학술논문

Virtual Agents and Proxemic Distances: How Social Interactions Affect Our Spatial Representations
Document Type
Conference
Source
2023 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE) Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE), 2023 IEEE International Conference on. :922-927 Oct, 2023
Subject
Communication, Networking and Broadcast Technologies
Computing and Processing
Engineering Profession
General Topics for Engineers
Robotics and Control Systems
Signal Processing and Analysis
Extended reality
Psychology
Neural engineering
Metrology
Regulation
Complexity theory
Task analysis
Social Cognition
Proxemics
Spatial memory
Virtual Agent
Language
Abstract
Proxemic distance regulation allows us to understand, through nonverbal communication, how people psychologically represent the social space they maintain from others. For example, in comfortable interactions, the more familiar people feel with the interactants, the more they tolerate their spatial proximity. By contrast, in uncomfortable interactions, the more people feel unfamiliar with the interactants, the more they maintain a larger distance from them. Recent research has also shown that the proximity of others modulates the processing of objects in a social space. This suggests that the areas of social interaction between individuals may have a different effect on the processing of spatial context stimuli. According to the literature on spatial memory, we use egocentric (subject to object) and allocentric (object to object) reference frames to represent the position of stimuli in the space, even if they are individuals. As everyday communication is becoming increasingly digital, it is worth investigating how the complexity real-world social interactions can be reproduced in the virtual world. Our aim was to verify whether proxemic distance (intimate and personal) between virtual agents could affect the way we represent spatial stimuli according to reference frames. To this end, participants performed a modified version of the Ego-Allo task, in which objects were located between two virtual agents, who could be at intimate or personal proxemic distance from each other. We hypothesized that if spatial memory of stimuli is influenced by proxemic distance between interactants, then intimate and personal social space should produce a different performance in terms of egocentric and allocentric representations. Results showed an egocentric advantage when objects were placed between the agents at an intimate proxemic distance, whereas there was no effect in the personal space condition. Investigating this aspect is crucial to better understand how social dynamics, which are automatically activated in a real-world context, can also be reproduced in a digital context predominant in our times.