Rational models of information-seeking typically assume that people prefer more informative data over less informative data. This is reasonable on its surface, but it assumes that informativeness is only a property of the data, rather than a joint property of the data and a (potentially bounded) learner. We propose a model of rational information-seeking under representational constraints which simplifies complex data used to make inferences and seeks information that respects its own limitations. We find support for this account across six experiments using two strategy games: people reason imprecisely and make objectively uninformative queries in ways that are not easy to explain under standard rational accounts of information seeking, but are actually efficient given limited representational capacity. Additionally, when given a choice, participants increasingly prefer simpler, less informative queries as they gain experience, suggesting rapid adaptation to their cognitive limits. Our findings challenge theories of human information-seeking based on Optimal Experimental Design principles and instead support a more nuanced view: people are limited in what information they can represent but have information-seeking strategies that are efficient given those limits.