The mere presence of an attentive and emotionally responsive virtual character influences focus of attention and perceived stress

• Published in: International journal of human-computer studies Vol. 132; pp. 45 - 51
• Main Authors: Felnhofer, Anna, Kaufmann, Marlene, Atteneder, Katharina, Kafka, Johanna Xenia, Hlavacs, Helmut, Beutl, Leon, Hennig-Fast, Kristina, Kothgassner, Oswald David
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.12.2019
• Subjects: Agent; Avatar; Social support; Stress buffering; Trier Social Stress Test (TSST); Virtual reality; Stress buffering; Agent; Trier Social Stress Test (TSST); Avatar; Virtual reality; Social support
• ISSN: 1071-5819; 1095-9300
• DOI: 10.1016/j.ijhcs.2019.07.010

•The attentive presence of a virtual other seems to act as a stress buffer.•Avatar support may particularly help reduce tension and increase focus on others.•Unsupported individuals tend to show more self-focus and tension during stress.•There are no differences in levels of social presence evoked by avatars and agents. As social interactions increasingly shift towards digitally mediated communications, understanding the underlying mechanisms of virtually provided support and its effect on stress buffering is of central importance. Thus, this study's objective was to assess, whether a virtual character's (avatar vs. agent) attentive and (non-verbal) emotionally responsive presence during a phase anticipating a real-life social-evaluative stressor would attenuate subsequent stress responses and have a beneficial effect on attentional focus and self-esteem. In addition, social presence and physical presence experiences were assessed. 48 healthy young adults were either supported by a human controlled avatar, a computer agent or received no support prior to a real-life Trier Social Stress Test (TSST). Results revealed a moderate stress buffering effect of a virtual character's non-verbal attentive presence preceding a stressor. In particular, the presence of the human controlled avatar led to reduced feelings of tension as well as to a heightened attentional focus on others during the TSST. By trend, avatar support also seemed to protect the participants’ self-esteem. In contrast, unsupported controls showed the tendency to focus more on the self and report more tension. Finally, there were no group differences in social or physical presence. The attentive presence of a virtual character seems to be effective in terms of attenuating subjective stress responses, particularly if the user has the impression that the virtual other is controlled by a human. Hence, online stress prevention and treatment may profit from purposefully inserting virtual characters as supporters and motivators.