新兴技术与教育的融合正推动教育教学走向新的变革。近年来，增强现实（Augmented Reality，简称AR）技术在教育领域展现出独特的优势，且逐渐被应用于中小学课堂的科学探究活动。随着AR合作探究活动的深入开展，一些合作问题（如角色固化、偏离主题等）逐渐涌现，因此给学生提供协作脚本以提升学生的合作效果具有重要意义。本研究进行了AR合作探究环境下的协作脚本的设计并应用于中学生的科学课堂以提升学生的合作探究体验。然而学生之间的角色互动作为小组合作的内部因素，如何与外部提供的协作脚本交互影响学生的合作探究效果值得深入探讨。因此，本研究从合作探究过程和结果两个角度探索了学生在AR合作探究环境下的角色互动模式和协作脚本的交互作用。
首先，通过观察法、访谈法，本研究进行了AR合作探究情境下的协作脚本的设计和开发，并在实验中给学生提供了社会脚本，包括（角色分工、角色交换、流程控制和角色交互），和认知脚本。之后采用准实验研究法，以青岛五中的80名七年级学生为研究对象，开展了2（有无脚本）×3（角色互动模式）的双因素实验研究。本研究采集了学生AR合作探究的视频数据和音频数据，并对其进行角色交互模式编码，通过聚类分析，将学生的角色交互模式分为三类：平等互换型、一人主导型和被动消极型。最后，对于角色交互模式和协作脚本的交互效应的验证，本研究收集了学生的前后测科学知识测试、实验单成绩，并结合视频、音频数据从结果和过程两个角度进行了分析。合作结果方面，双因素方差分析的结果表明角色交互模式和协作脚本之间不存在交互效应，但各自存在显著的主效应。合作过程方面，本研究对学生的合作探究表现行为进行了编码，采用滞后序列分析法分析学生的行为模式，进一步揭示了两个因素对学生行为模式的影响。
本研究发现，角色交互模式对学生AR合作探究效果的影响存在主效应，不同交互模式的学生的合作探究成果和过程存在显著差异。平等互换型模式下成员具有平等的话语地位，能够深入探究；一人主导模式缺乏任务协调，但并不影响学生的合作探究成果；被动消极模式缺乏深入探究，交互较为被动，较多偏离主题。此外，协作脚本也对学生AR合作探究效果的影响存在主效应。协作脚本能够促进合作探究的知识习得、提升合作体验，降低认知负荷；促进探究的深入和深层的知识建构；通过任务协调引发频繁的角色转换，使得平等互换型模式更多；影响合作氛围和任务意识，使得交互更加密切。除此之外，本研究还发现协作脚本的作用会随着时间逐渐消退。
本研究关注AR合作探究的合作过程，对学生的角色互动、合作探究过程、成果进行分析，总结出协作脚本的设计策略，为教师有效合理开展AR教学活动提供教学策略方面的建议，帮助学生提升AR合作探究的体验，对AR教学具有实践指导意义。
 

The integration of emerging technologies and education is driving a new wave of educational reform. Augmented Reality (AR) technology has exhibited unique advantages in the education sector and has gradually been applied to science exploration activities in primary and secondary school classrooms. However, as AR collaborative inquiry activities continue to evolve, some collaboration problems such as role solidification and deviation from the theme have emerged, making it important to provide students with collaborative scripts to improve collaboration outcomes. This study designd collaborative scripts in an AR collaborative inquiry environment and applied them to science classrooms for secondary school students to improve their collaborative inquiry experience. However, students’ role interaction in group collaboration is an internal factor, and how it interacts with the collaborative scripts provided externally deserves further investigation. Therefore, this study explores the interaction effects of role interaction pattern and collaborative scripts among students in an AR collaborative inquiry environment from the perspectives of the collaborative process and results.

Firstly, using observation and interview methods, this study designed and developed social scripts (including role distribution, role exchange, process control, and role interaction) and cognitive scripts for AR collaborative inquiry. Then, a quasi-experimental design was employed with 80 seventh-grade students from Qingdao No.5 Middle School as the participants to investigate the interaction effects between role interaction patterns and collaborative scripts on AR collaborative inquiry. The experiment was conducted with a 2 (with or without scripts) × 3 (role interaction patterns) factorial design. Video and audio data of the students' AR collaborative inquiry were collected and analyzed using cluster analysis to categorize the role interaction patterns into three types: equal exchange pattern, one-person domination pattern, and passive negativity pattern. Finally, in terms of the interaction effects between role interaction patterns and collaborative scripts, this study analyzed the students' scientific knowledge tests, group worksheet scores, and behavior performance in terms of both learning results and processes, and found that there was no interaction effect between the two factors, but each had a significant main effect on both the results and processes of the collaborative inquiry. In terms of the collaboration process, this study coded the students' collaborative inquiry performance behavior, and analyzed the students' behavior patterns by using the lag sequence analysis method, which further revealed the influence of two factors on the students' behavior patterns.

This study found that the influence of role interaction patterns on the effectiveness of collaborative AR exploration among students is significant, and there are significant differences in the collaborative exploration outcomes and processes of students in different role interaction patterns. In the equal exchange pattern, members have equal speaking status and can conduct in-depth exploration. The one-person domination pattern lacks task coordination but does not affect the collaborative inquiry outcomes of students. The passive negative pattern lacks in-depth exploration, and the interaction is more passive and often deviates from the main topic. In addition, collaborative scripts also have a significant impact on the effectiveness of collaborative AR exploration among students. Collaborative scripts can promote knowledge acquisition and collaboration experience, reduce cognitive load, promote in-depth exploration and deep knowledge construction, and trigger frequent role transitions through task coordination, leading to more equal exchange patterns. Collaborative scripts also affect the collaborative atmosphere and task awareness, resulting in more close interactions. Furthermore, this study found that the effect of collaborative scripts gradually diminishes over time.

This study focuses on the collaborative process of AR inquiry and analyzed students' role interaction, collaborative processes, and outcomes and summarized the design strategies of collaborative scripts. The study provided insights into the design of collaboration scripts for effective AR teaching practices and offered practical guidance for enhancing students' AR collaborative inquiry experiences.