生成式人工智能与二语写作研究的热点、演进与前沿——基于CiteSpace的可视化分析

Hotspots, Evolution and Frontiers of Research on Generative Artificial Intelligence and Second-Language Writing—A CiteSpace-Based Visualization Analysis

摘要: 本研究借助CiteSpace文献计量工具对2016~2026年中国知网(CNIK)收录的379篇有关生成式人工智能与二语写作相关研究文献进行可视化分析。通过发文量统计、关键词共现、聚类分析、时间线图谱及突现词探测等方法，系统梳理该领域近十年的发文趋势、研究热点与演进脉络。研究发现：该领域发文量总体呈现“前期低位徘徊、近三年爆发式增长”的态势，2025年发文量达到近十年峰值；研究热点可归纳为智能写作反馈与教学应用、二语写作与差异化教育情境、写作评价与学术写作、技术支撑与教师角色转变等领域；研究演进大致经历了萌芽探索期(2016~2022年)、技术介入期(2023年)与快速发展期(2024~2026年)三个阶段，“高中英语写作”“活动理论”“人机协同”等议题正逐渐成为该领域新兴的研究前沿。本研究清晰呈现了生成式人工智能与二语写作研究在中国中文学术界的整体图景，可为该领域后续研究与教学实践提供参考。此外，本研究结合国际(英文)文献综述，初步比较了中外学界在该领域研究趋势上的异同，发现二者在写作反馈研究议题上具有较高共识，但在方法规范化程度、批判性议题关注及研究情境分布等方面存在差异。

Abstract: This study employed the CiteSpace bibliometric visualization tool to conduct a systematic analysis of 397 valid research articles on generative artificial intelligence (GenAI) and second-language (L2) writing indexed in the China National Knowledge Infrastructure (CNKI) database from 2016 to 2026. Through publication-trend statistics, keyword co-occurrence, cluster analysis, timeline mapping, and citation-burst detection, this study systematically examined the publication trends, research hotspots, and evolutionary trajectory of this field over the past decade. The findings show that the volume of publications remained low before 2023 and then grew explosively over the past three years, peaking in 2025. The research hotspots can be summarized into four core areas: intelligent writing feedback and pedagogical applications, L2 writing across differentiated educational contexts, writing evaluation and academic writing, and technological support alongside the transformation of teacher roles. The evolution of this field can be roughly divided into three stages: an embryonic exploration stage (2016~2022), a technology-intervention stage (2023), and a rapid-development stage (2024~2026), with senior high school English writing, activity theory, and human-AI collaboration emerging as new research frontiers. This study presents a clear overview of the landscape of research on generative AI and second language writing within the Chinese academic community, offering a reference for future research and pedagogical practice in the field. In addition, drawing on international English-language reviews, this study further compares domestic and international research trends, finding a high degree of consensus on AI-assisted writing feedback alongside notable differences in methodological rigor, attention to critical issues, and educational-stage coverage.