Abstract This paper proposes an account of the interpretive effects of two discourse particles in Hungarian, talán and vajon, within the view of context and context change developed in Farkas & Roelofsen (2017), and shows that the restrictions on their distribution follow from their interpretive properties. Building on Gyuris (2022), talán will be treated as signaling epistemic bias in both declaratives and interrogatives. Following Farkas (2022), vajon will be treated as a non-intrusive question marker, which, in the account proposed, is incompatible with bias markers. The restrictions on the sentence types in which these particles occur, as well as the fact that there are restrictions on their co-occurence, will be derived from their interpretive contribution.