Hedonic vs Utilitarian No.6

1. Title & Reference:

   Bickmore, T., & Cassell, J. (2005). Social dialongue with embodied conversational agents. In Advances in natural multimodal dialogue systems (pp. 23-54). Springer, Dordrecht.

2. Agent Type:

   Woman. life-sized character (EMBODIED condition). Talking with REA over a telephone

3. Theme of conversation (domain) : Real estate sales

4. Result / Implication:

   • Results indicate that subjects’ perceptions of the agent are sensitive to both interaction style (social vs. task-only dialogue) and medium.
   • We observed introverts liking and trusting REA more when she behaved consistently introverted and extroverts liking and trusting her more when she behaved consistently extroverted

5. Memo:

   • The purpose of small talk is primarily to build rapport and trust among the interlocutors

   • One strategy for effecting changes to the familiarity dimension of the relationship model is for speakers to disclose personal information about themselves – moving it into the common ground – and induce the listener to do the same. (친해지려면 비밀을 이야기 해라)

   • Subjects reported being less relaxed and assured when interacting with the embodied interface than when interacting with the text interface

   • richer media – such as face-to-face, video-, and audio-mediated communication – leads to higher trust levels than media with lower bandwidth such as text chat.

   • audio-only communication encouraged negotiators to behave impersonally, to ignore the subtleties of self-presentation, and to concentrate primarily on pursuing victory for their side.

   • face-to-face discussions were more protracted and wide-ranging while subjects communicating via audio-only kept much more to the specific issues on the agenda

   • Subjects felt that REA knew them (F=3.95; p<.06) and understood them (F=7.13; p<.05) better when she used task-only dialogue face-to-face (e reversed for phone-based interactions)

   • Task-only dialogue was more fun and less tedious when embodied, while social dialogue was more fun and less tedious on the phone.

   • extroverts trusted the system more when it engaged in small talk, while introverts were not affected by the use of small talk (introverts는 영향을 안받음)

   • the substantial difference between introverts and extroverts in the face-to-face task-only condition

   • Introverts trusted REA significantly more in the faceto-face task-only condition

   • extroverts liked REA more when she used social dialogue, while introverts liked her more when she only talked about the task

   • Passive subjects felt more comfortable interacting with REA than active subjects did, regardless of whether the interaction was face-to-face or on the phone, or whether REA used social dialogue or not

6. Remark:

   • Extrovertedness was an index composed of seven Wiggins [Wiggins, 1979] extrovert adjective items: Cheerful, Enthusiastic, Extroverted, Jovial, Outgoing, and Perky.
   • Introvertedness was an index composed of seven Wiggins [Wiggins, 1979] introvert adjective items: Bashful, Introverted, Inward, Shy, Undemonstrative, Unrevealing, and Unsparkling.