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

Abstract Unlike many human senses, multimedia primarily engages the visual and auditory faculties. To broaden the sensory experience influenced by multimedia, it has incorporated olfactory stimulation to enhance the sense of reality. Odors are typically matched with objects in scenes. However, it is impractical to include all odors corresponding to every object in a scene for viewers. Alternatively, researchers propose presenting a singular odor from a category, representative of others within that category. Yet, it remains uncertain whether viewers' reactions to videos featuring multiple odors (e.g., rose, lavender, and lily) from a specific category (e.g., flower) are comparable. Therefore, we investigated whether odors within a given category could exhibit similarity in congruency based on the electroencephalogram (EEG) data's five frequency bands (delta, theta, alpha, beta, and gamma) collected as viewers watched videos. Through questionnaires and EEG experiments, we sought to comprehend the impact of similar odors within categories. It was observed that odors within a specific category were more congruent with videos than those from different odor categories. The delta and theta bands predominantly clustered in EEG data when odors from similar categories were presented to viewers. The theta band, linked to neural signals of odors during olfactory processing, played a significant role. However, despite their association with human emotional responses, the alpha, beta, and gamma bands did not exhibit clustering based on category. Our findings demonstrate the viability of selecting odors based on their categories in multimedia.