Source-linked AI summary
Risk-driven migration and the collective-risk social dilemma
Xiaojie Chen, Attila Szolnoki, Matjaz Perc
TL;DR
The paper asks how cooperation can be sustained when inadequate collective contributions threaten everyone’s endowments and introduces risk-driven migration as a response to that danger. In a spatial collective-risk game, movement is determined by experienced group risk rather than a manually fixed rate. The model promotes cooperation more effectively than manually determined migration and produces successful diluted cooperative patterns alongside other invasion modes.
Problem
Collective-risk dilemmas threaten all group members when contributions miss a common target, raising the question of how mobility can respond to unfavorable local risk.
Method
The paper models a collective-risk social dilemma on a spatial lattice where migration probability equals players’ average experienced group risk.
Results
Risk-driven migration promotes cooperation more effectively than fixed-rate random migration, while diluted cooperators can outperform compact clusters in some spatial processes.
Takeaways & Limitations
Self-organized movement can support cooperation under collective risk and reveals spatial patterns in which noncompact cooperative distributions succeed.
Abstract
from arXiv · showhide
A collective-risk social dilemma implies that personal endowments will be lost if contributions to the common pool within a group are too small. Failure to reach the collective target thus has dire consequences for all group members, independently of their strategies. Wanting to move away from unfavorable locations is therefore all but surprising. Inspired by these observations, we here propose and study a collective-risk social dilemma where players are allowed to move if the collective failure becomes too probable. More precisely, this so-called risk-driven migration is launched depending on the difference between the actual contributions and the declared target. Mobility therefore becomes an inherent property that is utilized in an entirely self-organizing manner. We show that under these assumptions cooperation is promoted much more effectively than under the action of manually determined migration rates. For the latter, we in fact identify parameter regions where the evolution of cooperation is incredibly inhibited. Moreover, we find unexpected spatial patterns where cooperators that do not form compact clusters outperform those that do, and where defectors are able to utilize strikingly different ways of invasion. The presented results support the recently revealed importance of percolation for the successful evolution of public cooperation, while at the same time revealing surprisingly simple ways of self-organization towards socially desirable states.
I. INTRODUCTION
The paper situates cooperation within collective-risk social dilemmas, where failing to meet a target can cause severe losses, and introduces mobility that responds to dynamically changing local risk. Risk-driven migration makes movement a self-organizing response to unfavorable environments rather than a manually imposed process.
- Collective-risk dilemma: Collective-risk dilemmas extend public-goods games by making failure to reach a declared target potentially impose severe long-term consequences on all group members.Players contribute from initial endowments to a common pool; if the target is missed, remaining endowments may be lost with a specified probability.
- Dynamic risk: The model makes group risk depend dynamically on the difference between actual contributions and the declared target, with β tuning the risk function from step-like to flat.This parameterization allows the collective-risk function to vary continuously with group performance.
- Risk-driven migration: Risk-driven migration links movement to immediate environmental risk, so players are more likely to leave locations where contributions fall farther below the target.The risk changes with spatial patterns and group composition, making mobility responsive to local conditions.
- Mobility: The study frames mobility as an inherent feature of structured populations while acknowledging that social ties, travel, and adaptation costs can inhibit migration.This motivates modeling movement as forced by unfavorable environments rather than as an explicit preference.
II. MODEL
The model places cooperators and defectors on a partially occupied square lattice, where cooperation contributes to group targets and failure creates risk. Players move according to experienced risk, then update strategies through noisy imitation in Monte Carlo simulations.
- Population structure: Players occupy or leave sites on a periodic square lattice, with fixed density ρ and equal random initialization of cooperators and defectors.Each player begins with endowment b, set to 1 without loss of generality.
- Group interaction: Cooperators contribute c from their endowment, defectors contribute nothing, and each group sets its target as T = nα based on active-player count n.The parameter α weights the collective threshold, while vacant sites allow group sizes to differ.
- Risk function: If a group misses its target, members lose their remaining endowments with probability r_i; β controls how that risk scales with the contribution shortfall.β = 0 yields a step-like risk function, whereas β = 1 gives a linear relation; the study focuses on 0 ≤ β ≤ 1.
- Risk-driven migration: After payoff accumulation, a player moves to a random empty neighboring site with probability r_m = Σ_i r_i/n, the average risk across its groups.This rule introduces no additional mobility parameter and makes movement vary across players and over time.
- Strategy updating: Strategies are updated by noisy imitation of a randomly chosen neighbor, with κ = 0.5 making better-performing strategies more likely but not certain to spread.The reported results remain qualitatively identical under a best-takes-over update rule.
III. RESULTS
Cooperation depends jointly on collective-risk parameters, donation costs, population density, and migration dynamics. The results reveal parameter-sensitive spatial patterns in which diluted cooperators can outperform compact clusters, while risk-driven migration generally supports cooperation more effectively than fixed migration.
- Parameter dependence: At c/b = 0.1, cooperation vanishes when α = 0 regardless of β, whereas higher α can support full cooperation; increasing c/b progressively narrows the cooperative region.For c/b = 0.5, cooperation survives only at large α and small β, while still larger donation ratios yield full defection.
- Parameter dependence: Increasing α raises collective-failure risk and migration, reducing exploitation of cooperators and making cooperation more likely.At small α, low migration leaves cooperators vulnerable; at larger α, dissatisfied defectors move more frequently and can destabilize defector clusters.
- Spatial patterns: At β = 0.2 and α = 0.9, diluted cooperators become successful at mixed-domain edges and ultimately invade defector domains, outperforming compact cooperative clusters.The high threshold makes compact cooperative domains resistant to drifting defectors, while diluted cooperators can expand once their local groups succeed.
- Spatial patterns: At α = 0.8, high β enables defectors to invade cooperative domains, whereas low β again favors diluted cooperators and causes compact clusters to lose.The two β values produce distinct invasion mechanisms: defectors can avoid risk at high β, while low β leaves defectors exposed to high risk.
- Spatial patterns: A stationary mixed phase emerges when diluted cooperators compete with temporarily aggregated defectors, producing dynamical coexistence.Rare successful defectors invade at pattern edges, whereas dense unsuccessful defectors are invaded.
- Population density: As population density increases, the cooperator fraction first sharply rises and then falls; full cooperation is attainable at sufficiently low densities but fails as density approaches one.At densities approaching zero, cooperators cannot form domains large enough to resist defection, so cooperation also declines.
IV. SUMMARY
The paper studies risk-driven migration in a collective-risk social dilemma, making mobility a self-organized response to dynamically changing, group-dependent failure risk. This framework reveals spatial cooperation patterns and supports self-organization toward socially desirable states.
- IV. SUMMARY: Risk-driven migration is introduced on a square lattice as a parameter-free mobility mechanism governed by players’ experienced, group-dependent risk.Risk depends on the difference between actual contributions and the declared target in each group.
- IV. SUMMARY: The model makes mobility an inherent property of the collective-risk social dilemma rather than imposing manually determined migration rates.
- IV. SUMMARY: The resulting dynamics support self-organization toward socially desirable states.