Empathy in Virtual Agents: How Emotional Expressions can Influence User Perception

Loading...
Thumbnail Image
Date
2024
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association
Abstract
With advancements in natural language processing, intelligent virtual agents (IVAs) are increasingly integrated into various sectors like education, customer service, personal assistance, and healthcare. Medical counseling and digital therapy, fields that require trusting relationships between patient and practitioner, profit immensely from the use of IVAs. A key component of these social relationships is empathy, which helps build trust and understanding. This paper investigates whether simulated empathy through emotional expressions in IVAs can improve interactions and influence users' emotional contagion. Additionally, it explores the correlation between self-reported empathy and users' expressiveness. Participants alternate reading an emotional story with a virtual agent (VA) which mirrors the users' emotional expressions in one condition, while remaining neutral in the baseline condition. The results show that simulated emotions can animate participants to elicit more facial expressions in response to the VA's, while correlating with the users' self reported empathy.
Description

CCS Concepts: Human-centered computing → Human computer interaction (HCI); Collaborative interaction; User studies; Applied computing → Consumer health

        
@inproceedings{
10.2312:egve.20241361
, booktitle = {
ICAT-EGVE 2024 - International Conference on Artificial Reality and Telexistence and Eurographics Symposium on Virtual Environments
}, editor = {
Hasegawa, Shoichi
and
Sakata, Nobuchika
and
Sundstedt, Veronica
}, title = {{
Empathy in Virtual Agents: How Emotional Expressions can Influence User Perception
}}, author = {
Rings, Sebastian
and
Schmidt, Susanne
and
Janßen, Julia
and
Lehmann-Willenbrock, Nale
and
Steinicke, Frank
}, year = {
2024
}, publisher = {
The Eurographics Association
}, ISSN = {
1727-530X
}, ISBN = {
978-3-03868-245-5
}, DOI = {
10.2312/egve.20241361
} }
Citation
Collections