Design of an Emotion Elicitation Framework for Arabic Speakers

The automatic detection of human affective states has been of great interest lately for its applications not only in the field of Human-Computer Interaction, but also for its applications in physiological, neurobiological and sociological studies. Several standardized techniques to elicit emotions have been used, with emotion eliciting movie clips being the most popular. To date, there are only four studies that have been carried out to validate emotional movie clips using three different languages (English, French, Spanish) and cultures (French, Italian, British / American). The context of language and culture is an underexplored area in affective computing. Considering cultural and language differences between Western and Arab countries, it is possible that some of the validated clips, even when dubbed, will not achieve similar results. Given the unique and conservative cultures of the Arab countries, a standardized and validated framework for affect studies is needed in order to be comparable with current studies of different cultures and languages. In this paper, we describe a framework and its prerequisites for eliciting emotions that could be used for affect studies on an Arab population. We present some aspects of Arab culture values that might affect the selection and acceptance of emotion eliciting video clips. Methods for rating and validating Arab emotional clips are presented to derive at a list of clips that could be used in the proposed emotion elicitation framework. A pilot study was conducted to evaluate a basic version of our framework, which showed great potential to succeed in eliciting emotions.