A strong predictor of children's language is performance on non‐word repetition (NWR) tasks. However, the basis of this relationship remains unknown. Some suggest that NWR tasks measure phonological working memory, which then affects language growth. Others argue that children's knowledge of language/language experience affects NWR performance. A complicating factor is that most studies focus on school‐aged children, who have already mastered key language skills. Here, we present a new NWR task for English‐learning 2‐year‐olds, use it to assess the effect of NWR performance on concurrent and later vocabulary development, and compare the children's performance with that of an experience‐based computational model (CLASSIC). The new NWR task produced reliable results; replicating wordlikeness effects, word‐length effects, and the relationship with concurrent and later language ability we see in older children. The model also simulated all effects, suggesting that the relationship between vocabulary and NWR performance can be explained by language experience‐/knowledge‐based theories.

中文翻译：

---

模拟 2 岁儿童非单词重复表现与词汇量增长之间的关系：来自语言 0-5 项目的证据


儿童语言能力的一个强有力的预测因素是非单词重复（NWR）任务的表现。然而，这种关系的基础仍然未知。一些人认为 NWR 任务测量语音工作记忆，从而影响语言的发展。其他人则认为儿童的语言知识/语言体验会影响 NWR 的表现。一个复杂的因素是，大多数研究都集中在已经掌握了关键语言技能的学龄儿童。在这里，我们为学习英语的 2 岁儿童提出了一项新的 NWR 任务，用它来评估 NWR 表现对并发和后期词汇发展的影响，并将儿童的表现与基于经验的计算模型（CLASSIC ）。新的 NWR 任务产生了可靠的结果；复制我们在年龄较大的儿童中看到的词相似效应、词长度效应以及与并发和后期语言能力的关系。该模型还模拟了所有效果，表明词汇量和 NWR 表现之间的关系可以通过基于语言经验/知识的理论来解释。