This study leverages sentence processing results
for relative clause attachment ambiguities in Mandarin Chinese and Korean, and incorporates them in a computational model to compare two contrasting analyses of Relative clauses.
Specifically, we compare the ability of each analysis to predict the correct sentence
processing preferences, by exploiting a top-down parser for Minimalist
grammars, connecting syntactic choices to processing difficulty.
We argue that transparent computational models grounded in rich grammar
formalisms can advise us on our theoretical choices, and strengthen the
connection between theoretical syntax and psycholinguistics results.