ИСТИНА

Recent studies investigate the relation between acceptability judgements and frequencies of occurrence. Though the ability to capture the gradient acceptability is the clear benefit of the probabilistic approach (Sprouse et. al. in press), the relation between frequency/probability measures and acceptability ratings is still relatively undefined. In this paper we explore the correlation between production and perception data in the context of language variation. Our aim is to find circumstances in which the two sources of data diverge and investigate how we can interpret this inconsistency.