An Interplay Between Noise and Temperature in a Simple Coupled Model of Opinion Dynamics
Anna Chmiel1 ✉️, Julian Sienkiewicz2 ✉️ 1Faculty of Physics, Warsaw University of Technology, Poland 2Faculty of Physics, Warsaw University of Technology, Poland
Cite as: Chmiel, A. and Sienkiewicz, J. (2025, May). An Interplay Between Noise and Temperature in a Simple Coupled Model of Opinion Dynamics. In SCS 2025, 2nd International Conference on Social Contexts of Science (p. 25). Wrocław University of Science and Technology, Poland.
Abstract
Opinion modeling, despite decades of efforts by sociologists, psychologists, and even physicists, still remains a challenge. This task has become even more complex in today’s world, where communication channels between people are diverse and multifaceted. Social physicists have adapted their flagship model from magnetism - the Ising model - to opinion dynamics. In this model, the temperature becomes a key parameter that controls how likely spins/agents are to adopt the opinions of their neighbors (encoded as +1 or -1). In the language of physics, we talk about a continuous phase transition, a shift from an ordered to a disordered state. It has been shown that a small modification of the Ising model, in which a randomly chosen spin interacts only with its q neighbors (q lobby), leads to a switch from a continuous to a discontinuous phase transition with hysteresis at q >= 4. Similarly, in the generalized voter model, instead of just one neighbor, we choose q neighbors of the node and change its state only if all neighbors share the same opinion. In the presence of stochastic noise, interpreted as “independence,” a continuous phase transition is observed only for q ≤ 5, while for q > 5, the transition becomes discontinuous. In real social relationships, we often observe nodes belonging to many different networks. We communicate on the Internet via email or through social media. Each tool forms a separate network, yet a single individual can be a member of several of them simultaneously. To accurately model the dynamics of node/agent states, it is necessary to account for the influence of multiple networks, which becomes possible when we describe the system using a multiplex network. Both the q-voter model and the q-Ising model have been analyzed on multiplex networks. Depending on the value of the parameter q, interesting changes in the nature of phase transitions have been observed. Here we focus on a case where dynamics follow the q-voter model on one layer and the q-Ising model on the other, and an
Keywords
Opinion dynamics, Multiplex network, Agent based model, Phase transition, Sociophysics
Current status of the research is: Work-in-progress
Potential collaboration with Authors
I am open to collaboration