16 May New work on network effects on coordination with conflicting preferences
A few days ago, a work I am specially proud of resulting from a collaboration with Joris Broere, Vicent Buskens and Henk Stoof at Universiteit Utrecht came out in Scientific Reports. The reason for my pride is that it is experimental work, nicely done mostly by Joris, in which computational predictions about how people with different preferences will coordinate in a network are tested. This is a topic that I’ve been interested in since several years ago, when along with Penélope Hernández and Manu Muñoz Herrera we introduced a model for conflicting preferences on networks. We carried out experiments with 240 human subjects, and the resulting data shows that, as expected, the less ‘random’ the network structure, the better the experimental results are predictable by the computational models. This opens the door to one of my long-term goals, namely the design of realistic computational models of societal phenomena that can be used to, e.g., test policies before trying to take them to real life. On top of that, it was a real pleasure to collaborate with this Utrecht gang, and I do hope that this will not be our last joint paper.