CoChat

Small groups and long memories promote cooperation

Open access
Stewart, Alexander J., Plotkin, Joshua B. · 2016 · Scientific Reports

The DOI matches its Crossref record; 44 of the 45 references it cites were found in a registry or an index.

Abstract

Complex social behaviors lie at the heart of many of the challenges facing evolutionary biology, sociology, economics, and beyond. For evolutionary biologists the question is often how group behaviors such as collective action, or decision making that accounts for memories of past experience, can emerge and persist in an evolving system. Evolutionary game theory provides a framework for formalizing these questions and admitting them to rigorous study. Here we develop such a framework to study the evolution of sustained collective action in multi-player public-goods games, in which players have arbitrarily long memories of prior rounds of play and can react to their experience in an arbitrary way. We construct a coordinate system for memory-m strategies in iterated n-player games that permits us to characterize all cooperative strategies that resist invasion by any mutant strategy, and stabilize cooperative behavior. We show that, especially when groups are small, longer-memory strategies make cooperation easier to evolve, by increasing the number of ways to stabilize cooperation. We also explore the co-evolution of behavior and memory. We find that even when memory has a cost, longer-memory strategies often evolve, which in turn drives the evolution of cooperation, even when the benefits for cooperation are low.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/srep26889

Cites

44 of 44 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Multiplayer games and memory. We study the evolution of behavior in iterated n-player public-goods games in which players use strategies with memory capacity m. We consider a replicating population of N individuals who each receive a payoff from engaging in an infinitely iterated game with all possible subsets of (n − 1) opponents in the population. Players then reproduce according to a “copying process”, in which a player X copies another player’s strategy Y with a probability where SX and SY are the player’s respective payoffs and s scales the strength of selection. We consider the case of strong selection, such that a rare mutant who is at a selective disadvantage is quickly lost from the population31. We investigate the success of cooperative strategies as a function of group size and the length of players’ memories. We determine the frequency of robust cooperative strategies, which can resist invasion by any possible mutant. (Top) Depending on the size of the game n relative to the population N, the dynamics of public-goods games are different. In a two-player game, a series of pairwise interactions occur in the population at each generation (left). If the whole population plays the game each generation (right) all players interact simultaneously. (Bottom) Memory of past events results in strategies that update behavior depending on the histories of both players’ actions. This allows for more complex strategies, such as those that punish rare defection or reward rare cooperation.
Figure 1 Multiplayer games and memory. We study the evolution of behavior in iterated n-player public-goods games in which players use strategies with memory capacity m. We consider a replicating population of N individuals who each receive a payoff from engaging in an infinitely iterated game with all possible subsets of (n − 1) opponents in the population. Players then reproduce according to a “copying process”, in which a player X copies another player’s strategy Y with a probability where SX and SY are the player’s respective payoffs and s scales the strength of selection. We consider the case of strong selection, such that a rare mutant who is at a selective disadvantage is quickly lost from the population31. We investigate the success of cooperative strategies as a function of group size and the length of players’ memories. We determine the frequency of robust cooperative strategies, which can resist invasion by any possible mutant. (Top) Depending on the size of the game n relative to the population N, the dynamics of public-goods games are different. In a two-player game, a series of pairwise interactions occur in the population at each generation (left). If the whole population plays the game each generation (right) all players interact simultaneously. (Bottom) Memory of past events results in strategies that update behavior depending on the histories of both players’ actions. This allows for more complex strategies, such as those that punish rare defection or reward rare cooperation.CC BY · PubMed Central
Figure 2 A co-ordinate system for describing strategies in public-goods games. We consider the space of strategies of the form , such that players cooperate with a probability that depends on the number of times lo her opponents have cooperated and the number of times lp she has cooperated within her memory. We define the strategy of a focal player by coordinates as shown in the figure. The components of this coordinate system have an intuitive interpretation: the probability that a player cooperates depends on (1) her past tendency to cooperate, (2) a baseline tendency to cooperate (κ), (3) a tendency to cooperate in proportion to her own payoff (χ), (4) a tendency to punish (i.e. defect) in proportion to her opponents’ payoffs (ϕ) and (5) a contingent punishment that depends on the specific outcome of the game over the prior m rounds ().
Figure 2 A co-ordinate system for describing strategies in public-goods games. We consider the space of strategies of the form , such that players cooperate with a probability that depends on the number of times lo her opponents have cooperated and the number of times lp she has cooperated within her memory. We define the strategy of a focal player by coordinates as shown in the figure. The components of this coordinate system have an intuitive interpretation: the probability that a player cooperates depends on (1) her past tendency to cooperate, (2) a baseline tendency to cooperate (κ), (3) a tendency to cooperate in proportion to her own payoff (χ), (4) a tendency to punish (i.e. defect) in proportion to her opponents’ payoffs (ϕ) and (5) a contingent punishment that depends on the specific outcome of the game over the prior m rounds ().CC BY · PubMed Central
