社交线索革新:生成式社交机器人的人机互动联合效应——对微博“评论罗伯特”的修正性计算扎根与QCA分析  

Innovations in Social Cues:Joint Effects in Human-Computer Interaction with Generative Social Robots

作　　者：陈阳[1] 吕行 杜莉华 Chen Yang;Lv Xing;Du Lihua

机构地区：[1]中国人民大学新闻学院,北京100872

出　　处：《传媒观察》2025年第1期31-43,共13页Media Observer

基　　金：中国人民大学科学研究基金(中央高校基本科研业务费专项资金资助)项目成果“移动互联网时代新闻传播学研究方法体系创新研究”(23XNL016)成果之一。

摘　　要：生成式社交机器人给社交媒体平台人机交互带来了全新可能性。然而目前的研究忽视了生成式AI技术带给人机交互社交线索类型与权重的革新,以及不同维度的线索对于社交媒体中人机交互可能产生的联合效应。本研究从社交线索入手,采用修正性计算扎根与fsQCA方法,考察影响用户与微博生成式社交机器人“评论罗伯特”互动的混合效应。计算扎根结果表明,用户与生成式AI社交机器人进行人机交互主要受到用户、机器、情境、关系4个维度18类新旧社交线索的共同影响,并由此形成了支持型、抵抗型与修复型三种主要的互动模式。进一步的QCA路径分析解释了导致三类互动模式选择偏好的线索联合效应路径。本研究为重新思考生成式人工智能时代人机交互现象提供了必要的实证证据。Generative social robots open new avenues for human-computer interaction on social media platforms.However,existing research has largely overlooked innovations in the types and weights of social cues introduced by generative AI,as well as the potential joint effects of different cue dimensions on interaction dynamics.This study employs a modified computational grounded theory and fsQCA to analyze the combined effects of social cues influencing user interactions with the "Comment Robot",which is the Sina Weibo generative social bot.The findings reveal that human-computer interactions are shaped by eighteen new and traditional social cues spanning four dimensions:user,machine,situation,and relationship.These interactions manifest in three primary modes:supportive,resistant,and repairing.QCA path analysis further uncovers the joint cue effect pathways driving users' preferences for these interaction modes.This research offers empirical insights into the complexities of human-computer interaction in the generative AI era,shedding light on the nuanced role of social cues in shaping user behavior and engagement.

关 键 词：生成式社交机器人 社交线索 人机交互 计算扎根 QCA 

分 类 号：G206[文化科学—传播学]