Cheap Talk, explained.
Cheap talk is communication that directly changes neither feasible actions nor payoffs, though messages can affect beliefs and subsequent choices.
Why it happens
A sender observes six hidden states and follows either a truthful aligned policy or an always-high policy. The receiver chooses high or low, earning a unit for matching the actual state.
Cheap talk is communication that directly changes neither feasible actions nor payoffs, though messages can affect beliefs and subsequent choices.
Read the result
Receiver payoffs are scored against the hidden states, while the two sender policies remain disclosed scripts rather than strategic equilibrium solutions.
A worked example
Evaluate a message under different incentive policies
Choose an aligned truthful sender before the six-state sequence begins.
Select high or low after each message and inspect the final state-and-payoff ledger.
Reset with the always-high sender and identify which information its message no longer distinguishes.
OPTIONAL DEEPER DETAILGo deeper: inside the model
Inside this model
A sender observes six hidden states and follows either a truthful aligned policy or an always-high policy. The receiver chooses high or low, earning a unit for matching the actual state.
Where this idea is useful
A practical use
Ask how a sender’s incentives and credibility support an informative interpretation of a message.
A common misconception
“Any misalignment makes informative communication impossible.”
Cheap-talk models can support partial information under some preferences and beliefs. This always-high bot is only a particular uninformative policy.
What this explanation leaves out
- The bots are disclosed fixed policies, not computed equilibria. Misaligned preferences do not imply that all cheap-talk equilibria are uninformative.
Is the sender learning from my responses?
No. The specified message rule is fixed throughout the round and does not adapt strategically to the receiver.
What would make this sender willing to communicate a distinction truthfully?
Associated thinkers
Associations marked provisional are awaiting source review.
Further reading
Explore the original research or the teaching reference behind this experiment.