Swarm Lab · Experiment 01 · 叛逃细胞城
The city of freeloaders
If everyone wins individually, who keeps the city alive?
Two cities. The same tiny residents. In one, descendants tend to start together. In the other, everyone mixes. Guess which city still pays its maintenance bill.
Your challenge
Keep cooperation alive without confusing a green chart with a moral lesson.
- Play once at the defaults. Compare the two colony grids and their contributing shares.
- Keep seed 24. Raise Mixing toward 1, then Reset and play again.
- Set Public benefit to 0. Check whether the shared-founder advantage survives.
First time here? The buttons explained
Play starts or pauses. Step advances one frame; the time slider rewinds. Changing a control prepares a new run; press Play again. Reset replays the current settings. Seed chooses a random world, not difficulty; keep it while changing one rule, then try Next seed. Share saves your setup in a link. The downloads let you inspect the data.
What every parameter means
- Seed
- A repeatable random world, from 0 to 4294967295. The compared arms use the same seed.
- Mixing
- Mixing is the chance an offspring is drawn independently rather than from its colony founder. 0 keeps founder families together; 1 mixes completely. Only the second arm changes.
- Public benefit
- How much payoff everyone gets from contributing neighbors. Contributors still pay 0.6. It changes both arms.
Make a prediction. Then press Play.
Model 1.1.0 · simulation onlyGreen residents contribute; coral residents keep their share. Each small square is a colony. Population is fixed; this model measures cooperation, not survival. The dashed chart line is the first arm. Controls change both arms except the named treatment; the seed is shared.
Read the result: The vertical axis is the fraction contributing, not population growth. Each arm always contains 192 cells.
Loading local simulation…
Now break the story
Try benefit = 0. With nothing useful to maintain, should shared ancestry still rescue cooperation?
Move one control, keep the seed, and compare again. The batch below is a fixed baseline—not a live summary of your current settings.
Observed in this model · 32 matched seeds · all runs retained
A result, not just an animation.
Shared founders sustain far more cooperation under these rules. But take away the public benefit and the story changes. Assortment is doing work here; the green dots are not better people.
| Arm | Mean | Seed range |
|---|---|---|
| Shared founders | 94.2% | 81.8% – 99.0% |
| Mixed founders | 3.4% | 0.5% – 6.3% |
Ranges describe these runs, not confidence intervals for real-world systems.
Under the hood · the actual rules
Parents reproduce in proportion to payoff: max(0.05, 1 + benefit × colony cooperation − 0.6 × own contribution). Each new colony chooses one founder. A mixing draw replaces a founder-descendant with an independently sampled parent. Mutation flips the inherited choice with probability 0.015. The public-works index follows cooperation with a lag; it does not influence reproduction.
Read this model’s source · Read invariant tests · Reproduce the receipts
Where the analogy stops
Founders are an assortment mechanism, not a complete genetic model. Cells have no intentions, no politics and no moral scores. The population stays at 192: neither cancer nor an actual city is simulated.
Sources, methods & reproduction
Inspired by Robin’s diary. Background: Avida-ED · digital evolution for teaching. This is an original teaching model, not a replication of that research.
Experiment notes · Run locally with Node · Version & source hashes