To join the Zoom meeting, please click https://uva-live.zoom.us/j/87471603292.
Abstract
When communicating, people adapt their linguistic representations to those by their interlocutors. Previous studies have shown that this also works at the semantic level: speakers align their interpretations of both nouns and vague, context-dependent expressions such as quantifiers and uncertainty terms to those by an interlocutor. While adaptation is known to occur by passive exposure to a given speaker’s interpretation, we hypothesize that actively interacting with an interlocutor should strengthen semantic alignment and lead to higher communicative success. In particular, asking clarification questions can be helpful to reduce the uncertainty about the context in which a certain expression is uttered. In this study, we focus on gradable adjectives big and small and show that, in line with previous findings regarding quantifiers and uncertainty terms, speakers can align their representations to those by their interlocutors when being passively exposed to them. In addition, we show that higher alignment can be achieved when people are given the possibility of asking questions, but, crucially, only informative questions that entirely eliminate the uncertainty are found to improve communicative success. In line with previous work, the ability to ask informative questions proves not to be trivial, with speakers becoming better questioners as the interaction progresses.