Prisoner's Dilemma

Face off against a computer in the classic iterated Prisoner’s Dilemma. Each round you choose to Cooperate (stay silent) or Defect (testify). Your goal: minimize your total years in prison over a series of rounds.

How it works

- Payoff matrix (years, lower is better):
- Both cooperate: 1 year each
- You cooperate, AI defects: You 3, AI 0
- You defect, AI cooperates: You 0, AI 3
- Both defect: 2 years each
- Play fixed-length games (5, 10, 20, 50) or Endless mode
- Practice mode for single-round learning

AI opponents

- Always Defect
- Tit-for-Tat (mirrors your last move after starting with cooperation)
- Grim Trigger (cooperates until you defect once, then defects forever)
- Pavlov / Win-Stay, Lose-Shift (repeats if last outcome was good, otherwise switches)
- Random (Biased) with adjustable defect probability

Options and polish

- Noise: optional execution error that can flip a choice (0–20%)
- Haptic feedback toggle

Progress and stats

- Running totals each game and a clear Game Over summary
- Recent games history (last results with cooperation rates)
- Aggregate counts: total games, wins, losses, draws

All gameplay is local and offline