Figure 3 The impact of group size and memory capacity on cooperation. We calculated the relative volumes of robust cooperation – that is, the absolute volume of robust cooperative strategies divided by the total volume of robust cooperators and defectors – and compared this to the relative volume of defectors (solid lines) by numerically integrating Eqs. 2, 3. We also verified these analytic results by randomly drawing 106 cooperative and 106 defecting strategies and determining their success at resisting invasion from 105 random mutants with the same memory (points). We calculated player’s payoffs by simulating 2 × 103 rounds of a public-goods game. We then plotted the relative volumes of robust cooperators and robust defectors as a function of group size n (with fixed memory m = 1, left) and as a function of memory capacity m (with fixed group size n = 2, right). Increasing group size increases the relative volume of robust defection; while increasing memory length increases the relative volume of robust cooperation. In all calculations and simulations we used cost C = 1 and benefit B as indicated in the figure.
Figure 3 The impact of group size and memory capacity on cooperation. We calculated the relative volumes of robust cooperation – that is, the absolute volume of robust cooperative strategies divided by the total volume of robust cooperators and defectors – and compared this to the relative volume of defectors (solid lines) by numerically integrating Eqs. 2, 3. We also verified these analytic results by randomly drawing 106 cooperative and 106 defecting strategies and determining their success at resisting invasion from 105 random mutants with the same memory (points). We calculated player’s payoffs by simulating 2 × 103 rounds of a public-goods game. We then plotted the relative volumes of robust cooperators and robust defectors as a function of group size n (with fixed memory m = 1, left) and as a function of memory capacity m (with fixed group size n = 2, right). Increasing group size increases the relative volume of robust defection; while increasing memory length increases the relative volume of robust cooperation. In all calculations and simulations we used cost C = 1 and benefit B as indicated in the figure.CC BY · PubMed Central
Figure 4 Co-evolution of strategies and memory capacity. We simulated populations playing the iterated n-player public-goods game, proposing mutant strategies until reaching equilibrium, and then also proposing mutations to a player’s memory capacity m, each at rate μ/10. In these simulations all players initially have memory m = 1, with payoff parameters C = 1 and B = 1.2. Mutations to strategies were drawn uniformly from the full space of memory-m strategies. Mutations perturbing the memory m caused it to increase or decrease by 1, with a lower bound of m = 1. Evolution was modeled according to a copying process under weak mutation31 in a population of size N = 10 individuals. (a) When the group size is small, n = 2, defecting strategies are initially dominant in the population, but they are quickly replaced by cooperators as memory capacity evolves to higher values. (b) When group size is large, n = N = 10, defecting strategies initially dominate the population and they remain dominant as memory evolves. In both (a,b) the overall frequency of cooperators and defectors (that is, the fraction of robust cooperative (and defecting) strategies among all cooperative and (defecting) strategies) decline as the dimension of strategy space increases, in line with the decline in the overall volume of robust strategies (Fig. S4). (c) When the game size is small memory evolves rapidly to larger values, reflecting the greater success of longer-memory strategies at invading (Fig. S3), and driving the increase in cooperative as compared to defecting strategies. (d) When the group size is large memory does not evolve to large values, reaching only m = 2 across 50,000 generations, and reflecting the decline in long-memory strategies’ success as invaders in larger games. (e) As cooperation increases so does the average payoff of the population, by a factor of 5–10 fold. (f) The lack of increase in cooperation results in a much more modest (although still appreciable) increase in average payoff for the population as defectors become less frequent.
Figure 4 Co-evolution of strategies and memory capacity. We simulated populations playing the iterated n-player public-goods game, proposing mutant strategies until reaching equilibrium, and then also proposing mutations to a player’s memory capacity m, each at rate μ/10. In these simulations all players initially have memory m = 1, with payoff parameters C = 1 and B = 1.2. Mutations to strategies were drawn uniformly from the full space of memory-m strategies. Mutations perturbing the memory m caused it to increase or decrease by 1, with a lower bound of m = 1. Evolution was modeled according to a copying process under weak mutation31 in a population of size N = 10 individuals. (a) When the group size is small, n = 2, defecting strategies are initially dominant in the population, but they are quickly replaced by cooperators as memory capacity evolves to higher values. (b) When group size is large, n = N = 10, defecting strategies initially dominate the population and they remain dominant as memory evolves. In both (a,b) the overall frequency of cooperators and defectors (that is, the fraction of robust cooperative (and defecting) strategies among all cooperative and (defecting) strategies) decline as the dimension of strategy space increases, in line with the decline in the overall volume of robust strategies (Fig. S4). (c) When the game size is small memory evolves rapidly to larger values, reflecting the greater success of longer-memory strategies at invading (Fig. S3), and driving the increase in cooperative as compared to defecting strategies. (d) When the group size is large memory does not evolve to large values, reaching only m = 2 across 50,000 generations, and reflecting the decline in long-memory strategies’ success as invaders in larger games. (e) As cooperation increases so does the average payoff of the population, by a factor of 5–10 fold. (f) The lack of increase in cooperation results in a much more modest (although still appreciable) increase in average payoff for the population as defectors become less frequent.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 2

