L1 lexical attrition is the decline of L1 lexical-semantic abilities due to reduced L1 exposure and/or L2 interference. Semantic fluency tasks are central in this research, but traditional analyses are often inconclusive. To address this, we employed an innovative network science approach to investigate the bilingual lexicon's structural properties. Semantic fluency data were collected from immersed/non-immersed late bilinguals with comparable L2 proficiencies. Our results indicate that L2 immersion led to more integrated and efficient L2 networks. Crucially, immersion also resulted in L1 attrition (only evident in the network analysis). Immersed participants exhibited less densely connected and sparser L1 networks. Furthermore, network measures suggest that L1 lexical attrition occurs gradually, initially impacting network interconnectivity, while information flow and community structure remain more stable. Drawing from these insights, we introduce the Lexical Attrition Foundation (LeAF) framework, offering a network-based perspective on lexical attrition development and laying the groundwork for future research.

中文翻译：

---

网络科学揭示 L1 词汇磨损的早期迹象：引入词汇磨损基金会 (LeAF) 框架

L1 词汇损耗是由于 L1 暴露减少和/或 L2 干扰而导致的 L1 词汇语义能力下降。语义流畅性任务是这项研究的核心，但传统分析往往没有结论。为了解决这个问题，我们采用了创新的网络科学方法来研究双语词典的结构特性。语义流利度数据是从具有可比 L2 熟练程度的沉浸式/非沉浸式晚期双语者收集的。我们的结果表明，L2 沉浸带来了更加集成和高效的 L2 网络。至关重要的是，沉浸感还会导致 L1 损耗（仅在网络分析中明显）。沉浸式的参与者表现出连接密度较低且稀疏的 L1 网络。此外，网络测量表明，L1 词汇磨损逐渐发生，最初影响网络互连性，而信息流和社区结构保持更加稳定。根据这些见解，我们引入了词汇磨损基金会（LeAF）框架，提供了基于网络的词汇磨损发展视角，并为未来的研究奠定了基础。