Effects of avatar's blinking animation on person impressions

Kazuki Takashima, Yasuko Omori, Yoshiharu Yoshimoto, Yuich Itoh, Yoshifumi Kitamura, Fumio Kishino

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

39 Citations (Scopus)


Blinking is one of the most important cues for forming person impressions. We focus on the eye blinking rate of avatars and investigate its effect on viewer subjective impressions. Two experiments are conducted. The stimulus avatars included humans with generic reality (male and female), cartoon-style humans (male and female), animals, and unidentified life forms that were presented as a 20-second animation with various blink rates: 9, 12, 18, 24 and 36 blinks/min. Subjects rated their impressions of the presented stimulus avatars on a seven-point semantic differential scale. The results showed a significant effect of the avatar's blinking on viewer impressions and it was larger with the human-style avatars than the others. The results also lead to several implications and guidelines for the design of avatar representation. Blink animation of 18 blinks/min with a human-style avatar produces the friendliest impression. The higher blink rates, i.e., 36 blinks/min, give inactive impressions while Ihe lower blink rates, i.e., 9 blinks/min, give intelligent impressions. Through these results, guidelines are derived for managing attractiveness of avatar by changing the avatar's blinking rate.

Original languageEnglish
Title of host publicationProceedings - Graphics Interface 2008
Number of pages8
Publication statusPublished - 2008
EventGraphics Interface 2008 - Windsor, ON, Canada
Duration: 2008 May 282008 May 30

Publication series

NameProceedings - Graphics Interface
ISSN (Print)0713-5424


ConferenceGraphics Interface 2008
CityWindsor, ON


  • Blinking
  • Character
  • Computer-mediated communication
  • Facial animation
  • Graphics
  • Psychology


Dive into the research topics of 'Effects of avatar's blinking animation on person impressions'. Together they form a unique fingerprint.

Cite this