Bayesian Persuasion with Fact Checking:
An Experimental Analysis

Abstract

How does fact-checking influence a person’s ability to persuade another? We present a lab experiment based on a Bayesian persuasion framework in which messages are fact-checked with a positive probability. We introduce a fact-checking device that sends a signal to the receiver whenever it detects a false message, and remains silent otherwise. We find that not all subjects behave as predicted. Senders do not respond to increases in the probability of fact-checking, i.e., they neither increase nor decrease their frequency of lying when their messages have a higher probability of getting fact-checked. At the same time, receivers become increasingly credulous as the probability of fact-checking increases, even towards messages that are not checked. Lastly, we show that these results have important policy implications for regulation of misinformation, social media, and lobbying.