Authors: Stephanie Rich and Matt Wagers
Contact Email(s): skrich@ucsc.edu, mwagers@ucsc.edu
One thought on “Syntactic and semantic parallelism guides filler-gap processing in coordination”
Leave a Comment or Question Below
You must be logged in to post a comment.
Great talk, Stephanie! I want to re-phrase the question I asked now that I have gotten my thoughts together more 🙂
I guess I’ve really got two separate questions here:
(1) Have you tested whether the instrument manipulation did in fact decrease the filled-gap effect when you take all critical regions into account (rather than just looking at the N of the second conjunct [“dessert”])?
(2) Have you gamed out the predictability of the NP for the structures you use in your stimuli? In particular, two things are nagging me here:
It’s unclear to me whether the filled-gap effect in the -instr case is driven by comprehenders looking for the first available gap, but rather by the unexpectedness of repetition of the first NP (i.e., why would the speaker want to elaborate about the dessert for Dan when it was seemingly irrelevant for Carla?) Do you have a sense of how frequent these kinds of structures are?
Similarly, in the +instr condition, could the slowdown at the determiner by driven by the unexpectedness of the full NP repetition? “Carla ate the dessert with a spoon, but he wasn’t sure if Dan…” seems to be setting up a contrast, so people may be expecting a pronoun (“it”) rather than full NP repetition, which could potentially cause some processing difficulty at the determiner.
In both of these cases, you would expect some processing difficulty but it would primarily be driven purely by expectations.