My research is animated by the following two questions: 'How do we, humans, convey meaningful information through the use of language?' and 'How can we build machines that do the same thing, that is, how can we build machines that mean what they say?' To address these questions, I draw on both foundational work in the cognitive sciences and the philosophy of language, and empirical studies (e.g., by collecting and analysing video recordings of human-human communication). I work on translating the insights from these areas of study into computational models of communication in which inference plays a central role.