Source-linked AI summary
Conformity enhances network reciprocity in evolutionary social dilemmas
Attila Szolnoki, Matjaz Perc
TL;DR
The paper examines whether conformity can complement payoff maximization in evolutionary social dilemmas, where cooperation is difficult to sustain. It models a population containing conformity-driven and payoff-driven players and finds that an appropriate mixture enhances network reciprocity, subject to network and payoff-normalization conditions.
Problem
Existing evolutionary-game research largely assumes that every player maximizes individual payoff, despite evidence that conformity also guides social behavior.
Method
The study models a fraction ρ of conformity-driven players who adopt the most common neighborhood strategy, alongside payoff-driven players, across lattice and scale-free networks.
Results
Conformity enhances network reciprocity: cooperative clusters become compact with smooth interfaces, and payoff-driven players break interface symmetry in favor of cooperation.
Takeaways & Limitations
Conformity can benefit the resolution of social dilemmas when its prevalence is neither too low nor too high, allowing cooperative clusters to expand beyond traditional network-reciprocity boundaries.
Takeaways & Limitations
On strongly heterogeneous networks with absolute payoffs, conformity is negligible or negative; the reported cooperation benefit requires degree-normalized payoffs.
Abstract
from arXiv · showhide
The pursuit of highest payoffs in evolutionary social dilemmas is risky and sometimes inferior to conformity. Choosing the most common strategy within the interaction range is safer because it ensures that the payoff of an individual will not be much lower than average. Herding instincts and crowd behavior in humans and social animals also compel to conformity on their own right. Motivated by these facts, we here study the impact of conformity on the evolution of cooperation in social dilemmas. We show that an appropriate fraction of conformists within the population introduces an effective surface tension around cooperative clusters and ensures smooth interfaces between different strategy domains. Payoff-driven players brake the symmetry in favor of cooperation and enable an expansion of clusters past the boundaries imposed by traditional network reciprocity. This mechanism works even under the most testing conditions, and it is robust against variations of the interaction network as long as degree-normalized payoffs are applied. Conformity may thus be beneficial for the resolution of social dilemmas.
1. Introduction
The paper asks how conformity-driven players alter cooperation in evolutionary social dilemmas, challenging the assumption that all players maximize payoff. It proposes that a suitable mixture of conformity and payoff-driven behavior can enhance network reciprocity.
- Social interactions occur on networks and may pursue belonging, group identification, or conformity rather than individual payoff maximization alone.
- In social dilemmas, mutual cooperation benefits society, but unilateral defection yields a higher individual payoff against a cooperator.
- Network reciprocity is a major existing mechanism studied for sustaining cooperation in structured populations.
- Conformists adopt the most common strategy within their interaction range, reducing individual risk and fostering population-wide coherence.
- The study asks how conformity-driven players and their population fraction affect cooperation, predicting compact cooperative clusters and smooth interfaces.
2. Evolutionary games with conformists
The study simulates evolutionary social dilemmas on homogeneous and heterogeneous networks using payoff-driven and conformity-driven strategy updates. Conformists follow neighborhood majorities, while payoff-driven players update through payoff-sensitive imitation under uncertainty.
- Simulations use a square lattice and a Barabási-Albert scale-free network, both with average degree k = 4 and size N.The two networks represent homogeneous and strongly heterogeneous interaction topologies.
- Players begin as cooperators or defectors with equal probability in weak and true prisoner’s dilemma settings.The weak dilemma uses T > 1, R = 1, and P = S = 0; the donation game uses T = 1 + b, R = 1, P = 0, and S = −b.
- Each Monte Carlo update compares a randomly selected player with a random neighbor before applying strategy adoption.A full Monte Carlo step contains N elementary updates, giving each player one average opportunity to change strategy.
- Payoff-driven adoption follows a Fermi rule in which better-performing strategies are more readily copied, while worse-performing strategies remain possible.The uncertainty parameter is K = 0.1, representing imperfect information and errors in payoff evaluation.
- A fraction ρ of players is designated conformity-driven and usually adopts whichever strategy is most common in its interaction range.For ties, a conformist changes strategy with probability 1/2; minority-strategy adoption remains possible but very unlikely.
3. Results
In structured populations, conformity creates compact cooperative clusters with smooth interfaces, while payoff-driven players break symmetry and guide their expansion. The effect remains robust across dilemma types and network topologies when degree-normalized payoffs are used.
- In well-mixed populations, defectors dominate without conformists, whereas all-conformist populations undergo neutral evolution and average fC = 0.5.
- Conformity produces a bell-shaped, nonmonotonic dependence of cooperation on ρ: moderate conformity enhances cooperation, but ρ = 1 yields neutral evolution.At ρ = 1, the population ends in all-C or all-D states with equal probability.
- Conformity-driven players form compact cooperative clusters with smooth interfaces, while payoff-driven players reveal cooperation’s long-term benefits and promote cluster growth.The spatial sequence proceeds from random initialization to flocking, smooth interfaces, and near-complete cooperative dominance.
- When payoff-driven players are rare, interface transitions can be summed into an approximation showing that cooperation spreads below a threshold even when T > R.The approximation explains the sharp transition between full-C and full-D outcomes at high ρ.
- In the donation game, conformity-enhanced network reciprocity remains qualitatively similar to the weak prisoner’s dilemma and operates under more testing conditions where ordinary network reciprocity fails for T > 1.
- On scale-free networks, results remain qualitatively unchanged with degree-normalized payoffs, whereas absolute payoffs make conformity negligible or slightly negative on strongly heterogeneous networks.Absolute-payoff heterogeneity alone can provide maximal support for network reciprocity.
4. Discussion
Conformity enhances network reciprocity when conformists are neither too rare nor too common, because it smooths interfaces around cooperative clusters while payoff-driven players break symmetry in cooperation’s favor. The mechanism remains effective across social dilemmas and interaction networks under degree-normalized payoffs, but can be negligible or negative with absolute payoffs on strongly heterogeneous networks.
- Discussion: Conformity-driven players enhance network reciprocity by adopting the locally most common strategy rather than maximizing payoff.This behavior is motivated partly by belonging and fitting in, which are important goals in interactions among humans and social animals.
- Discussion: An intermediate fraction of conformists produces compact cooperative clusters with smooth interfaces, while payoff-driven players break interface symmetry and favor cooperation.Too few or too many conformists reduce effectiveness; the payoff-driven fraction is necessary because conformity itself is strategy-neutral.
- Discussion: Cooperative clusters can expand beyond traditional network-reciprocity boundaries because conformity dynamics resemble the majority-voter model and generate effective surface tension.Larger conformist fractions produce smoother interfaces between competing strategy domains.
- Discussion: The mechanism is robust across variations of the social dilemma and can sustain cooperation under conditions where traditional network reciprocity fails.The authors report that conformity promotes cooperation regardless of interaction-network properties when degree-normalized payoffs are used.
- Discussion: With absolute payoffs on strongly heterogeneous networks, conformity may have negligible or negative effects because network heterogeneity already maximally supports reciprocity.A defective hub designated as conformity-driven may be nearly impossible to revert to cooperation, disabling the key mechanism.