Alternative representations in formal semantics: A case study of quantifiers

How do semantic theories fit into the psychology of language more generally? A number of recent theoretical and experimental findings suggest that specifications of truth-conditions generate biases for different verification procedures. In this paper, we show how considerations of different representations of a visual scene in the semantic automata framework can generate predictions for differential working memory activation in proportional quantifier sentence verification. We present experimental results showing that different representations do impact working memory in sentence verification and that 'more than half' and 'most' behave differently in this regard.