Multimodal Wearable Dataset Advances Olfactory Emotion Recognition

A new extensive multimodal dataset has been developed by researchers to improve emotion recognition via olfaction, filling existing gaps in affective computing. This dataset, which includes data from 111 participants, records EEG, ECG, and PPG signals as subjects encounter odors categorized within a two-dimensional arousal-valence framework. Announced on arXiv (ID: 2608.00043), the study seeks to address challenges in olfactory emotion research, underscoring the significance of olfaction in emotion regulation. It enables the integration of central and peripheral signals for a deeper understanding of emotional responses and showcases the promise of wearable technology for discreet emotional monitoring, influencing areas such as human-computer interaction, mental health, and consumer neuroscience, ultimately leading to real-time responsive personalized technologies.