Selection of odors in multimedia based on odor categories watched with objects in scenes

Abstract

Multimedia has mainly focused on visual and auditory senses, although humans detect numerous stimuli using five senses. As one of the attempts to extend senses used in multimedia, olfactory stimulation has been used in multimedia content for enhancing the sense of multimedia's reality. Matching odors with objects in scenes is mainly conducted when selecting odors for multimedia. However, it is impractical to select all odors matched with all objects in scenes and offer them to viewers. As an alternative, offering an odor in a category was suggested to represent odors belonging to the category. However, it is still unclear whether viewers' responses to videos with multiple odors (e.g., rose, lavender, lily) from a category (e.g., flower) can be comparable. Therefore, we studied whether odors belonging to the same categories could be similar by monitoring congruency and five frequency bands (delta, theta, alpha, beta, and gamma) of the EEG data in videos. We conducted questionnaires and EEG experiments to validate the effects of odors belonging to similar categories. Our result showed that odors in the similar odor categories were higher congruency to video clips than odors in the different odor categories. In our EEG data, mainly delta and theta bands were clustered in both video clips when odors were offered in similar categories. However, alpha, beta, gamma bands were not clustered depending on the categories. Our studies suggested that choosing the odors based on odor categories in multimedia can be partially feasible.