1 The grass is always greener on the other side of the fence The effect of misperceived signalling in a network formation process Simone Giansante 1 , Alan Kirman 2 , Sheri Markose 1 and Paolo Pin 3 1 CCFEA, University of Essex 2 GREQAM, Universit´ e d’Aix Marseille 3 Economics Department, University of Venice, and ICTP, Trieste Summary. We consider a game–theoretical network formation model based on local interaction. Agents dynamically change the target of the links that give them lower utility and self-organize into stable and efficient unconnected networks. They have the opportunity to detect better profits, and will break their connections for new ones only if profitable. Computational results show that, up to a certain threshold, the more this signal is deliberately exagger- ated (so that better opportunities could in fact not be so), the more efficient equilibria are likely to be the outcome. 1.1 Introduction Social and economic networks are becoming increasingly popular in the last ten years, because of both the application of game theory to the network formation processes 1 , and the study of stochastic processes that fit the statistical properties of real world social networks. 2 In the very recent years there have also been attempts to combine the contri- bution of these two streams of research, trying to find strategic models whose equilibria resemble the empirical data. 3 A well known source of debate in the game theoretical approach is the incompatibility between 1 The seminal paper is Jackson and Wolinsky (1996), see Jackson (2005) for a survey of this literature. 2 The starting point of this second stream of research can be considered Albert and Barabasi (1999), see Newman (2003) as a survey. Let us refer to this second scientific contribution as the econo–physics approach. 3 As an example see Jackson and Rogers (2007).