Evolution of Coordination in Social Networks: A Numerical Study

Coordination games are important to explain efficient and desirable social behavior. Here we study these games by extensive numerical simulation on networked social structures using an evolutionary approach. We show that local network effects may promote selection of efficient equilibria in both pure and general coordination games and may explain social polarization. These results are put into perspective with respect to known theoretical results. The main insight we obtain is that clustering, and especially community structure in social networks has a positive role in promoting socially efficient outcomes.

💡 Research Summary

The paper investigates how the topology of social networks influences the evolutionary dynamics of coordination games. Using an agent‑based simulation framework, the authors study two game classes: a pure coordination game where one strategy yields a Pareto‑optimal payoff and the other is strictly inferior, and a general coordination game with multiple Nash equilibria that differ in efficiency. Strategies evolve according to a replicator rule combined with a small mutation probability, allowing occasional random strategy switches. Four network structures are examined: (1) Erdős‑Rényi random graphs, (2) regular two‑dimensional lattices, (3) Watts‑Strogatz small‑world networks characterized by high clustering and short average path length, and (4) hierarchical modular networks with clearly defined communities. All networks contain roughly 10,000 nodes and have the same average degree to ensure comparability.

Simulation runs span 10,000 generations, during which each node updates its strategy by sampling the payoffs of its neighbors and adopting a neighbor’s strategy with probability proportional to its relative fitness. The authors systematically vary the selection intensity (β) and mutation rate (μ) to assess robustness. Results show that on random graphs and lattices, inefficient strategies either dominate or coexist with efficient ones, leading to a lack of clear convergence. In contrast, small‑world networks exhibit a pronounced “cluster effect”: once a few tightly‑connected clusters adopt the efficient strategy, the high local homogeneity accelerates its spread to adjacent nodes. This effect becomes especially strong when the clustering coefficient exceeds about 0.4, indicating an apparent critical threshold for rapid diffusion.

The most striking findings arise in modular networks. Within each community, efficient strategies quickly become fixed, but sparse inter‑community links allow different communities to settle on different equilibria, producing a macro‑level polarization: some modules coordinate on the efficient outcome while others remain stuck in the inefficient one. Increasing the density of inter‑community edges eventually triggers a global transition to the efficient equilibrium, highlighting the role of bridge ties in overcoming local lock‑ins.

Sensitivity analysis reveals that very low mutation rates (μ < 10⁻⁴) make the system highly path‑dependent on initial conditions, whereas moderate mutation rates (10⁻³ – 10⁻²) enable occasional “innovation” events where an efficient strategy emerges in a peripheral community and spreads network‑wide. Higher selection intensity (β > 1) sharpens the transition, allowing a small seed of efficient players to trigger a cascade.

The authors relate these empirical observations to existing theoretical work on network‑based selection and structural equilibria, noting that their simulations provide quantitative boundaries for when clustering and modularity promote socially optimal outcomes versus when they entrench suboptimal coordination and foster polarization. Policy implications are discussed: designing social platforms that encourage moderate clustering can boost cooperative behavior, but excessive compartmentalization may exacerbate societal divides. The study thus bridges numerical experimentation with theoretical insights, underscoring the pivotal influence of network architecture on the evolution of coordination.