gaips_bea image1 image2 image3 image4 image5 gaips_ecute_beach_bar_banner gaips_ecute_train_incorrect_ticket_banner
Social Intelligence in Virtual Groups


Abstract Autonomous virtual agents have the potential to promote social engagement of users in virtual environments, thus enhancing their interaction experience. This effect is supported by the interactions users and virtual agents perform together. These interactions are often in group scenarios, where both users and agents perform collaborative tasks. However, in order to have successful group interactions, it is not enough to assure that the characters behave in a coherent manner from an individual perspective, it is also necessary that they exhibit behaviours that are coherent with the group’s composition, context and structure. Furthermore, nurturing social intelligence in virtual agents has been a major concern in the community of multi-agent systems, specially to endow them with skills to perform well in group tasks. However, when building agents to interact in group with users, to perform well in the task does not, necessarily, assure a good interaction experience. While, it is true that users expect the members of their group to perform well in the task, they also expect believable dynamics of the interpersonal relations. In this chapter, we will present a model (SGD Model) to support the dynamics of group interactions that incorporates task related interactions as well as social emotional ones. The model defines the knowledge that each individual should build about the others and the group, and how this knowledge influences its action decision. We will also describe a study that was performed to assess the effect of the SGD Model in the interaction experience of users while playing a collaborative game with virtual counterparts.
Year 2008
Keywords group dynamics, teamwork, social intelligence, autonomous synthetic characters, believability, computer–human interaction
Authors Rui Prada, Ana Paiva
Publisher Springer Berlin / Heidelberg
Chapter New Advances in Virtual Humans: Artificial Intelligence Environment
Volume 140
Series Studies in Computational Intelligence
Month September
Edition Nadia Magnenat-Thalmann, Lakhmi C. Jain, Nikhil Ichalkaranje
Pages 113-132
Pdf File \"pdf
BibTex bib icon or see it here down icon

@inbook { prada08, abstract = {Autonomous virtual agents have the potential to promote social engagement of users in virtual environments, thus enhancing their interaction experience. This effect is supported by the interactions users and virtual agents perform together. These interactions are often in group scenarios, where both users and agents perform collaborative tasks. However, in order to have successful group interactions, it is not enough to assure that the characters behave in a coherent manner from an individual perspective, it is also necessary that they exhibit behaviours that are coherent with the group’s composition, context and structure. Furthermore, nurturing social intelligence in virtual agents has been a major concern in the community of multi-agent systems, specially to endow them with skills to perform well in group tasks. However, when building agents to interact in group with users, to perform well in the task does not, necessarily, assure a good interaction experience. While, it is true that users expect the members of their group to perform well in the task, they also expect believable dynamics of the interpersonal relations. In this chapter, we will present a model (SGD Model) to support the dynamics of group interactions that incorporates task related interactions as well as social emotional ones. The model defines the knowledge that each individual should build about the others and the group, and how this knowledge influences its action decision. We will also describe a study that was performed to assess the effect of the SGD Model in the interaction experience of users while playing a collaborative game with virtual counterparts.}, chapter = {New Advances in Virtual Humans: Artificial Intelligence Environment}, edition = {Nadia Magnenat-Thalmann, Lakhmi C. Jain, Nikhil Ichalkaranje}, keywords = {group dynamics, teamwork, social intelligence, autonomous synthetic characters, believability, computer–human interaction}, month = {September}, pages = {113-132}, publisher = {Springer Berlin / Heidelberg}, series = {Studies in Computational Intelligence}, title = {Social Intelligence in Virtual Groups}, volume = {140}, year = {2008}, author = {Rui Prada and Ana Paiva} }

up icon hide this content