Every ambiguity isn't syntactic in nature: Testing the Rational Speech Act model of scope ambiguity

Pragmatic reasoning and semantic convention: A case study on gradable adjectives

“…This exercise remains a preliminary exploration of the possibility that one can potentially combine the Bayesian approach with semantic threshold conventions to capture the full pattern of human judgments. We note that the general spirit of this possibility is broadly in line with results from van Tiel et al (2021), which compared different models on their ability to capture human's production choices of quantifiers. The best model emerged from their study was the one that combined both truth-conditional semantics and probabilistic pragmatic reasoning, which outperformed those models that only contained a single module.…”

“…The mean slider ratings for each object were then normalized to derive a probability distribution over the five objects in the same object set. The study was carried out on IbexFarm using a slider controller developed in Chen & van Tiel (2021). More details of this experiment are provided in the Supplementary Material (Section S2).…”