📄 evolutionary biology

The evolution of family reputation extends indirect reciprocity

This paper demonstrates that the evolution of family-based reputation creates a positive feedback loop between kin selection and indirect reciprocity, thereby expanding the conditions under which cooperation can be sustained, particularly when interactions are infrequent or individual behavior is difficult to observe.

Original authors: Dos Santos, M., Ohtsuki, H., Mullon, C.

Published 2026-08-29

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Dos Santos, M., Ohtsuki, H., Mullon, C.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Human beings are social creatures who rely heavily on the trust of others to get things done. When we help a stranger, we often do so with the hope that this good deed will be remembered, making it more likely that someone else will help us in the future. This mechanism, known as indirect reciprocity, is a powerful engine for cooperation among people who are not related. It works because a good reputation acts like a social currency; those who are known to be generous are trusted more and receive more support. For decades, scientists have studied how this system works, usually assuming that a person's reputation is built entirely on their own actions. If you help someone, you get a good mark; if you don't, you get a bad one.

However, in many real-world societies, people are not judged solely on their own deeds. They are also judged by the reputation of their family. A person might be trusted because their parents were known to be honest, or they might be shunned because their relatives were known to be deceitful. This phenomenon, where a family's standing influences how an individual is treated, is common across diverse cultures, from ancient feuds to modern business networks. Yet, until now, it was unclear how this kind of family-based reputation actually affects the evolution of cooperation. Does it help people work together, or does it just create unfair biases? A new study by evolutionary biologists Miguel Dos Santos, Hisashi Ohtsuki, and Charles Mullon uses computer simulations to explore this question, revealing that family reputation does more than just influence first impressions; it fundamentally changes the rules of how cooperation survives and thrives.

The researchers built a detailed computer model to simulate how people interact, make decisions, and pass on their traits to the next generation. In their virtual world, individuals play a game where they can choose to help a partner at a personal cost, or they can refuse to help. The amount of help they give depends on two things: their natural tendency to be generous, and how they respond to the reputation of the person they are helping. Crucially, the model allowed individuals to use a new piece of information: the reputation of their partner's parents. If a person had not yet been seen acting, others could look at their family's history to decide whether to trust them. The researchers then watched to see how these traits—generosity, responsiveness to reputation, and the reliance on family history—changed over thousands of generations.

What they found was that relying on family reputation creates a powerful positive feedback loop that stabilizes cooperation. In a world where people only look at an individual's own past actions, cooperation can be fragile. If interactions are rare, or if it is hard to see what people are doing, a person might start life with no reputation at all, making it difficult for them to prove they are trustworthy. In these situations, cooperation often collapses because there is no way to reward good behavior early on. But when individuals use family reputation, the story changes. A child born to a cooperative parent starts life with a good reputation, inherited from their parent's final actions. This head start means they are more likely to receive help immediately, which reinforces their own tendency to be generous.

This system creates a unique bridge between two different ways that evolution usually explains cooperation. One way is through kin selection, where we help relatives because they share our genes. The other is through reciprocity, where we help non-relatives because we expect a return favor. The study shows that family reputation links these two ideas. By helping a stranger, an individual not only builds their own reputation but also improves the reputation their children will inherit. This means that the benefits of a good deed can ripple across generations, benefiting descendants even before they have acted themselves. The researchers found that this cross-generational benefit makes it much easier for cooperation to evolve, even in difficult conditions where people interact only a few times or where it is hard to observe their behavior.

The simulations also showed that the reliance on family reputation is not just a passive trait; it evolves on its own. When cooperation is already present in a population, natural selection favors individuals who pay attention to family history. This is because using family information allows people to make better decisions about who to trust, especially when they don't know a partner's personal history. As more people start using this information, it becomes even more valuable to be a good family member, which in turn strengthens the pressure to be cooperative. The result is a self-reinforcing cycle where family reputation and personal cooperation support each other, making the whole system more robust against collapse.

The study also explored what happens when family members share a single, common reputation, a scenario seen in some societies where the actions of one sibling affect the standing of all. In these cases, the pressure to cooperate is even stronger because a single bad act can damage the reputation of the entire family. The researchers found that this shared reputation can sustain high levels of cooperation, provided that information about the family is widely available. However, if the information is scarce, relying on family reputation can become a liability, as people might make poor decisions based on incomplete data. This suggests that the way societies organize their reputational systems—whether they judge individuals or families—depends heavily on how much information is available to the community.

Ultimately, this research provides a new evolutionary framework for understanding why family-based institutions, feuds, and honor codes exist in human societies. It suggests that these systems are not just cultural quirks but are deeply rooted in the mechanics of how cooperation evolves. By extending the reach of reciprocity across generations, family reputation allows societies to maintain trust and cooperation even when direct observation of behavior is difficult. It turns out that in the complex game of human social life, your family name is not just a label; it is a vital piece of information that can determine whether you are trusted, helped, or left out, and it plays a crucial role in keeping the whole system of cooperation alive.

Technical Summary: The Evolution of Family Reputation Extends Indirect Reciprocity

Problem Statement
Reputation is a cornerstone of indirect reciprocity, enabling cooperation among unrelated individuals by allowing helpers to build a personal reputation that yields future benefits. However, standard evolutionary models of indirect reciprocity assume that an individual's reputation is based solely on their own observed behavior. In many human societies, individuals are also judged by their family's reputation, and their behavior affects the standing of their relatives. While kin selection and indirect reciprocity are often treated as distinct mechanisms, it remains unclear how family reputation influences the evolution of cooperation, under what conditions reliance on family information evolves, and how shared reputational consequences among relatives alter the stability of cooperative systems.

