The Quantity-Competence Presumption: Children Use Informativeness to Calibrate Trust and Infer Speakers' Knowledge.

Choosing whom to learn from is critical for human learners. A well-established strategy to do so is to track informants' past accuracy. Yet this requires both a track record and independent knowledge of the truth-neither of which is available when encountering a new informant in an unfamiliar domain. This paper proposes that the amount of information an individual communicates can serve as a proxy signal for their trustworthiness, and tests this hypothesis in children. Study 1 (N = 225, ages 4-11) pitted an informant making necessarily accurate but uninformative tautological claims (e.g., "a dibleur is a dibleur") against one making informative claims of unknown accuracy (e.g., "a dibleur is a bird"). Children as young as four preferred to seek and endorse information from the more informative speaker-even though her accuracy had never been established. This preference was bounded: when the non-tautological speaker's claims were clearly inaccurate, children shifted their trust to the uninformative but accurate speaker. Study 2 (N = 159, ages 4-11) tested the cognitive bases of reliance on informative individuals. It assessed whether children apply the quantity-competence presumption: all else equal, a more informative individual implies she knows more. Children attributed greater knowledge to more informative speakers across four conditions varying in the degree and nature of the informativeness contrast. Together, these findings reveal that children treat informativeness as a graded signal of reliability, using it to bootstrap trust in unfamiliar informants when other epistemic tools are unavailable.