The evolution of indirect reciprocity under action and assessment generosity

Laura Schmid, Pouya Shati, Christian Hilbe, Krishnendu ChatterjeeScientific 2021 · DOI

Abstract

Indirect reciprocity is a mechanism for the evolution of cooperation based on social norms. This mechanism requires that individuals in a population observe and judge each other's behaviors. Individuals with a good reputation are more likely to receive help from others. Previous work suggests that indirect reciprocity is only effective when all relevant information is reliable and publicly available. Otherwise, individuals may disagree on how to assess others, even if they all apply the same social norm. Such disagreements can lead to a breakdown of cooperation. Here we explore whether the predominantly studied 'leading eight' social norms of indirect reciprocity can be made more robust by equipping them with an element of generosity. To this end, we distinguish between two kinds of generosity. According to assessment generosity, individuals occasionally assign a good reputation to group members who would usually be regarded as bad. According to action generosity, individuals occasionally cooperate with group members with whom they would usually defect. Using individual-based simulations, we show that the two kinds of generosity have a very different effect on the resulting reputation dynamics. Assessment generosity tends to add to the overall noise and allows defectors to invade. In contrast, a limited amount of action generosity can be beneficial in a few cases. However, even when action generosity is beneficial, the respective simulations do not result in full cooperation. Our results suggest that while generosity can favor cooperation when individuals use the most simple strategies of reciprocity, it is disadvantageous when individuals use more complex social norms.

Introduction

Cooperation is a fundamental pillar of human society, yet explaining its persistence when individuals help those who may never reciprocate remains a central question in evolutionary biology and social science. Indirect reciprocity offers a solution: individuals cooperate to build a positive reputation, which increases their likelihood of receiving future assistance from others. This mechanism is governed by social norms—rules that dictate how individuals should be judged based on their past behavior.

Previous research identified a set of eight social norms, known as the "leading eight," which are theoretically capable of sustaining high levels of cooperation. However, these norms were primarily evaluated in idealized settings with perfect, public information. In realistic scenarios where information is noisy or private, individuals often disagree on reputations, leading to a collapse in cooperation. This study investigates whether introducing "generosity"—a form of forgiveness—can stabilize these complex social norms in imperfect environments.

Methodology

The researchers employed individual-based simulations and evolutionary game theory to evaluate two distinct forms of generosity within the framework of the "leading eight" norms:

  1. Assessment Generosity ($g_1$): The probability that an observer assigns a "good" reputation to an individual who would otherwise be judged "bad" according to the norm.
  2. Action Generosity ($g_2$): The probability that a donor chooses to cooperate with a recipient who has a "bad" reputation.
Conceptual framework of assessment and action generosity within the leading eight social norms.

The simulations modeled populations of $N=50$ and $N=90$ individuals interacting in social dilemmas. Reputations were maintained privately in image matrices, and the evolutionary success of norms was tracked through selection-mutation dynamics. The researchers explored a wide parameter space, including varying benefit-to-cost ratios ($b/c$) and levels of observation error ($\epsilon$).

Results and Discussion

The Detrimental Impact of Assessment Generosity

A primary finding of the study is that assessment generosity ($g_1$) is universally detrimental to cooperation. Because assessments are private, an individual's "forgiving" judgment is not shared by the rest of the community. This creates a divergence in reputation tracking that cascades through the population, eventually providing an opening for defectors to invade. In every simulation, increasing $g_1$ led to lower cooperation rates and increased vulnerability to invasion.

The Limited Efficacy of Action Generosity

In contrast, action generosity ($g_2$) proved slightly more robust because it involves a public act. When a donor helps a "bad" recipient, the action is visible to the entire community. While this may be viewed as a "mistake" under the strict norm, it does not create the same level of informational entropy as private reassessments. However, its benefits are highly constrained; for one specific norm ($L_7$), a small amount of action generosity (approximately 8%) slightly improved cooperation, but full cooperation was never achieved.

Robustness and Evolutionary Dynamics

The "Consistent Standing" ($L_2$) norm emerged as the most robust under noisy conditions, but its performance peaked when generosity was entirely absent ($g_1 = g_2 = 0$). The study highlights a "two-fold cost of kindness" in complex norms: a generous donor not only pays the immediate cost of helping but also risks their own reputation if the community judges them "bad" for helping a "bad" individual. Evolutionary dynamics often resulted in cycles where leading eight norms were invaded by unconditional cooperators, who were subsequently exploited by defectors.

Conclusion

While generosity and forgiveness are known to stabilize simple reciprocal strategies like "Generous Tit-for-Tat," they appear to be counterproductive for more complex social norms. The "leading eight" norms require a high degree of informational alignment to function effectively. Adding generosity introduces noise that disrupts this alignment, particularly when information is private. Future research may explore whether communication mechanisms, such as gossip or shared reputation systems, can mitigate these disagreements and allow generosity to play a more constructive role in complex indirect reciprocity systems.

Summary updated

Abstract

Indirect reciprocity is a mechanism for the evolution of cooperation based on social norms. This mechanism requires that individuals in a population observe and judge each other's behaviors. Individuals with a good reputation are more likely to receive help from others. Previous work suggests that indirect reciprocity is only effective when all relevant information is reliable and publicly available. Otherwise, individuals may disagree on how to assess others, even if they all apply the same social norm. Such disagreements can lead to a breakdown of cooperation. Here we explore whether the predominantly studied 'leading eight' social norms of indirect reciprocity can be made more robust by equipping them with an element of generosity. To this end, we distinguish between two kinds of generosity. According to assessment generosity, individuals occasionally assign a good reputation to group members who would usually be regarded as bad. According to action generosity, individuals occasionally cooperate with group members with whom they would usually defect. Using individual-based simulations, we show that the two kinds of generosity have a very different effect on the resulting reputation dynamics. Assessment generosity tends to add to the overall noise and allows defectors to invade. In contrast, a limited amount of action generosity can be beneficial in a few cases. However, even when action generosity is beneficial, the respective simulations do not result in full cooperation. Our results suggest that while generosity can favor cooperation when individuals use the most simple strategies of reciprocity, it is disadvantageous when individuals use more complex social norms.
BibTeX
@inproceedings{schmid2021evolution,
  author = {Laura Schmid and Pouya Shati and Christian Hilbe and Krishnendu Chatterjee},
  title = {{The evolution of indirect reciprocity under action and assessment generosity}},
  booktitle = {Scientific},
  year = {2021},
  doi = {10.1038/s41598-021-96932-1},
}

Loading related content...

Figure