Methodology
The authors develop an evolutionary game theory model to investigate the coevolution of cooperative behavior, reciprocity, and reliance on family reputation.

Population Structure: A large, well-mixed population of size $N$ engages in $T$ pairwise donation games with randomly selected partners.
Traits: Individuals possess three evolving traits:
1. Baseline donation ( $\alpha$ ): The unconditional tendency to cooperate.
2. Reciprocity ( $\beta$ ): The responsiveness to a partner's reputation.
3. Reliance on family reputation ( $h$ ): The weight placed on a partner's inherited reputation when their personal behavior has not yet been observed.
Reputation Dynamics:
- Personal reputation ( $\rho$ ) follows a first-order image scoring norm, reflecting the donation made in the previous interaction.
- In the first round (or when personal reputation is unavailable), an individual $i$ evaluates partner $j$ using a weighted combination of a default reputation (zero) and the partner's inherited family reputation ( $\rho_{F,j}$ ), determined by the parent's final donation: $\rho_j = (1-h_i)\rho_{ini} + h_i\rho_{F,j}$ .
- Once a partner's personal behavior is observed, the inherited reputation is replaced by the personal reputation.
Evolutionary Mechanism: The model uses adaptive dynamics to derive selection gradients for $\alpha$ , $\beta$ , and $h$ . It decomposes selection into intragenerational effects (fitness within a lifetime) and transgenerational effects (fitness benefits to descendants via inherited reputation).
Extensions: The study includes individual-based stochastic simulations to test robustness against:
- Imperfect observability (probability $q$ that an interaction is public).
- Imperfect knowledge of family reputation (probability $s$ ).
- Execution and assessment errors.
- Alternative reputation systems where siblings share a common, dynamically updated family reputation.
- Discrete strategy models (Unconditional Defectors, Unconditional Cooperators, Discriminators, and Discriminators using Family Reputation).

Key Contributions and Results

Expansion of the Cooperative Basin:
The introduction of family reputation ( $h > 0$ ) lowers the critical reciprocity threshold ( $\beta^*$ ) required for selection to favor increased cooperation. In classical models, cooperation often collapses if initial reciprocity is too low or interactions are too few. Family reputation expands the "basin of attraction" for cooperation, allowing it to evolve and stabilize under conditions (e.g., low $T$ or low $b/c$ ) where personal reputation alone would fail.
Positive Feedback Loop:
The model identifies a positive evolutionary feedback mechanism:
- Increased reliance on family reputation ( $h$ ) improves the initial treatment of offspring, enhancing their personal reputation and future returns.
- This strengthens selection for higher reciprocity ( $\beta$ ).
- Higher reciprocity, in turn, generates better inherited reputations for descendants, further favoring higher $h$ .
  This feedback stabilizes cooperation, particularly when interactions are infrequent or personal behavior is difficult to observe.
Linking Kin Selection and Reciprocity:
The selection gradient for reciprocity is decomposed into intragenerational and transgenerational components. The transgenerational component captures how an individual's cooperative behavior benefits their descendants by improving the reputation they inherit. This effectively links indirect reciprocity with kin selection, showing that reputation inheritance provides a route where these two mechanisms reinforce one another.
Evolution of Reliance ( $h$ ):
Selection favors the evolution of reliance on family reputation ( $h$ ) when baseline donation is positive and reciprocity exceeds a specific threshold ( $\beta > c/b$ ). This evolution is most robust when personal information is scarce (low observability $q$ ) but family information is available (high $s$ ).
Robustness to Imperfect Information:
When personal behavior is imperfectly observed, family reputation compensates for the lack of data. Inherited reputation allows cooperative tendencies to be tracked across generations even when they cannot be reliably tracked within a single lifetime. The model shows that reliance on family reputation evolves and sustains cooperation even when $q$ is low, provided $s$ is sufficiently high.
Discrete Strategy Stability:
In a discrete-strategy model (including Unconditional Defectors, Cooperators, and Discriminators), the introduction of a strategy that uses family reputation (DiscH) stabilizes cooperation. Without family reputation, image-scoring models often exhibit cycles of cooperation and collapse. With DiscH, cooperation remains high and stable, as defectors cannot easily exploit the system because the reputational consequences of their actions are transmitted to their offspring.

Significance and Claims
The paper claims that family reputation is not merely a cultural artifact but an evolutionary mechanism that extends the effective horizon of indirect reciprocity. By allowing reputational consequences to cross generations, family reputation:

Stabilizes cooperation in environments where personal reputation is insufficient (e.g., rare interactions, low observability).
Reconciles kin selection and reciprocity, demonstrating how they can co-evolve to support cooperation among non-kin.
Explains the persistence of family-based institutions (e.g., feuds, honor codes, marriage alliances) across diverse human societies. The model suggests that these systems may have evolved because they provide a cognitive shortcut for evaluating partners and stabilizing social norms, even if they occasionally sustain harmful cycles of violence or collective punishment.

The authors conclude that their framework offers a unified evolutionary explanation for phenomena organized around family standing, suggesting that the inheritance of reputation is a critical factor in the social evolution of human cooperation.

More like this