Understanding the nonverbal behavior of socially anxious people during intimate self-disclosure

  • Authors:

  • Sin-Hwa Kang;Albert (Skip) Rizzo;Jonathan Gratch

  • Affiliations:

  • Institute for Creative Technologies, University of Southern California, Playa Vista, CA;Institute for Creative Technologies, University of Southern California, Playa Vista, CA;Institute for Creative Technologies, University of Southern California, Playa Vista, CA

  • Venue:
  • IVA'12 Proceedings of the 12th international conference on Intelligent Virtual Agents
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

This study explores the types of nonverbal behavior exhibited by socially anxious users over the course of an interview with virtual agent counselors that talked about themselves. The counselors provided self-disclosure using human back stories or computer back stories. The video data was collected from a previous study. We defined nine types of nonverbal behavior to investigate the associations between the types of nonverbal behavior and users' anxiety levels. The results of preliminary data analysis show that five features out of the nine features are positively correlated with different levels of users' anxiety in the "computer back stories" condition. These five types of nonverbal behavior are gaze aversion, moving arms and hands, constant rocking, shaking a head, and fidgeting arms and hands. There are no significant relationships between the kinds of nonverbal behavior and users' anxiety levels in the "human back stories" condition.