TABLE 4.

Questions asked to users who have had direct text chats with the dialogue system to evaluate the system. Q2 measures the impression of the system’s cognitive empathy and Q3 measures that of one’s affective empathy. These questions were answered on a 7-point Likert scale (1–I do not agree at all, · · ·, 7–I totally agree). Q1–Q4 are used for the main analysis, but Q5 is listed here because it is used in the interpretation of the results.

Qid	Notation	Question (To what extent do you agree with the following statements?)
Q1	TalkOwnExp	The system chatted based on its own experience
Q2	Understanding	The system understood what you were chatting about
Q3	Empathizing	The system empathized with you
Q4	Satisfaction	You satisfied with the chat conversation with the system
Q5	Reliability	The system was reliable