project the Principal investigators will investigate the social network effects on coordination and cooperation. While most existing work provides insights on the influence of group size on coordination and cooperation, we examine the role of group structure. First we will analyze a laboratory experiment on a coordination game in which each player engages with an exogenously connected ?local? subset of the player population, the pattern of connections defining a social network. The main research question we will address is How does the social network structure affect coordination behavior? We identify sources of influence on coordination from both global network (the density of overall network) and local network (the connectedness of individuals). In our experiment, both full-coordination and no-coordination are pooling equilibria in all laboratory networks, while in one case there is also a separating equilibrium, where players coordinate only if they have an enough number of local social links. In a pilot study we observe nearly complete coordination in denser networks but less so in sparser networks. Via a laboratory public goods game, we will study the impact of social network architecture on cooperation. Our research foci are the fairness and social welfare in equilibria as affected by network configurations. To analyze the data collected from experiments, we shall apply various statistical techniques including nonparametric methods and logit regressions (adjusted to individual heterogeneity).
In terms of broader impacts, this research examines how the social network, which determines ?who plays the game with whom?, affects the outcome of coordination and cooperation. Our results might guide Internet service and platform providers in developing effective online social networks to promote service and products, and assist policy makers in seeking for efficient network layouts to facilitate the spread of goods or activities that are socially beneficial. The products of the proposed study will be made publicly available for research and educational uses at no cost.
This project investigates the social network effects on the coordination and cooperation of individual decision makers (called agents). The term coordination / cooperation here respectively refers to the case where an individual’s incentive to take a certain action increases / decreases if more individuals connected to her in the network takes the same action. Typical applications include technology adoptions under network effects (coordination), and the supply of informational goods in Peer-to-Peer networks (cooperation). Research Outcomes Through controlled laboratory experiments, we find that the aggregate coordination level is close to the theoretical prediction (or equilibrium). Furthermore, our work suggests that a simple decomposition of the complex institution into global network and local network might be sufficient for predicting the level of successful coordination. Other things held equal, the level of agent coordination increases with her number of connections, as well as the density of the network. Also by means of laboratory experiments, we find that individual cooperation is primarily affected by one's local network, rather than global network. In some cases, providing localized network information helps enhance the cooperation of agents, and thereby increases the social welfare. The observed agent behavior can be explained as a consequence of myopic learning. Intellectual Merit Existing studies on network externalities often focus on the case that everyone coordinates / cooperates with everyone else, so that individual decisions hinge on network size. [1] My research, however, concerns situations where one interacts with a subset of agent population. [2] As a result, what matters to decision making is the network structure determining "who interacts with whom". Broader Impact Our study sheds insights on how coordination and cooperation behaviors are shaped and shifted by social networks. These insights will guide platform providers in the pricing and promotion of services and products throughout social networks. Our results also inspire the policy maker in optimizing the social / organizational infrastructure to achieve the efficient outcomes of coordination / cooperation. [1] As an example, a customer will buy a phone of certain brand only if the brand achieves a large enough user base. [2] For example, the Iphone adoption of a certain user only raises the value of the Iphone to a subset of the user population (e.g. the friends or colleagues of the adopting user), since the user will use her Iphone to contact with her friends and colleagues (for whom the phone becomes more valuable for compatibility reasons) but not the rest of people.
