This repository contains scripts for the analyses reported in
Divjak, D., Milin, P., Ez-Zizi, A., Józefowski, J., and Adam, C. (2020). What is learned from exposure: an error-driven approach to productivity in language. To appear in Language, Cognition and Neuroscience.
The paper-package is split into R script files which are hosted here on GitHub and the supporting datasets in rda and csv formats which are available at the University of Birmingham UBIRA eData platform: UBIRA eData link.
The study analyses are split in three parts corresponding to the paper's subsections: A corpus-based model of -a/-u variation (Section 3), A discriminative computational model (Section 4), and Experimental verification (Section 5).
For further inquiries contact: p.milin@bham.ac.uk