The value of conceptual knowledge

TL;DR

The paper addresses how conceptual knowledge about data-generating relationships affects the value of information in Bayesian decision problems. It develops a tractable framework that separates conceptual from statistical knowledge and derives how conceptual knowledge guides optimal signal design. Key findings show that conceptual knowledge is more valuable when states are highly reducible, the value is non-monotone in the number of signals and vanishes with infinitely many signals, and deeper knowledge generally adds value though with diminishing returns; it also shows that more concepts can reduce the needed number of signals to achieve a target welfare. These insights have practical implications for information design, human-AI knowledge allocation, and inform empirical work on teaching concept-based reasoning to improve decision outcomes.

Abstract

We study the instrumental value of conceptual knowledge when making statistical decisions. Such knowledge tells agents how unknown, payoff-relevant states relate. It is distinct from the statistical knowledge gained from observing signals of those states. We formalize this distinction in a tractable framework used by economists and statisticians. Conceptual knowledge is valuable because it empowers agents to design more informative signals. It is more valuable when states are more "reducible": when they can be explained with fewer common concepts. Its value is non-monotone in the number of signals and vanishes when agents have infinitely many signals. Agents who know more concepts can attain the same payoffs with fewer signals. This is especially true when states are highly reducible.