Unpacking the black box: Examining the (de)Gender categorization effect in human-machine communication
作者：
Highlights：
• This study examines the communication approach when conversing with a digital interlocutor.
• Subjects were asked to assign gender category to targets based on conversation transcripts.
• Subjects had 68.98% of chance to correctly assign gender based on the target-human talks.
• Subjects had 42.86% of chance to correctly assign gender based on the target-AI talks.
• There were significant differences between conversation approaches with human and with AI.
摘要
•This study examines the communication approach when conversing with a digital interlocutor.•Subjects were asked to assign gender category to targets based on conversation transcripts.•Subjects had 68.98% of chance to correctly assign gender based on the target-human talks.•Subjects had 42.86% of chance to correctly assign gender based on the target-AI talks.•There were significant differences between conversation approaches with human and with AI.
论文关键词：Gender categorization effect,Human-machine communication,Computer-mediated communication,Chatbot,Social media
论文评审过程：Received 1 November 2017, Revised 15 July 2018, Accepted 25 August 2018, Available online 27 August 2018, Version of Record 18 November 2018.
论文官网地址：https://doi.org/10.1016/j.chb.2018.08.049