  1. Nowak, M. A. Evolutionary dynamics: exploring the equations of life (Belknap Press of Harvard University Press, Cambridge, Mass., 2006).
  2. Sigmund, K. The calculus of selfishness Princeton series in theoretical and computational biology (Princeton University Press, Princeton, 2010).

Cited by

See all 78

Metadata

Authors
Stewart, A. J., & Plotkin, J. B.
Affiliations
University of Pennsylvania (US)
Corresponding author
Stewart, Alexander J.
Venue
Scientific Reports
Published
2016-06-01
Volume
6
Issue
1
Pages
26889-26889
ISSN
2045-2322 (ISSN-L)
OpenAlex
W2261291220
Cited by
78
References
45
Citations by year
2026: 8 · 2025: 6 · 2024: 6 · 2023: 7 · 2022: 5 · 2021: 6 · 2020: 5 · 2019: 7 · 2018: 15 · 2017: 9 · 2016: 4
FWCI
33.28
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • UCL Discovery (University College London) (submittedVersion)
  • St Andrews Research Repository (St Andrews Research Repository) (submittedVersion, cc-by)
Funders
  • David and Lucile Packard Foundation (W911NF-12-1-0552)
  • U.S. Department of the Interior (D12AP00025)
  • Royal Society
  • Army Research Office (W911NF-12-1-0552, W911NF)
Topics
  • Social Sciences › Sociology and Political Science › Evolutionary Game Theory and Cooperation
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Experimental Behavioral Economics Studies
  • Social Sciences › Decision Sciences › Management Science and Operations Research › Game Theory and Applications
Keywords
evolution of cooperation, repeated games, public goods game, multiplayer games, small group size, collective action, evolutionary stability
MeSH terms
Cooperative Behavior, Decision Making, Game Theory, Games, Experimental, Humans, Interpersonal Relations, Memory, Models, Psychological, Models, Statistical, Sociobiology

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.