The uncanny valley: No need for any further judgments when an avatar looks eerie

Mincheol Shin, Se Jung Kim, Frank Biocca

Research output: Contribution to journalArticlepeer-review

40 Scopus citations


Considering that a 3D scanning technology allows to create avatars that embody more authentic and rich information about its users, it could be expected that enhancing avatar realism will increase the accuracy of thin-slice judgments of the person represented by the avatar. However, previous studies suggest that the objective realism of avatars may not always lead to positive outcomes, and the activation of negative affect and the aversive motivational system through the uncanny valley effects may rather harm the accuracy of thin-slice judgments by making people distance themselves from available information in avatars. To validate this speculation, a 2 (Realism: Cartoonish vs. Hyper-realistic) x 2 (Animacy: Still vs. Animate) between-subjects experiment (N = 134) was conducted. In support of our prediction, results from an online experiment confirmed that hyper-realistic and animate avatars can induce a greater feeling of eeriness as compared to cartoonish and still avatars. This feeling of eeriness, evoked by the uncanny valley effects, suppressed the amount of information processing oriented for thin-slice judgments, and subsequently decreased the accuracy of extraversion and agreeableness judgments. Further theoretical implications of the findings are discussed.

Original languageEnglish (US)
Pages (from-to)100-109
Number of pages10
JournalComputers in Human Behavior
StatePublished - May 2019

All Science Journal Classification (ASJC) codes

  • Arts and Humanities (miscellaneous)
  • Human-Computer Interaction
  • General Psychology


  • Avatars
  • Information processing
  • Personality perceptions
  • The dual systems model
  • The uncanny valley effects
  • Thin-slice judgments


Dive into the research topics of 'The uncanny valley: No need for any further judgments when an avatar looks eerie'. Together they form a unique fingerprint.

Cite this