We extend the artificial language learning experimental paradigm from psycholinguistics and apply it to pre-trained language models -- specifically, BERT (Devlin et al., 2019). We treat the model as a subject in an artificial language learning experimental setting: in order to learn the relation between two linguistic properties A and B, we introduce a set of new, non-existent, linguistic items, give the model information about their variation along property A, then measure to what extent the model learns property B for these items as a result of training. We show this method at work for degree modifiers (expressions like "slightly", "very", "rather", "extremely") and test the hypothesis that the degree expressed by modifiers (low, medium or high degree) is related to their sensitivity to sentence polarity (whether they show preference for affirmative or negative sentences or neither). Our experimental results are compatible with existing linguistic observations that relate degree semantics to polarity-sensitivity, including the main one: low degree semantics leads to positive polarity sensitivity (that is, to preference towards affirmative contexts). The method can be used in linguistics to elaborate on hypotheses and interpret experimental results, as well as for more insightful evaluation of linguistic representations in language models.

翻译：我们从精神语言学中推广人工语言学习实验模式,并将其应用到预先培训的语言模式 -- -- 具体来说,BERT(Devlin等人,2019年)。我们将该模式作为人工语言学习实验环境中的一个主题处理:为了了解两种语言属性A和B之间的关系,我们引入了一套新的、不存在的语言项目,在属性A上提供它们差异的模型信息,然后测量模型通过培训而学习这些物项的属性B的程度。我们为学位修正者(“浅度”、“非常度”、“非常度”、“偏向”、“极端”、“极端”)工作展示了这种方法,并测试了以下假设:改变者(低度、中度或高度)所表现的程度与其对判词极的敏感性相关(无论是偏爱肯定或否定的句子,还是两者都没有)。我们的实验结果与现有的语言观测结果是兼容的,这些观察结果与极度对极性敏感度有关,包括主要观察结果:低度的语义导致积极的极性敏感性(偏向肯定环境的偏爱度)。在语言模型中可以使用这种方法,在语言模型中进行更深入的判读和语言分析时使用。