A Computational Theory of Collective Action

This project will apply computational modelling and simulation to investigate the problem of collective action: explaining how self-interested parties cam be motivated to coordinate their action to achieve a common benefit. Problems of this sort include management of a common resource pool (such as river water or a fishery) and collectively reducing carbon emissions. Mathematical models from game theory predict that free-riding behaviour will dominate and cause the collective action to fail. However, a range of social factors, such as the existence of norms, a desire to earn "social capital", and leadership mechanisms have been proposed to explain why collective action will often succeed in practice.

This project will use a computational approach to gain new understanding of the social reasoning underlying collective action problems. As with much of the prior research on collective action, we are interested in understanding how self-interested agents can be motivated to participate in a collective action. We will investigate a range of proposed mechanisms. However, unlike existing work, we will take a social computing perspective and consider the following question:

How can software assist user communities to coordinate themselves to achieve collective action?

A complete answer to this question requires developing computational modelling and decision-making techniques, as well as human-computer interface and interaction design. This project will focus on the former, with the following high level objectives:

* To develop a generic computational model for collective action problems, in which a range of individual and social motivations for collective action and individual action-selection strategies can be expressed.

* To build computational models of rich social structures and mechanisms that are thought to contribute to the success of collective action; in particular, social capital, social roles and mechanisms for aggregating knowledge and coordinating individual actions.

* To use software simulations of societies to compare the effectiveness of different internal motivations and social factors on achieving collective action.

This work is timely given the potential for social mechanisms (e.g. norm emergence and social capital) to help communities contribute to pressing global and national agendas, such as the reduction of greenhouse gas emissions and promotion of lifestyle changes to reduce obesity. Given the rapid growth of smartphone use, there is great potential for readily accessible social software to foster collective action within electronically connected communities. This research will provide a rich computational model that can be used for reasoning about how to act in social settings where individual and social goals may be in conflict. Our simulation experiments will identify the social mechanisms that can best foster collective action in different settings, and will provide important design criteria for future social software systems.

The University of Otago is New Zealand’s oldest university, and has a higher proportion of PhD students than any other NZ university. It is ranked 151 in the QS World University Rankings 2018 and has the highest QS Stars rating of 5+ for overall university quality.

For more information about PhD study at the University of Otago, see http://www.otago.ac.nz/graduate-research/study/phddoctoral/. Further information for international candidates is available at http://www.otago.ac.nz/international.