Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 84 of the 89 references it cites were found in a registry or an index; 47 of the 100 sources its claims cite were checked for support.

Cooperation without punishment

Open access
Battu, Balaraju, Rahwan, Talal · 2023 · Scientific Reports

Abstract

A fundamental question in social and biological sciences is whether self-governance is possible when individual and collective interests are in conflict. Free riding poses a major challenge to self-governance, and a prominent solution to this challenge has been altruistic punishment. However, this solution is ineffective when counter-punishments are possible and when social interactions are noisy. We set out to address these shortcomings, motivated by the fact that most people behave like conditional cooperators-individuals willing to cooperate if a critical number of others do so. In our evolutionary model, the population contains heterogeneous conditional cooperators whose decisions depend on past cooperation levels. The population plays a repeated public goods game in a moderately noisy environment where individuals can occasionally commit mistakes in their cooperative decisions and in their imitation of the role models' strategies. We show that, under moderate levels of noise, injecting a few altruists into the population triggers positive reciprocity among conditional cooperators, thereby providing a novel mechanism to establish stable cooperation. More broadly, our findings indicate that self-governance is possible while avoiding the detrimental effects of punishment, and suggest that society should focus on creating a critical amount of trust to harness the conditional nature of its members.

Overview

CoChat Summary

TL;DR

The paper uses an evolutionary agent-based model to show that stable cooperation in a repeated public goods game can emerge without punishment when a few altruists trigger positive reciprocity among conditional cooperators, provided social interactions have moderate noise. It argues self-governance may be possible by creating trust rather than punishing free-riders.

Key findings

  • Without altruists, cooperation declines over 20 generations regardless of the initial cooperation level, resembling behavioural regularities observed in repeated public goods games.
  • With altruists and moderate noise, cooperation rises above the initial altruist share; for example, when altruists can imitate with 50% probability, cooperation levels drop from 86% to 77% under the compared conditions but remain high.
  • Noise is essential: decreasing noise lowers asymptotic cooperation; with minimal noise, cooperation equals the percentage of altruists, while maximal noise makes cooperation driven by noise and altruists.
  • In the high-cooperation condition, nearly 80% of agents adopt a conditional cooperative criterion (CCC) value of 0, whereas in the low-cooperation condition nearly 25% do so, with the majority adopting CCC values greater than 50; a critical amount of altruism is needed, and under certain noise levels this may require 20% or even 30% altruists.

Methods

  • Evolutionary agent-based model with N agents (the provided text omits the numerical value of N), a small fraction of altruists, and the rest conditional cooperators whose CCC threshold determines cooperation based on the previous round’s cooperation level and a noise parameter.
  • Agents play a repeated public goods game each generation; conditional cooperators update via pairwise comparison and payoff-biased imitation with noise and 5% mutation, while altruists may be stubborn or imitate with probability w.
  • Simulations run for 20,000 generations and are repeated 30 times; outcomes are cooperation levels across generations, asymptotic cooperation over the last 5,000 generations, and the distribution of CCC values in the 20,000th generation.

Limitations

  • Stated: the mechanism may not work in small groups, where individuals are more likely to have perfect information about role models’ payoffs and thus imitate free-riders.
  • Stated: the model does not specify the conditions under which conditional cooperation evolves endogenously; future work includes spatial networks.
  • Evident and stated: the mechanism requires a considerable number of altruists (20% or even 30% under certain noise levels) and relies on assumptions that most people are conditional cooperators, that moderate noise stabilizes cooperation, and that a few altruists can influence the majority.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41598-023-28372-y

