Recognition of Emotions in Natural and Virtual Faces
Author Information
Author(s): Dyck Miriam, Winbeck Maren, Leiberg Susanne, Chen Yuhan, Gur Rurben C., Mathiak Klaus
Primary Institution: RWTH Aachen University
Hypothesis
It is possible to generate virtual emotional expressions that are recognized as well as natural emotional expressions.
Conclusion
Virtual and natural facial displays of emotion may be equally effective, with some virtual emotions being recognized better than natural ones.
Supporting Evidence
- Virtual sadness and fear achieved better recognition results than natural faces.
- Disgust was difficult to convey with the current avatar technology.
- Recognition rates decreased for virtual but not natural faces in participants over the age of 40.
Takeaway
The study shows that people can recognize emotions on virtual faces almost as well as on real faces, but some emotions like disgust are harder to recognize on virtual faces.
Methodology
Thirty-two healthy volunteers rated pictures of natural human faces and virtual characters on expressed emotions.
Limitations
Recognition rates for negative emotions were low, and the study only used static facial stimuli.
Participant Demographics
Participants were healthy volunteers aged 20 to 60, with equal gender distribution.
Statistical Information
P-Value
p<0.001 for sadness, p<0.002 for fear, p<0.0001 for disgust and neutral expressions.
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website