Accelerating consensus on co-evolving networks: the effect of committed individuals

Social networks are not static but rather constantly evolve in time. One of the elements thought to drive the evolution of social network structure is homophily - the need for individuals to connect with others who are similar to them. In this paper, we study how the spread of a new opinion, idea, or behavior on such a homophily-driven social network is affected by the changing network structure. In particular, using simulations, we study a variant of the Axelrod model on a network with a homophilic rewiring rule imposed. First, we find that the presence of homophilic rewiring within the network, in general, impedes the reaching of consensus in opinion, as the time to reach consensus diverges exponentially with network size $N$. We then investigate whether the introduction of committed individuals who are rigid in their opinion on a particular issue, can speed up the convergence to consensus on that issue. We demonstrate that as committed agents are added, beyond a critical value of the committed fraction, the consensus time growth becomes logarithmic in network size $N$. Furthermore, we show that slight changes in the interaction rule can produce strikingly different results in the scaling behavior of $T_c$. However, the benefit gained by introducing committed agents is qualitatively preserved across all the interaction rules we consider.

💡 Research Summary

The paper investigates how opinion dynamics unfold on a social network that co‑evolves with its structure under homophily‑driven rewiring. Building on the classic Axelrod model, each agent possesses (F) cultural features that can take one of (q) values. At each step a randomly chosen pair of neighboring agents interacts: if their similarity exceeds a threshold (\theta), one differing feature is copied (cultural assimilation); otherwise the link is broken and the focal agent reconnects to a more similar node, implementing homophilic rewiring. Simulations are performed on Erdős‑Rényi graphs with average degree (\langle k\rangle = 6) for system sizes ranging from (N=10^2) to (10^4).

The first set of results shows that homophilic rewiring dramatically slows down global consensus. The consensus time (T_c) grows exponentially with (N), indicating that the formation of tightly knit, culturally homogeneous clusters hinders information flow between disparate groups. This confirms the intuition that homophily can sustain opinion diversity by fragmenting the network.

To counteract this slowdown, the authors introduce a fraction (p) of committed agents—nodes that hold a fixed opinion (e.g., all features set to a particular value) and never change it, though they can influence others. By gradually increasing (p), they observe a sharp transition at a critical committed fraction (p_c \approx 0.07). Below (p_c) the exponential scaling persists; above (p_c) the consensus time collapses to a logarithmic dependence on (N) ((T_c \sim \log N)). Thus, a relatively small minority of inflexible individuals can steer the entire population toward rapid agreement even on a dynamically rewiring network.

The robustness of this phenomenon is tested by altering the interaction rules in two ways: (1) weighting the probability of cultural adoption by a decreasing function of cultural distance, and (2) relaxing the rewiring condition to allow random reconnections with a small probability. Both variations preserve the qualitative transition, although the exact value of (p_c) and the prefactors of the scaling laws shift. This demonstrates that the acceleration effect of committed agents is not an artifact of a specific rule set but a generic feature of co‑evolving opinion dynamics.

The authors discuss the broader implications. In real societies, homophily‑driven network restructuring can create echo chambers that impede consensus on public issues such as health measures or climate policy. Introducing a strategically placed core of steadfast advocates—experts, policymakers, or community leaders—could dramatically reduce the time needed for a population to adopt a desired stance, even when the underlying social fabric is fluid. The paper concludes by suggesting future directions: optimizing the placement of committed agents, extending the analysis to multiple competing opinions, and validating the model against empirical data from online social platforms.