Validation of dynamic virtual faces for facial affect recognition

Abstract

The ability to recognise facial emotions is essential for successful social interaction. The most common stimuli used when evaluating this ability are photographs. Although these stimuli have proved to be valid, they do not offer the level of realism that virtual humans have achieved. The objective of the present paper is the validation of a new set of dynamic virtual faces (DVFs) that mimic the six basic emotions plus the neutral expression. The faces are prepared to be observed with low and high dynamism, and from front and side views. For this purpose, 204 healthy participants, stratified by gender, age and education level, were recruited for assessing their facial affect recognition with the set of DVFs. The accuracy in responses was compared with the already validated Penn Emotion Recognition Test (ER-40). The results showed that DVFs were as valid as standardised natural faces for accurately recreating human-like facial expressions. The overall accuracy in the identification of emotions was higher for the DVFs (88.25%) than for the ER-40 faces (82.60%). The percentage of hits of each DVF emotion was high, especially for neutral expression and happiness emotion. No statistically significant differences were discovered regarding gender. Nor were significant differences found between younger adults and adults over 60 years. Moreover, there is an increase of hits for avatar faces showing a greater dynamism, as well as front views of the DVFs compared to their profile presentations. DVFs are as valid as standardised natural faces for accurately recreating human-like facial expressions of emotions.

Facial Affect Recognition by Patients with Schizophrenia Using Human Avatars

“…It is also true that some study found differences between young adults (20–39 years) and older adults (40–59 years) in favor of the younger group [ 36 ]. Therefore, and in line with a recent paper by the same authors validating the set of avatars in healthy population [ 24 ], three age groups were established for the analysis: 18–39 years, 40–59 years and above (here, 60–65 years old).…”

“…Some examples of DVFs are shown in Figure 1 . The complete description of the DVFs employed can be consulted in our previous experiment that validated the DVFs in 204 healthy people [ 24 ]. The software tool and the DVFs developed for this research are publicly accessible at Universidad de Castilla-La Mancha institutional repository RUIdeRA at , accessed on 20 April 2021.…”

“…The complete design process has been published accordingly [ 14 ]. Afterwards, these faces have been validated in 204 healthy people, in order to have a clear idea of the percentage of success in recognizing emotions shown in those faces [ 24 ]. The results showed that virtual humans were valid to recreate accurately the facial expressions of human-like emotions.…”

How Interpersonal Distance Between Avatar and Human Influences Facial Affect Recognition in Immersive Virtual Reality

“…We used an immersive version of the software originally designed for facial affect recognition under non-immersive virtual reality (García et al, 2020b;Fernández-Sotos et al, 2021). Therefore, the avatars used and the expressions they depict have already been validated by a sample of 204 healthy volunteers (50% women and 50% men) using a non-immersive display.…”

“…Very recently, our research team has carried out the validation of a set of dynamic avatars for the task of facial affect recognition (Fernández-Sotos et al, 2021). The validation was performed on a healthy population with the intention of being able to make a comparison in the near future with patients.…”