A decision-maker facing uncertainty can acquire costly information before choosing. Rational inattention models this: the agent selects a signal structure (how to partition the state space into distinguishable categories) that maximizes expected utility minus the mutual information cost of the signal. The output is a stochastic mapping from states to actions, where the randomness reflects deliberate imprecision — the agent doesn't pay for distinctions that don't affect the optimal action.
The paper reveals this as optimal transport. The prior distribution over states must be transported to a posterior distribution over state-action pairs, and the cost of transport is the entropic divergence. The rational inattention problem IS a regularized optimal transport problem, where the entropy regularization corresponds to the information cost.
Note (2026-11-13; read against the paper's full text): the nesting is not the one the essay describes. In the paper's formulation (its eq. 4) the problem splits into an inner matching problem and an outer marginal-choice problem: for a fixed distribution over actions, which it calls the 'bridgehead', the inner problem finds the Schrodinger bridge between that action marginal and the prior over states, and the outer problem chooses the optimal bridgehead, that is, the marginal distribution over actions. It is not an outer choice of how much to learn followed by an inner choice of the action for each realised signal. The paper presents itself as a comment: it says rational inattention is a nested regularized optimal transport problem, uses entropic optimal transport to establish the main results of Matejka and McKay (2015) and Caplin, Dean and Leahy (2019), which are for finite action sets, and extends them to arbitrary choice sets; it adds that this is not to say continuous action spaces are econometrically practical given finitely many observations; rather, it shows that it is valid to model an agent with a continuous action space as an agent with a finely discretized one. The essay's market-clearing line about bits, states and Bayes' rule, and its claim that the transport plan determines optimal attention everywhere simultaneously, are the essay's.
The nesting matters: the agent first chooses how much to learn (the transport plan), then chooses an action for each realized signal (the inner optimization). This two-stage structure makes it a nested transport problem — the outer problem selects the coupling between states and signals, the inner problem selects the action for each signal.
The transport formulation extends previous results from finite choice sets to arbitrary choice sets. When the action space is a continuum (how much to invest, where to allocate resources), the finite-choice-set theory breaks down. The optimal transport framework handles continuous actions by the same mechanism as discrete ones — the regularized transport plan determines the optimal level of attention everywhere simultaneously.
Attention as optimal transport. The currency is bits; the goods are states; the market clearing condition is Bayes' rule.