This research theme investigates how meaning emerges from the speaker’s choice of expressions and the addressee’s expectations about them. Languages offer different alternatives for expressing the same functional goal. The speaker’s choice concerns alternatives that may exist at every stage of linguistic production, ranging from phonetic units, intonational contours, morphological and lexical units, to syntactic constructions and semantic units of meaning. The focus of our research is on the constraints that govern the choice among alternatives. It requires a close integration of experimental and theoretical work, which is one of the strengths of the ZAS. The integration of different theoretical approaches – in particular of grammatical, game-theoretic and algorithmic models – remains a central goal, as does the diversification of our data collection methods. The theme continues the previous topic Speaker’s choice and language use. For the future, on the one hand we will strengthen the cross-linguistic perspective, and on the other hand we will intensify research on Bayesian and probabilistic modelling.