A particle swarm model of socio-psychological dynamics based on Heider’s balance theory

The emergence of collective behavior has been extensively studied in the field of artificial life. We propose a self-driven particle system with the dynamics of both social preferences and social relationships in the socio-psychological space to computationally understand the dynamics of human social relationships. Social preferences, represented as a matrix with the values of how much a person prefers another, is updated according to Heider’s balance theory. In addition, social relationships, represented as the distribution of particle agents in the two-dimensional space, is updated based on Kano’s model. Our experimental results show that if we assume the loop dynamics caused by the social preferences and social relationships, the community tends to converge to a state with two major subgroups accompanied by a few minor subgroups as a locally optimal solution.