在中国高中生的英语学习过程中，英语写作能力的培养尤为关键。尽管在二语写作教学中，同伴反馈被认为是提升写作能力的有效方法，但是现有研究往往忽视了小组环境中复杂的互动机制，同时也少有研究以中学生作为研究对象。因此，为了填补该研究领域的空白，本研究以中国高中生作为研究对象，探究二语写作中小组同伴反馈的互动模式及其对写作修订的影响。

本研究的主要目的是探讨中国高中生在小组同伴反馈过程中的互动模式及其对写作修订的影响。研究旨在解决以下两个问题：1) 在高中生的二语写作小组同伴反馈中，可以发现哪些互动模式？2) 不同的互动模式是否会影响写作修订，特别是关于反馈采纳率和修订类型？如果是的话，有哪些影响？

采用案例研究方法，本研究以中国某高中的37名学生作为研究对象。通过两轮的小组同伴反馈活动，收集小组讨论的录音、观察笔记、学生初稿和修订稿以及半结构访谈等数据。通过转录和定性分析录音来识别互动模式，同时对学生的初稿和修订稿进行定性与定量分析，以评估同伴反馈的采纳率和修订类型。

通过分析小组互动的“平等性”和“相互性”以及互动的显著特征，本研究发现十个被试小组在两轮小组同伴反馈中呈现了总共五种互动模式：合作型、专家/新手型、轮流型、被动/被动型和主导/主导/被动型。最常见的模式为合作型，该模式以高度的平等性和相互性为特征，所有小组成员都积极参与讨论并为同伴反馈任务做出贡献。专家/新手型中，平等性较低但相互性高，核心成员通常为经验相对较少的同伴提供指导和支持。轮流型则表现出平等性高但相互性低，成员间轮流发言，缺乏有效互动和协商。被动/被动型同样表现为高平等性和低相互性，但成员们在遇到困难时往往表现沮丧或无能为力。主导/主导/被动模式的平等性和相互性均低，成员中存在多个主导者，往往会发生争论并难以达成共识。

为了探究互动模式对写作修订的影响，本研究具体分析了学生对同伴反馈的采纳以及学生写作修订的类型。分析学生的反馈采纳率并对学生的修订进行分类，有助于揭示互动模式如何影响学生对反馈的采纳和修订的类型。研究结果显示，与其他模式相比，合作型和专家/新手型表现出更高的反馈采纳率。此外，合作型表现出对表层修订的明显重视，而其他模式中的修订分类则较为相似。

本研究的意义不仅限于小组同伴反馈的领域，还为写作指导和教学实践提供了宝贵的见解。通过阐明互动模式在塑造写作修订中的作用，教育工作者可以设计更有效的同伴反馈干预措施，以满足学生的需求。此外，研究结果强调了促进合作学习环境的重要性，以促进同伴之间的积极参与和相互支持。最终，本研究为合作学习的广泛讨论做出了贡献，并突显了同伴互动在促进高中生写作改进中的重要性。

The development of English writing proficiency is particularly crucial for Chinese high school students. Although peer feedback is widely recognized as an effective method for improving writing skills in second language (L2) writing instruction, existing research often overlooks the complex interaction dynamics within group settings and rarely focuses on secondary school students. To fill this research gap, this study investigates the interaction patterns of group peer feedback (GPF) in L2 writing among Chinese high school students and their impact on writing revision.

The primary objective of this study is to explore the interaction patterns during GPF among Chinese high school students and their impact on writing revision. The study aims to address the following questions: 1) What patterns of interaction can be found in group peer feedback of L2 writing among senior high school students? 2) Do different patterns of interaction impact writing revision, specifically regarding uptake of peers’ comments and types of revision? If so, what are the impacts?

Utilizing a case study approach, this research involved 37 students from a high school in China. Data were collected through two rounds of GPF activities, including recordings of group talks, observation notes, students’ first and second drafts, and semi-structured interviews. Interaction patterns were identified through qualitative analysis of the recordings, while qualitative and quantitative analyses of the students’ drafts were conducted to evaluate the uptake of peers’ comments and the types of revision.

By analyzing the salient features as well as the “equality” and “mutuality” of group interactions, this study identified five interaction patterns in the ten participating groups: collaborative, expert/novice, alternate, passive/passive, and dominant/dominant/passive. The most common pattern was collaborative, characterized by high equality and mutuality, with all group members actively participating in discussions and providing feedback to peers. In the expert/novice pattern, there was low equality but high mutuality, with core members typically guiding and supporting less experienced peers. The alternate pattern exhibited high equality but low mutuality, with members taking turns to speak but lacking effective interaction and negotiation. Similarly, the passive/passive pattern showed high equality and low mutuality, with members often feeling frustrated or helpless when faced with difficulties. The dominant/dominant/passive pattern had low equality and mutuality, with dominant members leading to conflicts and difficulties in reaching consensus.

To investigate the impacts of distinct interaction patterns on writing revision, the study employed a multifaceted approach. The uptake of peers’ comments and the types of revision were thoroughly examined to uncover how interaction dynamics influenced the incorporation of feedback and the types of revisions made by students. The findings revealed that collaborative and expert/novice patterns were associated with higher uptake rates of peers’ comments compared to other patterns. Moreover, collaborative patterns demonstrated a notable emphasis on surface revisions, while the types of revision in other patterns appeared similar.

The implications of this study extend beyond the immediate context of GPF, offering valuable insights for writing instruction and pedagogical practices. By elucidating the role of interaction patterns in shaping writing revisions, educators can design more effective peer feedback interventions tailored to the needs of students. Additionally, the findings underscore the importance of fostering collaborative learning environments that promote active engagement and mutual support among peers. Ultimately, this study contributes to the broader discourse on peer-mediated learning and highlights the significance of interaction patterns in facilitating writing improvement among high school students.