Accéder directement au contenu Accéder directement à la navigation
Communication dans un congrès

Preliminary Evaluation of a Web-Oriented Assessment Tool for Emotion Recognition

Abstract : Perceiving and identifying emotions on facial expressions is one of the basic abilities that compose emotional intelligence, and is crucial for normal social functions. It is well documented that facial expression conveys information about felt emotion, and that expressive behavior can activate or regulate the emotion required by a given situation. Instruments measuring emotion perception based on facial expression have been found in literature either as stand-alone scales or as part of other tests. The proposed tool expands existing instruments to combine online availability while affording assessment of emotion recognition on a continuum of intensity. It was founded on Ekman's Facial Action Units, with two Virtual Characters (male and female) portraying five basic emotions Anger, Disgust, Fear, Joy, Sadness, plus Neutral expression. The user can navigate on the custom-made pentagon and choose the emotion and intensity level (1-5) through a single click. The preliminary evaluation of the tool on thirty normal subjects provided threshold data that can later be used as benchmarks to assess emotion perception sensitivity in psychiatric disorders such as depression and schizophrenia characterized by emotional dysfunction.
Liste complète des métadonnées

https://hal-univ-bourgogne.archives-ouvertes.fr/hal-01438363
Contributeur : Le2i - Université de Bourgogne <>
Soumis le : mardi 17 janvier 2017 - 16:15:32
Dernière modification le : vendredi 17 juillet 2020 - 14:54:10

Identifiants

Citation

Anastasia Pampouchidou, Eleni Kazantzaki,, Kostas Marias, Manolis Tsiknakis, Fabrice Meriaudeau, et al.. Preliminary Evaluation of a Web-Oriented Assessment Tool for Emotion Recognition . 13th International Conference on Wearable Micro and Nano Technologies for Personalised Health (pHealth), May 2016, Heraklion, Greece. pp.95-100, ⟨10.3233/978-1-61499-653-8-95⟩. ⟨hal-01438363⟩

Partager

Métriques

Consultations de la notice

117