Behavioral Economics in Network Games
BEING is a research project funded by a Marie-Curie Fellowship. The key objective of this research project is to develop a novel theoretical framework in which network formation and information diffusion can be analysed taking into account aspects of modern behavioural economics.
Psychologists and Experimental Economists have been providing models for humans' risk perception and interpretation of information that explain inconsistencies with rational decisions since the 1970's. But these models have not been used in the micro-foundation of dynamics at the macroeconomic level: for instance network formation games and information diffusion processes crucially depend on the individuals' incentives, and even small changes in the latter can potentially lead to crucial changes in the overall outcome. Take, as an illustration, a referendum on an important national decision. In a perfectly rational world, people would adjust their beliefs according to the information they receive, bearing in mind that the same piece of information may reach them via various channels and that the providers of information might have a self-interest. They would then form an opinion on the topic based on their preferences and their belief, and finally would strategically spread information so as to achieve their own best interest. But each link in this chain is affected by irrational behaviour: individuals tend to ignore information they don't like, they might even change their circle of friends depending on their opinion. They often ignore moral hazard problems, and they only spread the information they believe in without taking into account other people's reaction. In the end we might see a deeply divided but badly informed society, rather than a well informed society where a legitimate democratic instrument will lead to the social optimum. Understanding these dynamics, modelling them accurately, and developing measures to mitigate the risk of segregation requires to merge our knowledge on decision making, information aggregation, network formation, and diffusion processes from psychology, economics, and the hard sciences.