Fitting in and breaking up: A nonlinear version of coevolving voter models

A Bounded-Confidence Model of Opinion Dynamics with Heterogeneous Node-Activity Levels

“…To capture the effect of some agents having more frequent interactions (such as from greater sociability or a stronger desire to share their opinions), we implement such a node-based agent-selection procedure in our study. The choice between edge-based and nodebased agent selection can have substantial effects on the dynamics of voter models of opinion dynamics [36], and we expect that this is also true for other types of opinion models. We are not aware of a comparison of edge-based and node-based agent selection in asynchronous BCMs (and, in particular, in DW models), and it seems both interesting and relevant to explore this issue.…”

“…(See Ref. [36] for a discussion of node-based updates versus edge-based updates in the context of voter models.) In a node-based approach, one randomly selects a first node and then randomly selects a second node from its neighbors.…”

“…To determine how a network with an underlying block structure affects the dynamics of our node-weighted BCM, we consider stochastic-block-model (SBM) networks [39] with 2 × 2 blocks, where each block consists of an ER graph. Inspired by the choices of Kureh and Porter [36], we consider two types of SBM networks. The first has a two-community structure, in which there is a larger probability of edges within a community than between communities.…”

“…In our BCM, to incorporate node weights (e.g., to encode heterogeneous sociabilities or activity levels of individuals), we instead used a node-based selection mechanism. For voter models of opinion dynamics, it is known that the choice between edge-based and node-based agent selection can substantially affect a model's dynamics [36]. We are not aware of a comparison of edge-based and node-based agent selection in asynchronous BCMs (and, in particular, in DW models), and it seems interesting to investigate this issue.…”

Qualitative features of a nonlinear, nonlocal, agent-based PDE model with applications to homelessness

“…22 Human interactions with their environment can be studied both from a social science perspective and through mathematical and statistical methodologies. Models of human social interaction have studied in many contexts, for example voting, 27 lobbying; 18 leader-follower dynamics; 1 and evolutions of opinions, 20,50 including through agent-based models of continuous-valued opinions within a finite group of peers. 36 Models relating humans and the environment have also been put forth, such as those pertaining to climate change.…”