This study asks whether first and second language speakers would converge equally across dialect boundaries. To investigate this we ran a shadowing task in which first (American and Australian) and second (Russian) language speakers of English repeated isolated words after an American and Australian model talker. These realizations were compared to the participants’ baseline productions in a word reading task. We found that second language speakers’ baseline production generally fell between the two native-speaking groups. Second language speakers converged across more variables than did first language speakers, and all accommodated more to the American model. There was a minimal effect of speaker exposure to dialect, which was conceptualized as residence experience in America and Australia. We argue that these findings speak to second language speakers’ relative convergence flexibility in comparison with first language speakers. We explain other model, variable, and exposure effects through relative status differences between the two language varieties.

这项研究询问第一语言和第二语言的使用者是否会跨越方言边界平等地收敛。为了调查这一点，我们运行了一个影子任务，其中第一个（美国和澳大利亚）和第二个（俄罗斯）英语母语者在一个美国和澳大利亚模范演讲者之后重复孤立的单词。这些实现与参与者在单词阅读任务中的基线作品进行了比较。我们发现，第二语言使用者的基线产量通常介于两个母语群体之间。说第二语言的人比说第一语言的人收敛到更多的变量，并且都更适应美国模式。说话者接触方言的影响很小，这被概念化为在美国和澳大利亚的居住体验。我们认为，与第一语言使用者相比，这些发现说明第二语言使用者的相对收敛灵活性。我们通过两种语言变体之间的相对状态差异来解释其他模型、变量和暴露效应。