Authors: Jakub Dotlacil
Contact Email(s): j.dotlacil@gmail.com
One thought on “Cue-based retrieval model of parsing”
Leave a Comment or Question Below
You must be logged in to post a comment.
Authors: Jakub Dotlacil
Contact Email(s): j.dotlacil@gmail.com
You must be logged in to post a comment.
Hi! Thanks for the great talk! I had a question that is maybe not super well formulated yet.
I was wondering about the impact of the results for the RCs mentioned briefly towards the end.
So you co contrast was between the ACTR predictions with cue based and without cue based, right?
But then how much do you think the fact that the model without cue based performs worse just speaks to the way ACTR specifically encodes retrieval cost and the relation between activation and RTs vs the more general contribution of cue-based to parsing models? E.g. what happens if we compare with transition-based models that associate RTs to different memory types (e.g. maintenance)? You kinda touched on that mentioning the expectation based accounts, and I know it is a bit beyond the current scope but I was wondering if you had thought re: what is model specific vs what is generalizable to our theories of parsing more broadly, if it makes sense.