sciencebriefs
9:47in productionCh. 1 · The Origin/ 9:47 · ceiling 15 min
Evolution · Ecology

Evolutionary game theory

John Maynard Smith

Evolutionary game theory proves animals don’t need minds to outwit each other — just genes, maths, and the right kind of contest.

Evolutionary game theory is a precise, minimal framework that explains how genetically fixed behaviours persist without cognition. Its strength lies in its constraints — no rationality, no learning, no flexibility — and its explanatory reach is limited to frequency-dependent contests where inheritance and replication are well defined.

EvolutionEcologyEvolutionAnimalsBehavioural economicsEvolution across the network →
Chapters & takeaways4
  1. 1:20
    The Origin

    Evolutionary game theory began in 1973 with Maynard Smith and Price’s formalisation of the evolutionarily stable strategy in Nature.

  2. 2:44
    No Brains Required

    It replaces rational choice with inherited strategies and replicator dynamics — success depends only on birth, inheritance, and relative fitness.

  3. 4:17
    Ritual Over Ruin

    The hawk-dove game explains why most animal fights stop at display — not because they’re peaceful, but because injury is evolutionarily unstable.

  4. 5:56
    From Paper to Textbook

    The framework culminated in the 1982 book Evolution and the Theory of Games — not as speculation, but as a working mathematical toolkit.

Worth your time?

Yes. Study the whole thing.

4.5/ 5
What works
  • predicting stable strategy frequencies
  • explaining ritualised conflict
  • framing selection without intention
What does not
  • explain learning
  • model individual choice
  • address multi-level selection
Study it if
  • biologists studying frequency-dependent selection
  • modellers of adaptive systems
  • students of animal behaviour
Skip it if
  • psychologists seeking cognitive mechanisms
  • economists assuming rational agents
  • philosophers of mind
The written brief1 min read

What the work claims

That natural selection can stabilise behavioural strategies when their fitness depends on what others do — no rationality, no learning, no foresight required. A strategy is evolutionarily stable if it cannot be invaded by any rare alternative.

How it was done

Maynard Smith and Price modelled animal contests as inherited strategies, not rational choices. They used replicator dynamics to track how strategy frequencies change with fitness. They formalised the evolutionarily stable strategy (ESS) in 1973, building on Price’s verbal argument. Their first case study was the hawk-dove game, published in Nature in 1973.

What holds up

The hawk-dove model robustly explains why many animal contests escalate only to ritual, not injury. The ESS concept correctly predicts stable strategy frequencies under frequency-dependent selection. Replicator dynamics remain a foundational tool for modelling evolutionary change in strategy distributions.

What does not

It does not explain learning, plasticity, or individual decision-making. It assumes fixed genetic strategies and ignores developmental noise, spatial structure, or environmental stochasticity. It makes no claims about human behaviour, cooperation beyond pairwise contests, or multi-level selection.

Why it matters beyond the lab

It severed behavioural explanation from psychology in biology. It enabled rigorous analysis of signalling, cooperation thresholds, and arms races — from bacterial phage resistance to cancer cell competition — wherever heritable traits interact frequency-dependently.

Is it worth your time

Yes — it redefined how biologists model adaptation without invoking cognition. It remains indispensable for interpreting ritualised conflict, frequency-dependent selection, and stability in evolving populations.

Same field · Evolution4 of 69
11:26
Konrad Lorenz1903–1989Lorenz showed imprinting is a time-limited, species-specific mechanism in nidifugous birds—not a universal learning principle. His methods prioritised natural context over control, and his theory centred on innate releasing mechanisms—not neural circuits or genetics. Much of what is popularly attributed to him (e.g., human bonding, critical periods in mammals) goes beyond his verified claims.
13:00
Kin selectionW.D. Hamilton set out in two 1964 papers a mathematical condition, since called Hamilton's rule, for when a gene favouring altruistic behaviour can spread: it needs the benefit to the recipient, discounted by how closely related the two individuals are, to exceed the cost to the individual behaving altruistically. This gave a genetic explanation for puzzles such as sterile worker ants and bees helping to raise their sisters rather than reproducing themselves, and later studies, including work on red squirrels adopting orphaned pups, found behaviour lining up with the rule's prediction fairly closely. The theory, popularised for a wide audience by Richard Dawkins, became a central plank of how evolutionary biologists explain cooperation, though it has not gone unchallenged: a prominent 2010 paper by E.O. Wilson and colleagues argued the mathematics is unnecessarily complicated and proposed alternatives, prompting a rebuttal signed by well over a hundred researchers defending the original framework.
10:17
Nikolaas TinbergenTinbergen established that instinctive behaviour is triggered by discrete, measurable sensory features—and that those features can be isolated, exaggerated, and tested. His experiments with beewolves, bees, birds, and sticklebacks showed how motivation cascades through neural centres, released stepwise by innate mechanisms. His four-question framework remains foundational. But the material does not support claims about brain anatomy, human applications, or resolution of broader philosophical debates.
10:32
John Maynard Smith1920–2004John Maynard Smith transformed evolutionary biology from narrative to calculation. He did not observe organisms — he derived conditions under which strategies persist. His models are still used because they constrain what evolution *must* do, not just what it *might* do.
Up next in Science

Georg Cantor

1845–1918 · 9:44

Cantor didn’t discover infinity — he broke it into pieces you can count, compare, and order.

Play nextBack to Evolution
9:44