This study closely replicates Wong, Zhao, & MacWhinney (2018), who found that cognitive linguistics–inspired instruction (i.e., schematic diagram feedback) demonstrated a superiority effect over traditional instruction (i.e., rule and exemplar feedback or corrective feedback) on the translation test but not the cloze test. While the original study adopted the null hypothesis testing approach, the current study adopted Bayesian mixed effects logistic models to investigate how different variables might affect the learnability of prepositions among 81 Chinese-speaking learners of English. The research design, materials, and procedure are nearly identical to those of the original study except for an added delayed posttest. Our findings are generally consistent with the results reported in the original study, indicating that the cognitive linguistics–informed instruction demonstrates superiority effect. Furthermore, these positive learning outcomes persist over time, as evidenced by the results of the delayed posttest.

中文翻译：

---

一种贝叶斯方法，用于（重新）审视认知语言学启发教学的学习效果


这项研究紧密复制了Wong， Zhao和MacWhinney （2018），他们发现认知语言学启发的教学（即示意图反馈）在翻译测试中表现出优于传统教学（即规则和示例反馈或纠正反馈）的效果，但在完形填空测试中没有。虽然最初的研究采用了零假设检验方法，但目前的研究采用了贝叶斯混合效应 logistic 模型来研究不同变量如何影响 81 名中文英语学习者的介词可学习性。研究设计、材料和程序与原始研究几乎相同，除了增加了延迟后测。我们的研究结果与原始研究中报告的结果基本一致，表明认知语言学知情教学表现出优越效应。此外，这些积极的学习成果会随着时间的推移而持续存在，延迟后测的结果证明了这一点。