Study 1

"MALD_semrich_full_nat.rda" is the dataset with the experiment data and semantic richness predictors added to it for Study 1 (replication). This dataset is analyzed using the script "titled semrich_replication.R". The script prepares the dataset following the guidelines from Goh et al. (2016) and runs replications of hierarchical linear regressions and also a GAMM model for comparison.

Study 2

The script "MALDsem_GAMMs.R" loads the "MALD1_1_AllData.Rda" data, which is the MALD1.1 dataset that can be found here: https://doi.org/10.7939/r3-kh0r-r116. It also loads all external semantic richness predictors cited in the paper (these should be downloaded following their respective references). A final dataset used in the analyses is created and named "MALDsem.Rda". Someone interested only in analyzing this final dataset can therefore skip the data merging procedure, load "MALDsem.Rda" and start from that point in the "MALDsem_GAMMs.R" script.