Abstract

This paper presents an experiment investigating the impact of behavior and responsiveness on social responses to virtual humans in an immersive virtual environment (IVE). A number of responses are investigated, including presence, copresence, and two physiological responses-heart rate and electrodermal activity (EDA). Our findings suggest that increasing agents' responsiveness even on a simple level can have a significant impact on certain aspects of people's social responses to humanoid agents.Despite being aware that the agents were computer-generated, participants with higher levels of social anxiety were significantly more likely to avoid "disturbing" them. This suggests that on some level people can respond to virtual humans as social actors even in the absence of complex interaction.Responses appear to be shaped both by the agents' behaviors and by people's expectations of the technology. Participants experienced a significantly higher sense of personal contact when the agents were visually responsive to them, as opposed to static or simply moving. However, this effect diminished with experienced computer users. Our preliminary analysis of objective heart-rate data reveals an identical pattern of responses.

A common measure of effectiveness of a virtual environment (VE) is the amount of presence it evokes in users. Presence is commonly defined as the sense of being there in a VE. There has been much debate about the best way to measure presence, and presence researchers need and have sought a measure that is reliable, valid, sensitive, and objective. We hypothesized that to the degree that a VE seems real, it would evoke physiological responses similar to those evoked by the corresponding real environment, and that greater presence would evoke a greater response. To examine this, we conducted four experiments, each of which built upon findings that physiological measures in general, and heart rate in particular, are reliable, valid, sensitive, and objective presence measures. The experiments compare participants' physiological reactions to a nonthreatening virtual room and their reactions to a stressful virtual height situation. We found that change in heart rate satisfied our requirements for a measure of presence, change in skin conductance did to a lesser extent, and that change in skin temperature did not. Moreover, the results showed that significant increases in heart rate measures of presence appeared with the inclusion of a passive haptic element in the VE, with increasing frame rate (30 FPS > 20 FPS > 15 FPS) and when end-to-end latency was reduced (50 ms > 90 ms).

No abstract