Social Teaching: Being Informative vs. Being Right in Sequential Decision Making

TL;DR

This paper explores how in sequential social learning, agents benefit from intentionally misrepresenting initial beliefs, especially when private signals are Gaussian, to improve the final decision's accuracy.

Contribution

It reveals that agents should adopt inaccurate priors to optimize collective decision-making in sequential Bayesian hypothesis testing.

Findings

Optimal initial beliefs are systematically biased towards the unlikely hypothesis.

Agents acting as if the prior is larger than reality improve overall decision accuracy.

Inaccurate priors can be more beneficial than correct ones in sequential social learning.

Abstract

We show that it can be suboptimal for Bayesian decision-making agents employing social learning to use correct prior probabilities as their initial beliefs. We consider sequential Bayesian binary hypothesis testing where each individual agent makes a binary decision based on an initial belief, a private signal, and the decisions of all earlier-acting agents---with the actions of precedent agents causing updates of the initial belief. Each agent acts to minimize Bayes risk, with all agents sharing the same Bayes costs for Type I (false alarm) and Type II (missed detection) errors. The effect of the set of initial beliefs on the decision-making performance of the last agent is studied. The last agent makes the best decision when the initial beliefs are inaccurate. When the private signals are described by Gaussian likelihoods, the optimal initial beliefs are not haphazard but rather follow a systematic pattern: the earlier-acting agents should act as if the prior probability is larger than it is in reality when the true prior probability is small, and vice versa. We interpret this as being open minded toward the unlikely hypothesis. The early-acting agents face a trade-off between making a correct decision and being maximally informative to the later-acting agents.