Differential Physiological Responses to Proxemic and Facial Threats in Virtual Avatar Interactions

📅 2025-08-14
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This study investigates how avatar intrusion into personal space (approach vs. standing phase) and facial expression (neutral vs. angry) jointly modulate users’ physiological responses—specifically skin conductance response (SCR) and heart rate variability (HRV)—as well as subjective discomfort in virtual reality (VR). A 2×2 within-subject VR experiment was conducted, employing individualized personal space boundary calibration and synchronous multimodal physiological recording. Results reveal that SCR responds selectively to spatial intrusion—particularly during the standing phase—whereas HRV significantly decreases only under angry-expression conditions, aligning with heightened subjective discomfort. Notably, angry expressions do not amplify SCR, indicating functional dissociation: SCR reflects spatial threat detection, whereas HRV indexes social–emotional threat processing. This work provides the first empirical evidence of such physiological response segregation during social threat processing in VR. It advances theoretical understanding of threat perception mechanisms and offers methodological foundations for multidimensional assessment of virtual social interactions.

Technology Category

Application Category

📝 Abstract
Proxemics, the study of spatial behavior, is fundamental to social interaction and increasingly relevant for virtual reality (VR) applications. While previous research has established that users respond to personal space violations in VR similarly as in real-world settings, phase-specific physiological responses and the modulating effects of facial expressions remain understudied. We investigated physiological and subjective responses to personal space violations by virtual avatars, to understand how threatening facial expressions and interaction phases (approach vs. standing) influence these responses. Sixteen participants experienced a 2x2 factorial design manipulating Personal Space (intrusion vs. respect) and Facial Expression (neutral vs. angry) while we recorded skin conductance response (SCR), heart rate variability (HRV), and discomfort ratings. Personal space boundaries were individually calibrated using a stop-distance procedure. Results show that SCR responses are significantly higher during the standing phase compared to the approach phase when personal space was violated, indicating that prolonged proximity within personal space boundaries is more physiologically arousing than the approach itself. Angry facial expressions significantly reduced HRV, reflecting decreased parasympathetic activity, and increased discomfort ratings, but did not amplify SCR responses. These findings demonstrate that different physiological modalities capture distinct aspects of proxemic responses: SCR primarily reflects spatial boundary violations, while HRV responds to facial threat cues. Our results provide insights for developing comprehensive multi-modal assessments of social behavior in virtual environments and inform the design of more realistic avatar interactions.
Problem

Research questions and friction points this paper is trying to address.

Investigates physiological responses to personal space violations in VR
Examines effects of facial expressions on proxemic threat reactions
Compares approach versus standing phases in avatar interactions
Innovation

Methods, ideas, or system contributions that make the work stand out.

Individual calibration of personal space boundaries
Multi-modal physiological response measurement
Phase-specific analysis of proxemic responses
🔎 Similar Papers
No similar papers found.
B
Birgit Nierula
Fraunhofer Heinrich-Hertz-Institute, Vision and Imaging Technologies Department, Interactive & Cognitive Systems Group, Einsteinufer 37, 10587 Berlin, Germany
M
Mustafa Tevfik Lafci
Fraunhofer Heinrich-Hertz-Institute, Vision and Imaging Technologies Department, Interactive & Cognitive Systems Group, Einsteinufer 37, 10587 Berlin, Germany
A
Anna Melnik
Fraunhofer Heinrich-Hertz-Institute, Vision and Imaging Technologies Department, Interactive & Cognitive Systems Group, Einsteinufer 37, 10587 Berlin, Germany
M
Mert Akgül
Fraunhofer Heinrich-Hertz-Institute, Vision and Imaging Technologies Department, Interactive & Cognitive Systems Group, Einsteinufer 37, 10587 Berlin, Germany
F
Farelle Toumaleu Siewe
Fraunhofer Heinrich-Hertz-Institute, Vision and Imaging Technologies Department, Interactive & Cognitive Systems Group, Einsteinufer 37, 10587 Berlin, Germany
Sebastian Bosse
Sebastian Bosse
Head of Interactive & Cognitive Systems, Fraunhofer HHI, Germany
computer visionhuman-computer interactionhybrid modelsmachine learningcognition modelling