Cites

84 of 84 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 The evolution of conditional cooperation. The x-axis represents the generation, the y-axis represents the cooperation level, and each color-coded trajectory depicts the evolutionary dynamics of cooperation under different experimental conditions, given β=0.5. a, Evolution of cooperation in the absence of altruists, given different percentages of cooperation decisions made in the first generation. b, Evolution of cooperation given different percentages of altruists present across generations when w=1.
Figure 1 The evolution of conditional cooperation. The x-axis represents the generation, the y-axis represents the cooperation level, and each color-coded trajectory depicts the evolutionary dynamics of cooperation under different experimental conditions, given β=0.5. a, Evolution of cooperation in the absence of altruists, given different percentages of cooperation decisions made in the first generation. b, Evolution of cooperation given different percentages of altruists present across generations when w=1.CC BY · PubMed Central
Figure 2 Asymptotic cooperation levels. a, The x-axis represents the percentage of altruists, the y-axis represents the average cooperation level over the last 5000 generations out of 20,000 with w=1, with each color corresponding to a different level of noise, β. b, The same as (a) but for w=0.5 instead of w=1. c, The same as (a) but while varying w (x-axis) and fixing α=30%.
Figure 2 Asymptotic cooperation levels. a, The x-axis represents the percentage of altruists, the y-axis represents the average cooperation level over the last 5000 generations out of 20,000 with w=1, with each color corresponding to a different level of noise, β. b, The same as (a) but for w=0.5 instead of w=1. c, The same as (a) but while varying w (x-axis) and fixing α=30%.CC BY · PubMed Central
Figure 3 Distribution of the conditional cooperative criterion (CCC) values. Every altruists is born into the initial population with a conditional cooperative criterion CCCi=0, while every conditional cooperator k is born with CCCi∼{0,…,N}. The figure depicts the distribution of all CCC values in the 20,000th generation for different percentages of altruists (α), different levels of noise (β), and different values of w. a, Results when α=30%, β∈{0.5,∞}, and w=1. b, The same as (a) but for w=0.5 instead of w=1. c Results when α∈{10%,30%}, β=0.5, and w=1. d, The same as (c) but for w=0.5 instead of w=1.
Figure 3 Distribution of the conditional cooperative criterion (CCC) values. Every altruists is born into the initial population with a conditional cooperative criterion CCCi=0, while every conditional cooperator k is born with CCCi∼{0,…,N}. The figure depicts the distribution of all CCC values in the 20,000th generation for different percentages of altruists (α), different levels of noise (β), and different values of w. a, Results when α=30%, β∈{0.5,∞}, and w=1. b, The same as (a) but for w=0.5 instead of w=1. c Results when α∈{10%,30%}, β=0.5, and w=1. d, The same as (c) but for w=0.5 instead of w=1.CC BY · PubMed Central

Verification

Claims the paper makes

21 Valid · 3 Not valid · 13 Not checked · 51 Own results and interpretations

61 more claims · 11 with their cited sources checked

Create a free account to see all 88 claims and to verify the 6 references still open.

See all claims — free account
  • Probably not a substantial claim. 8
  • It may state two separate claims as one. 13
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Citations

Cites

Show all 89 references

Research works printed but not linked: 1

  1. Sarkar, R. Decentralised forest governance in central himalayas: A re-evaluation of outcomes. Econ. Pol. Wkly 43, 54–61 (2008).

Cited by

See all 12

Metadata

Authors
Battu, B., & Rahwan, T.
Affiliations
New York University Abu Dhabi (AE)
Corresponding author
  • Battu, Balaraju
  • Rahwan, Talal
Venue
Scientific Reports
Published
2023-01-21
Volume
13
Issue
1
Pages
1213-1213
ISSN
2045-2322 (ISSN-L)
OpenAlex
W4317662182
Cited by
12
References
89
Citations by year
2026: 4 · 2025: 3 · 2024: 3 · 2023: 2
FWCI
5.15
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
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
cooperation, altruistic punishment, conditional cooperation, public goods game, noise, free-riding, positive reciprocity, evolutionary game theory, self-government, trust, role models, imitation, heterogeneous populations
MeSH terms
Social Interaction, Altruism, Cooperative Behavior, Game Theory, Humans, Punishment

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.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Cooperation without punishment. https://app.cochat.ai/papers/10.1038/s41598-023-28372-y
The paper
Battu, B., & Rahwan, T. (2023). Cooperation without punishment. Scientific Reports. https://doi.org/10.1038/s41598-023-28372-y

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.