MG Parsing as a Model of Effort in Online RC Processing

Aniello De Santo

Keywords: Minimalist Grammars, Relative Clauses, Memory, Online Effort, Self-Paced Reading

TL;DR: We show how memory-based measures of effort that explicitly consider minimalist-like structure-building operations improve our ability to account for word-by-word (online) data from a large Self-paced reading dataset..

Abstract: Past research has shown that a top-down parser for Minimalist grammars [MGs; 1] captures sentence processing preferences across an array of languages and phenomena, when combined with complexity metrics connecting its behavior to memory usage [2, 3, 4, a.o.]. This approach (henceforth: MG Model) helps probe the link between generative syntactic theory and sentence processing, by offering a fully-specified theory of how fine-grained grammatical structure affects cognitive cost. While work in this framework has focused on modeling off-line asymmetries, here we show how measures of effort that explicitly consider minimalist-like structure-building operations can account for word-by-word (online) behavioral data.

Submission Number: 22