SInfoNiA: modeling information, modeling the intangible
In the streets, rumors are swirling. The local bank is on the verge of bankruptcy. Customers of the bank, which has been in business for decades, are increasingly worried. Panicked, they rush to the counter to withdraw their money. The bank collapses. This tried-and-tested scenario shows that the economy is not limited to the exchange of physical goods or services. Information, and particularly its dissemination, can have a major impact on the coordination of actions within a group, on collective choices and on individuals.
The SInfoNiA project, which the European Union is supporting with an ERC Advanced grant for the next 5 years, focuses on this phenomenon. It is based on the development of an original interactive information model adapted to game theory. This highly interdisciplinary field of research is fundamental to economics. It enables us to study, in a system where the choices of everyone have an influence on the others, the decisions taken by several individuals or objects (countries, companies, etc.) defending their interests.
"The first challenge for us is to describe the information," says Olivier Gossner, Director of Research at CREST* and project leader. In concrete terms, this can take the form of what circulates on social networks, in newspapers, in interpersonal communications... in short, everything that shapes an individual's beliefs and vision of the world. "And everyone's decisions depend on their beliefs about the decisions of others. What I believe you believe influences what I believe you're going to do, and therefore my decisions," laughs the researcher.
An immaterial universe, with very real consequences
But while economics is capable of modeling choices, there is yet no tool capable of describing how information circulates and impacts the players in a situation. "SInfoNiA has the profound ambition of modeling this uncontrollable, immaterial universe, which nonetheless has very real consequences”.
In the project, beliefs, opinions, news, etc. are represented by a combination of probabilistic and algorithmic mathematical tools. The aim is to create a scientific language that is relevant, representative of reality and usable by as many people as possible.
These methods and tools will be used in the engineering component of SInfoNiA. They will enable Olivier Gossner and his team to study the impact of information circulation within various organizations. This work will raise the question of the level of information to which the players in these structures must have access if they are to function properly. "For example, some online sites encourage their customers to choose priority delivery for a few euros. However, the customer does not know whether other people have made the same choice, whether there are enough delivery drivers, and therefore whether they really have priority. The company plays with ambiguity, in its own interest. This is an example of the very real impact of disseminating information", explains the researcher.
SInfoNiA also proposes to study new models of financial stability, making it possible, for example, to predict investor commitment based on data circulating about a company's health. Another part of the project will focus on the robustness of social equilibria. "Let's start from the principle that society only exists because everyone is committed to it. Can misinformation and fake news destabilize or even break this social contract?" asks Olivier Gossner. Finally, we'll be looking at how the spread of information and individual beliefs can help social norms evolve or, conversely, freeze them in place. "One study showed that in Saudi Arabia, most men are in favor of women driving, but everyone is convinced that others are not. The analysis of beliefs and their impact could break down everyone's preconceptions and move things forward".
*CREST : a joint research unit CNRS, GENES, École Polytechnique - Institut Polytechnique de Paris, ENSAE Paris - Institut Polytechnique de Paris
Funded by the European Union (ERC, SInfoNIa). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.