Agent‐based Computer Simulations of Language Choice Dynamics

Abstract

We use agent-based Monte Carlo simulations to address the problem of language choice dynamics in a tripartite community that is linguistically homogeneous but politically divided. We observe the process of nonlocal pattern formation that causes populations to self-organize into stable antagonistic groups as a result of the local dynamics of attraction and influence between individual computational agents. Our findings uncover some of the unique properties of opinion formation in social groups when the process is affected by asymmetric noise distribution, unstable intergroup boundaries, and different migratory behaviors. Although we focus on one particular study, the proposed stochastic dynamic models can be easily generalized and applied to investigate the evolution of other complex and nonlinear features of human collective behavior.

Evolution of tag-based cooperation on Erdős–Rényi random graphs

“…neural-based) mechanisms for their recognition and storage. For example, such studies could address issues previously discussed in the context of competition between languages that is usually motivated by various interethnic di®erences, 48,49 with speci¯c linguistic features such as speech accent serving the role of the major inter-group distinguishing tags. 9, 17 We note here that our model is rather°exible as to its possible future modi¯ca-tions, but more importantly, it follows an often recommended but still debated practice in computational modeling 50 according to which a novel model should be able to successfully deal with all the tasks that the model it is replacing can account for, but also to cover at least several additional ones (for a related discussion of this principle in macroscopic modeling, see also Ref.…”

Cooperation in the snowdrift game on directed small-world networks under self-questioning and noisy conditions

“…Another line of future generalizations that could potentially make our model more realistic would include simulations with reproducing agents [36], different migratory behaviors of interacting individuals [50,53], bottleneck and ageing effects on cooperation [62], or the influence of the approaching extinction of a studied population [63].…”

“…Considering such noise effects in models of social dynamics [48,49,50,51] might be useful for a better understanding of the influence of bounded rationality in evolutionary snowdrift and other games [52]; see also Ref. [53] for cooperation under noisy conditions in the PD game and Ref.…”

“…To prevent deterministic behavior of the model and allow for irrational decision making, we study the effects of noise, implemented here as a low probability with which an agent shifts from its own strategy to the strategy of a randomly chosen neighbor. Considering such noise effects in models of social dynamics [48,49,50,51] might be useful for a better understanding of the influence of bounded rationality in evolutionary snowdrift and other games [52]; see also Ref. [53] for cooperation under noisy conditions in the PD game and Ref.…”

Randomness in the evolution of cooperation

“…For example, in addition to frequently studied mechanisms for the evolution of cooperation (Nowak, 2006), more realistic aspects, such as detailed psychological mechanisms (Stevens et al, 2011) and higher level of inter-individual variation, e.g. through different types of agent migrations (Fu and Nowak, 2013;Hadzibeganovic et al, 2009;Han et al, 2014;Helbing and Yu, 2009), diversity of reproduction rates and strategyselection time scales (Rong et al, 2013;Wu et al, 2009), varied learning motivation (Zhang et al, 2010), or other random factors (McNamara et al, 2004) need to be included in next model generalizations. Moreover, the influence of individual development (Gottlieb, 2002) and social learning (Laland et al, 2000;Zhang et al, 2012) on behavioral variation and the resulting novel adaptations should more explicitly be addressed in future computational models.…”