A Study on Effective Emotion Expression of Metaverse Avatars through Bio-signal Analysis - Focused on Facial Expression and Gesture

Communication spaces are moving from social networks to metaverse virtual spaces. Metaverse users express their emotions to their counterparts through motions of various body sites based on avatars, which are nonverbal communications. Such nonverbal body motions make it possible to express emotions through avatars' actions and situations, so that they play key roles in communication. The purpose of this study is to find an effective method of expressing emotions based on the results of bio-signal experiments when expressing and delivering emotions through the human-like 3D avatar of the metaverse and presenting them to virtual character creators. The viewer's bio-signal data according to the expression factors were analyzed by deriving three types of emotional expression types according to the avatar's movement: facial expression, gesture and gesture with facial expression. As a result of the study, the average of GSR(galvanic skin response) value of facial expression type was high and the gaze was fixed for longer in the facial expression part. Through the result of immersion and likability, it was found that viewers' attention lingered more on facial expressions rather than gesture, and the sympathetic nervous system was activated. It was observed that viewers read information through facial expressions. Character creation focusing on face expression of human-like 3D avatar would help effective emotion expression and delivery. This study is expected to prepare the academic foundation for emotion expression and communication of avatars with face expression and motion, and to extend to the studies on communication of various platform avatars and characters.