Publication on Smell in special issue Frontiers in Psychology
NOSE-members Jasper de Groot, Ilja Croijmans, and Monique Smeets published their perspective paper “More data, please: Machine learning to advance the multidisciplinary science of human sociochemistry” in Frontiers in Psychology. The paper fits in a special issue on advances and obstacles in contemporary nonverbal communication research.
Human smells are increasingly being recognized as communicative medium. In this paper, Jasper, Ilja, and Monique highlight the complexity of human odors, which consist of thousands of molecules, and which are influenced by our genes, hormones, mood, skin bacteria, diet, smoking, hygiene habits, clothing, and use of fragranced products. Past research in this field consisted of small-scale psychological experiments conducted in sterile labs, which formed the first stepping stones by strongly suggesting that social information can be communicated via smell under tightly controlled settings; yet, to identify the molecules transmitting the chemical message and to assess their real-world impact, we need multidisciplinary ways of working by integrating psychology and chemistry towards a science of human sociochemistry, and we need to move outside of the sterile lab to test subjects with diverse backgrounds in their natural settings. The sociochemistry we advance is a multidisciplinary, ecological approach that in view of its inherent complexity requires an ecosystem of academic institutions around the world to flourish, by working together to create speed and scale. We propose the building of open access databases holding information that spans across chemistry (e.g., chemical composition of sweat odor) and psychology (e.g., capturing the states and traits of those participating in the chemosignaling as well as their unique contextual information). We introduce machine learning techniques that can be applied to generate models that may accurately predict molecules’ sway on our social lives across diverse contexts and samples, with technological, societal, and clinical applications following suit.
De Groot, J. H., Croijmans, I., & Smeets, M. A. (2020). More data, please: Machine learning to advance the multidisciplinary science of human sociochemistry. Frontiers in Psychology, 11, 2904. http://dx.doi.org/10.3389/fpsyg.2020.581701