Co-constructing intersubjectivity with artificial conversational agents: People are more likely to initiate repairs of misunderstandings with agents represented as human

作者：

Highlights：

• This article demonstrates a method for evaluating human-agent interaction against human–human benchmarks.

• An experiment assesses the effort people exert toward building common ground with a conversational agent.

• Believing an interlocutor is a person (vs. an agent) augments efforts to establish common ground.

• Interfacing with a human body (vs. a text screen) augments efforts to establish common ground.

摘要

•This article demonstrates a method for evaluating human-agent interaction against human–human benchmarks.•An experiment assesses the effort people exert toward building common ground with a conversational agent.•Believing an interlocutor is a person (vs. an agent) augments efforts to establish common ground.•Interfacing with a human body (vs. a text screen) augments efforts to establish common ground.

论文关键词：Common ground,Conversational repair,Echoborg,Human-agent interaction,Intersubjectivity,Psychological benchmarks

论文评审过程：Received 24 September 2015, Revised 15 December 2015, Accepted 17 December 2015, Available online 29 January 2016, Version of Record 29 January 2016.

论文官网地址：https://doi.org/10.1016/j.chb.2015.12.039