Even Tiny Amounts of Altruism Can Stop Epidemics, Game Theory Study Finds
A new mathematical study has found that people don’t need to be saints to justify staying home when they’re sick. In fact, caring about others even a tiny amount — valuing your own life as roughly equivalent to 100,000 strangers — is enough to make self-isolation the rational choice during an epidemic.
The research, published Thursday in the Proceedings of the National Academy of Sciences, uses game theory to model how individuals decide whether to social distance during disease outbreaks. The central question: how much do you need to care about other people before it makes sense to stay home when you’re infectious?
A surprisingly low bar
The answer, according to researchers from the University of Warwick, the University of Tokyo, and Kyoto University, is: barely at all.
Their model identifies a tipping point in altruism below which populations inevitably reach “herd immunity” — meaning most people get infected. Above this threshold, a much better outcome emerges: “indefinite suppression,” where infected individuals voluntarily isolate and the disease is kept in check without mass infection.
The remarkable finding is how low that threshold sits. For a COVID-19-like disease, the critical altruism level is on the order of one hundred-thousandth — meaning a person who values their own well-being a hundred thousand times more than any given stranger’s would still find it rational to self-isolate.
“Individuals self-isolate in order to avoid setting off chains of infections that they would perceive as costly to them even at such small altruism,” the authors write.
Why it works
The logic hinges on chain reactions. When an infected person goes about their normal routine, they risk not just infecting one other person, but triggering a cascade of subsequent infections. Even someone who cares only slightly about the welfare of strangers recognizes that the cumulative harm of these infection chains outweighs the personal cost of staying home for a week.
The researchers tested their model against parameters for several real diseases, including COVID-19, influenza, chickenpox, and measles. In every case, the altruism threshold required for disease suppression was extremely small — ranging from about one in a thousand to one in a hundred thousand.
Robust even with cheaters and silent spreaders
A key concern with any model relying on voluntary behaviour is what happens when some people don’t cooperate. The study addresses this directly.
Even when a fraction of the population is entirely selfish — refusing to modify their behaviour at all — the disease suppression equilibrium can hold, provided enough symptomatic individuals are willing to self-isolate. The selfish fraction effectively behaves like asymptomatic carriers in the model, and as long as the symptomatic proportion exceeds a critical threshold (roughly 1 minus 1 over the basic reproduction number), suppression remains achievable.
The results also hold up when a significant proportion of infections are asymptomatic. As long as enough cases eventually show symptoms, the self-isolation behaviour of symptomatic individuals alone can keep the disease suppressed.
Lessons for policymakers
The findings carry practical implications. The study shows that two factors under policy control can lower the altruism threshold further: early detection and fast vaccine development.
Providing accurate information about a disease early — when fewer people are infected — makes it easier for even weakly altruistic populations to self-organize into suppression. And the prospect of a vaccine arriving sooner shortens the planning horizon, making the personal cost of isolation feel more manageable.
“By providing accurate and complete information about the disease to the population as early as possible, they can lower the critical altruism and therefore better enable self-organization of Indefinite Suppression even when the altruism of the population is low,” the researchers write.
A heuristic even animals could follow
Perhaps most intriguingly, the optimal behaviour identified by the model is strikingly simple: when you know you’re sick, dramatically reduce your social contacts. That’s it. Despite emerging from a complex optimization problem involving game theory and optimal control, the actual behavioural rule is something that could be communicated in a single sentence — or, the authors suggest, could evolve naturally in social animal species where individuals share some genetic relatedness.
There is already evidence of such behaviour in eusocial insects. Sick honeybees and ants have been documented leaving their colonies to die in isolation. Whether similar behaviour occurs more broadly in social animals remains an open question, though the researchers note that common sickness behaviours like lethargy may partially reflect an evolved tendency toward infectious self-isolation.
Source: Lynch, M.P., Schnyder, S.K., Molina, J.J., Yamamoto, R., & Turner, M.S. “The theory of epidemics with altruism.” Proceedings of the National Academy of Sciences, Vol. 123, No. 9 (2026). doi:10.1073/pnas.2518893123
