Norms can grow from habit
Nobody in the model is told to trust or to share. Stable conventions still form, because people keep doing what has worked for them.
Interactive project · Roth–Erev reinforcement learning
Two players. No rules about right or wrong. Each one simply repeats whatever paid off last time. Run the simulation and watch norms like trust, fairness and “that’s mine” show up on their own, or fail to.
Round 1
Round 40
Round 600
The model comes from Roth and Erev’s work on how people actually learn in repeated games. It’s deliberately simple: no foresight and no morality, just habits that get stronger when they pay.
Every round, each player picks a strategy. Each strategy has a weight, and heavier strategies get picked more often.
Whatever a player earned that round is added to the weight of the strategy they used. Success is sticky.
Old weights fade a little each round (forgetting, δ). Now and then a player picks at random by mistake (ε).
weight ← (1 − δ) × weight + payoff
Pick a game, set the dials, press play. The grid shows how likely each pair of choices is at the current round.
Split a pie between two players. Each asks for a share. If the two requests fit, both get what they asked for; if together they ask for too much, nobody gets anything.
Payoffs row player · column player
| Row strategy | Ask ⅓ | Ask ½ | Ask ⅔ |
|---|---|---|---|
| Ask ⅓ | 2, 2 | 2, 3 | 2, 4 |
| Ask ½ | 3, 2 | 3, 3 | 0, 0 |
| Ask ⅔ | 4, 2 | 0, 0 | 0, 0 |
Round 1000 / 1000
Settled
Fair: ½ and ½
Row player
Column player
One run is a story. Many runs show the odds. Same game, same dials, 200 different strokes of early luck. Here’s where they ended up.
Philosophers like Brian Skyrms use models like this to ask how a social contract could grow out of ordinary, self-interested learning instead of being designed by anyone.
Nobody in the model is told to trust or to share. Stable conventions still form, because people keep doing what has worked for them.
The same rules can lead to different outcomes. Luck in the first few rounds decides which norm takes hold, and an unfair norm can last just as long as a fair one.
When two players split something, the 50/50 split is where they land most often. Equal treatment emerges without being imposed.
Cooperating would make everyone better off, yet it’s fragile. Caution or self-interest usually wins unless something else helps, like reputation, repeat partners or institutions.
The model explains how a norm could arise. It doesn’t tell you which norm is right. Real people also have language, culture, empathy and punishment, and none of that is in this model.
Roth & Erev, “Learning in Extensive-Form Games” (1995) · Skyrms, Evolution of the Social Contract (1996) · Skyrms, The Stag Hunt and the Evolution of Social Structure (2004)