Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Engaged and Affective Virtual Agents: Their Impact on Social Presence, Trustworthiness, and Decision-Making in the Group Discussion

Hanseob Kim, Bin Han, Jieun Kim, Muhammad Firdaus Syawaludin, Gerard Jounghyun Kim, Jae-In Hwang

TL;DR

The findings revealed that VA’s engagements effectively captured participants’ attention even in the group setting and enhanced group synergy, thereby facilitating more in-depth discussion and producing better consensus, and provides valuable insights for improving the VA’s behavioral design as a team member for collaborative tasks.

Abstract

This study investigates how different virtual agent (VA) behaviors influence subjects' perceptions and group decision-making. Participants carried out experimental group discussions with a VA exhibiting varying levels of engagement and affective behavior. Engagement refers to the VA's focus on the group task, whereas affective behavior reflects the VA's emotional state. The findings revealed that VA's engagements effectively captured participants' attention even in the group setting and enhanced group synergy, thereby facilitating more in-depth discussion and producing better consensus. On the other hand, VA's affective behavior negatively affected the perceived social presence and trustworthiness. Consequently, in the context of group discussion, participants preferred the engaged and non-affective VA to the non-engaged and affective VA. The study provides valuable insights for improving the VA's behavioral design as a team member for collaborative tasks.

Engaged and Affective Virtual Agents: Their Impact on Social Presence, Trustworthiness, and Decision-Making in the Group Discussion

TL;DR

The findings revealed that VA’s engagements effectively captured participants’ attention even in the group setting and enhanced group synergy, thereby facilitating more in-depth discussion and producing better consensus, and provides valuable insights for improving the VA’s behavioral design as a team member for collaborative tasks.

Abstract

This study investigates how different virtual agent (VA) behaviors influence subjects' perceptions and group decision-making. Participants carried out experimental group discussions with a VA exhibiting varying levels of engagement and affective behavior. Engagement refers to the VA's focus on the group task, whereas affective behavior reflects the VA's emotional state. The findings revealed that VA's engagements effectively captured participants' attention even in the group setting and enhanced group synergy, thereby facilitating more in-depth discussion and producing better consensus. On the other hand, VA's affective behavior negatively affected the perceived social presence and trustworthiness. Consequently, in the context of group discussion, participants preferred the engaged and non-affective VA to the non-engaged and affective VA. The study provides valuable insights for improving the VA's behavioral design as a team member for collaborative tasks.
Paper Structure (38 sections, 9 figures, 5 tables)

This paper contains 38 sections, 9 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Virtual Agents with Engaging Behavior
  4. Virtual Agents with Affective Behaviors
  5. Virtual Agent's Roles in Group Decision-Making
  6. Experiment
  7. Experiment Design
  8. Engaging and Non-Engaging behavior (E/NE)
  9. Affective and Non-Affective behavior (A/NA)
  10. Experimental Task: Group Decision-Making
  11. Experimental Set-Up and Control
  12. Virtual Agent Specification
  13. Interlocutor detection algorithm
  14. Semi-Wizard of Oz control
  15. Physical Set-up
  16. ...and 23 more sections

Figures (9)

  • Figure 1: The two main factors/traits in this "virtual agent for group discussion" design space study: (1) Engaging behavior and (2) Affective behavior. "Engaging" behavior is the VA's level of engagement in the group, expressed non-verbally. Examples include directing gaze and posture toward interlocutors while using co-speech gestures (a, b); and exhibiting listening gestures, such as a head nod and shake, in response to the interlocutor's statements (c, d). "Affective" behavior is the expression of VA's emotions through verbal and non-verbal behavior. Examples of facial expressions include neutral (e), happy (f), and angry (g); positive gestures, such as thumbs up (h) and cheering arms (i); and negative gestures, such as crossed arms (j) and a hand shrug (k). Affective behavior can also be shown by the tone of voice and emotional utterance, which may be positive or negative.
  • Figure 2: The mobile app for easily arranging and setting the priority orders of survival items as an experimental task.
  • Figure 3: Interlocutor Detection (ItD) algorithm: (a) Participant A is identified as the speaker because (c) Participant A has an open mouth and (d) sound is detected from Participant A, (b) despite both participants looking at the VA.
  • Figure 4: The left photo shows the group discussion; The right diagram depicts the control and experimental rooms.
  • Figure 5: The speaking order among two participants and the VA for each of the four phases within the single treatment. The order of PA-A and PA-B was counter-balanced.
  • ...and 4 more figures