Reanalyzing L2 Preposition Learning with Bayesian Mixed Effects and a Pretrained Language Model
This work addresses second language acquisition research by providing incremental insights into preposition learning for Chinese learners of English.
The study analyzed Chinese learners' English preposition learning using Bayesian and neural models, replicating prior frequentist results and uncovering interactions between student ability, task type, and stimulus sentences, with Bayesian methods proving most effective due to data sparsity and learner diversity.
We use both Bayesian and neural models to dissect a data set of Chinese learners' pre- and post-interventional responses to two tests measuring their understanding of English prepositions. The results mostly replicate previous findings from frequentist analyses and newly reveal crucial interactions between student ability, task type, and stimulus sentence. Given the sparsity of the data as well as high diversity among learners, the Bayesian method proves most useful; but we also see potential in using language model probabilities as predictors of grammaticality and learnability.