Cross-Cultural Impressions and Cue Weighting of Avatars for Emergency Announcements: Evidence from China and Japan

Abstract

Avatars are increasingly used for serious emergency announcements, but cross-cultural differences in impressions and cue weighting remain unclear. We compared participants from China (n=40) and Japan (n=40) evaluating four emergency-announcement avatars. In Part 1, participants watched avatar-based announcement videos and completed comprehension checks; understanding was generally high, while the Chinese participants scored higher than the Japanese with avatar-dependent differences. In Part 2, participants rated the avatars using static reference images (for identification) and reported the importance of six cues (clothes, gender, voice, style, speed, race). Mixed-design analyses showed that cross-cultural differences were most robust for authority, whereas trustworthiness and affability showed weaker effects. The Chinese participants assigned higher overall cue importance than the Japanese, and authority cue-importance patterns differed by country. A supplementary dlib-based facial landmark profile provided descriptive context for avatar-specific China–Japan gaps. These findings highlight the need for country aware validation in designing serious-communication avatars.

Keywords: Serious Communication, Emergency Announcements, Cross-cultural Comparison, Authority Impression, Cue Importance

DOI: 10.54941/ahfe1007426