Proceedings of the Annual Meeting of the Cognitive Science Society

Scalar inferences are commonly assumed to involve both lit-eral semantic interpretation and social cognitive reasoning.However, the precise way to characterize listeners’ represen-tation of context - including the space of possible utterance al-ternatives as well as the space of possible conventional mean-ings associated with linguistic forms - is a matter of ongoingdebate. We report a partial replication of a scalar inferencepriming study by Rees and Bott (2018), introducing a novelbaseline condition against which to compare behavior acrossdifferent priming treatments. We also investigate the effectof raising participants’ awareness of communicatively strongeralternatives that explicitly encode an exhaustive meaning (e.g.some but not all with respect to some). Our results suggestthat exhaustive alternatives (which are ‘symmetric’ to canoni-cal alternatives) can modulate the availability and strength ofscalar inferences, and that semantic uncertainty is an indepen-dent channel through which scalar inferences are modulated.We discuss implications for theories of pragmatic competence.