Instructor: Jakub Dotlačil (j.dotlacil@gmail.com)
A majority of research in generative linguistics idealizes how grammar is embedded in general cognition. For example, while it is sometimes noted that memory limitations might play a role in syntax (e.g., in constructing phases or wh dependencies) or semantics (e.g., in anaphora resolution) it is not standard to analyze how syntax/semantics and a cognitive framework could be combined together to derive precise, quantitative predictions.
In this course, we will work on fixing this gap. We will combine a cognitive architecture, ACT-R (Adaptive Control of Thought-Rational, cf. Anderson et al, 2004), with a linguistic framework to model various experimental findings that are related to memory retrieval of linguistic objects.
On a more general level, the goal of the course is to show how linguistic analyses could be tightly connected to well-documented, carefully collected experimental data.
The course has a theoretical and a practical part. The course is followable if you only focus on the theoretical part, but you will get most out of it if you also go through the practical part. The practical part involves running Python programmes using the cognitive model pyactr. If you want to actively participate in the practical part, you have to install pyactr before the course starts. The installation procedure is described in the first chapter of Brasoveanu and Dotlacil (in prep.), pages 5 – 7 (the chapter can be downloaded here). Once more, installing this is *not needed* in order to follow the course, but you will enjoy the course more if you do this.
Tentative plan:
1: Informal and formal judgements
2: Basics of cognitive architecture, ACT-R
3: Retrieving from the lexicon: lexical decision task, reading and word retrieval (Brasoveanu and Dotlacil, in prep., Dotlačil, 2017)
4: Memory retrieval in syntax: Lewis et al., 2005, Reitter et al., 2011
5: Memory retrieval in semantics: Brasoveanu and Dotlačil, 2015
Literature:
Brasoveanu, A. and Dotlačil, J., in prep. Computing dynamic meanings. Springer. (the first chapters available here.
Dotlačil, 2017. Building an ACT-R reader for eye-tracking data. Proceedings of ICCM 2017.
Lewis and Vasishth, 2005. An activation-based model of sentence processing as skilled memory retrieval. Cognitive Science 29:1–45.
Reitter, D., Keller, F., and Moore, J. D., 2011. A computational cognitive model of syntactic priming. Cognitive science, 35(4), 587-637.