Opinion dynamics in a group-based society

Many models have been proposed to analyze the evolution of opinion structure due to the interaction of individuals in their social environment. Such models analyze the spreading of ideas both in completely interacting backgrounds and on social networks, where each person has a finite set of interlocutors. In this paper we analyze the reciprocal feedback between the opinions of the individuals and the structure of the interpersonal relationships at the level of community structures. For this purpose we define a group based random network and we study how this structure co-evolves with opinion dynamics processes. We observe that the adaptive network structure affects the opinion dynamics process helping the consensus formation. The results also show interesting behaviors in regards to the size distribution of the groups and their correlation with opinion structure. Introduction. – Social science analyzes the large scale behaviour of the society. On the other side the interactions among the individuals are studied by social psychology. The main interest of social modeling is to create a bridge between the two levels of description, observing how, starting from the microscopic description of the interactions among the agents, the global phenomena observed in the society can be reproduced. This kind of modeling has recently attracted the attention of the physicists both for the interest in the characterization of complex systems and for the rich critical behavior that such models exhibit. In this paper we are going to focus on two main aspects observable in the society: the process of membership of groups and the process of opinion formation. The main rules of the interactions among the agents are dictated from social psychology. Other’s opinion is a source of cognitive inconsistency! According to Festinger [8] such inconsistency is experienced as dissonance, a psychological discomfort that people are motivated to reduce. Recently, [24] observed that this phenomenon is amplified inside the social groups. The groups are, in fact, suitable places for interaction between people. An individual can choose three main ways to reduce the dissonance created by the difference of its opinion with the one of the other members of its group [24]: changing its opinion to agree with others in the group, influencing others to change their opinions, or joining a different, attitudinally more congenial group. The first two suggest the basic rules governing the individual interactions. The third one can be linked to the local topology of the network surrounding the individual. The present paper studies a simulation model reproducing the coupled effect of these three strategies in order to better understand the link between the individual choices and the organization of the society into groups. Social networks coupled to various reaction processes, have been intensively studied in the last decades [28]: from epidemics [26], [27], to malware diffusion in electronic technology [19], collective behaviors [16], innovation diffusion and opinion dynamics [3], [11]. Moreover, many opinion dynamic models have been proposed to study the spreading of opinions: some of these models describe opinion as a discrete Boolean choice, like the Voter model [4], [17] or the Sznajd model [29]. The first model describing continuous opinion interaction is known as the Deffuant model [5]. This model is based on the concept of bounded confidence: two individuals having their opinion less far than a threshold are going to have closer opinion after the interaction. Some different implementations of this model taking into account a rejection process [20] or a different type of tolerance threshold connected to the opinion [12] have been proposed in the last years. In our model we incorporate the Deffuant model to describe the opinion dynamics contact process. The interest toward opinion dynamics increases coupling these phenomena with the investigation on the topological structure of social networks: different kinds of netp-1 Author-produced version of the paper published in EPL Europhisics letters, 2010 , vol.91, 58004. Original publication available at http://iopscience.iop.org/0295-5075/91/5/58004/