Opinion Diversity and the Resilience of Cooperation in Dynamical Networks

Model and reality

The last talk before the summer break was given by Adam Miles from Manchester Metropolitan University.

He investigated how opinion diversity contributes to the mechanics of a dynamical network model and to the resilience of cooperation, by utilising a game theory approach where individuals make use of both public and private information to make decisions. He was able to show that increasing diversity leads to more stable, less connected and less prosperous networks and more frequent, but shallower information cascades. Moreover, the outcome of the conflict between cooperators and cheaters depends on the interplay between population structure, individual decision making and individual opinions.

Adam’s aim was to show us a model that was trying to capture human behaviour in its complexity. But, with those kinds of models it is often difficult to find an unique and general way to test them towards reality due to the presence of technical issues such as the need of having consistent distributed computing requirements.

In social networks it is often difficult to isolate the features we want to test and doing experiments with real people is not always possible, especially at the level of a PhD student when you may not have any additional funding to make them.

So… what’s the point in creating social models?

Well, a possibility could be the engineering side, such as the application of reinforcement learning techniques to model individual decision making. We can use these findings to create a desired social scenario in order to reach some common goals.

What are your thoughts about it ?

Share with us your experience on Twitter or on Slack !