Costly Signaling & Bayesian Updating Simulator Spence Game Model

Bayesian Game Insight: In cheap talk (costless badge), both Genuine (30%) and Opportunistic (70%) actors emit the signal because cost is zero. Posterior belief equals prior P(G|Signal) = 0.30, providing zero receiver precision. Raising signal cost threshold above V=40 deters opportunistic actors while genuine actors continue signaling, triggering a Separating Equilibrium with P(G|Signal) = 1.00.
Agent Population & Signal Emission
Genuine Prosocial
Opportunistic
Bayesian Receiver Updating
Prior
Posterior P(G|Signal)
Receiver Classification & Payoff Workbench Posterior P(Genuine | Signal) = 100%
Prior P(Genuine) 30.0% Population baseline
P(Signal | Genuine) 100% Genuine signal rate
P(Signal | Opportunist) 0% Incentive filter check
Posterior P(G | Signal) 100% Bayes updated belief
Receiver Expected Payoff +50.0 Per signaled interaction

RECEIVER DECISION MATRIX & BELIEF BREAKDOWN

Actor Type Prior Prob Signal Decision Receiver Posterior P(Type|Signal) Receiver Optimal Action Expected Receiver Payoff

Trace the saved watermark counting rules and fixture limits

Read the explanation

The signaling source says an agent signals exactly when payoff pool minus its assigned cost is nonnegative. With illustrative pool forty, a genuine cost twelve leaves twenty-eight; an opportunistic cost forty-five leaves minus five. Both bars show cost at twelve pixels per payoff unit, with the pool forty stated as the threshold. These deterministic branches produce separation under these inputs. They are an authored game model, not observations about actual people or a guarantee that costly signals are truthful. Bayes combines the prior type fraction with whether each type sends a signal. If both always signal and the genuine prior is point three, the posterior remains point three. If only genuine types signal it becomes one. Both bars show posterior probability at five hundred pixels per unit. When neither type signals, the source falls back to the prior instead of deriving a conditional probability from a zero-probability event. Receiver acceptance uses posterior times correct payoff plus one minus posterior times incorrect payoff, and accepts only if this sum is positive. With illustrative payoffs fifty and minus thirty, posterior one half gives expected payoff ten and posterior point eight gives thirty-four. Both bars show payoff at sixteen pixels per unit. The zero crossing is probability point three seven five. This is the source payoff rule, not a universal trust or moral decision rule.

Super generates helpful tools and automates fact-checking across the internet proactively. If you enjoyed this tool, build your own with Super and share it with a friend.