An Experimental Study of Truth-Telling in a Sender-Receiver Game

A recent experimental study of Cai and Wang (2003) on strategic information transmission games reveals that subjects tend to transmit more information than predicted by the standard equilibrium analysis. To evidence that this overcommunication phenomenon can be explained in some situations in terms of a tension between normative social behavior and incentives for lying, we show that in a simple sender-receiver game subjects incurring in costs to punish liars tell the truth more often than predicted by the equilibrium analysis whereas subjects that do not punish liars after receiving a deceptive message play equilibrium strategies. Thus, we can partition the subject pool into two groups, one group of subjects with preferences for truth-telling and another group taking into account only economic incentives.