PhD Defense thesis Valentine Brémond-Bostoen «Chemical and perceptual analysis of the attractiveness of human body odors: characterization and effect of perfume and of odor gender in virtual reality.»

Valentine Brémond-Bostoen, doctorante NEUROPOP

A l'invitation de

Valentine Brémond-Bostoen, doctorante NEUROPOP

Valentine Brémond-Bostoen

Keywords: body odor, chemical communication, face perception, mate choice, sex differences, age differences, multisensory integration


Abstract: This thesis investigates how factors potentially relevant to mate choice—age, sex, and odor source—shape the chemical composition and perceptual characteristics of human body odors. Through a series of interconnected studies combining chemical analysis, perceptual evaluation, and virtual reality, we examined how these factors influence body odor chemistry and perception, and how they affect attractiveness.

Addressing our first methodological aim, we developed and validated ABOV (Analysis of
Body Odor Volatiles), a novel sampling device designed to collect airborne volatile compounds from the axilla and neck. Study 1 demonstrated that ABOV successfully captures individual-specific chemical profiles. Side, Source, Sex, and Subject accounted for 52.00% of the total variance in chemical compositions. The device enabled reliable differentiation between anatomical sources, and moderate differentiation between sex profiles providing a foundation for subsequent investigations.


Our second aim explored chemical differences in body odor profiles. Studies 1 and 2 revealed that while individual identity was the strongest determinant, significant differences existed between anatomical sources. Chemical analysis revealed distinct profiles between neck and axillary regions, likely reflecting their different glandular composition and microbiological environments. Sex-based differences while also significantly contributing to the chemical variability, were more subtle, with moderate discrimination accuracy. Study 2 looked further into this differentiation, finding significant chemical sex differences for pre-pubertal children and adults. It also showed age was another factor explaining chemical variability that the discrimination between age groups was more important than between sexes.

The third aim examined perceptual characteristics through complementary approaches using untrained raters and trained panelists. Female untrained raters discriminated sex-
based differences in neck odors, particularly for older adult samples. Male raters evaluated female body odors as more attractive. The trained panel, after one month of standardized training, provided detailed descriptive profiles revealing age and sex-related patterns. Young adults' odors were characterized by terms like "fatty/buttery" (40.5% for male samples vs 25.5% for female samples) and "rancid," while older adults' samples received more "perfume" and "soap" descriptors.


Our final aims investigated body odor perception in a close to real life context presented in virtual reality and the influence of perfume and gender labelling. In Study 4, while body odors did not significantly impact explicit attractiveness or gender ratings during video presentation, physiological responses were observed in response to perfume when odors were evaluated independently after the immersion in virtual reality session. When evaluating odors explicitly post-VR, perfume alone or combined with female body odors
received higher ratings for attractiveness, familiarity, and pleasantness compared to body odors alone. Gender labeling showed strong effects on perception independent of the actual odor source - male participants rated female-labeled odors as more attractive, while female participants rated them as less attractive, suggesting possible influences of gender-based cognitive biases in odor perception.


These findings advance our understanding of human body odors as chemical signals carrying information about individual identity, sex, and age. Our results demonstrate that
while humans can detect and discriminate these chemical signals when attention is directed to them, they did not seem to influence the perception of peoples presented in a close to real life setting (in VR setup). The strong effects of cognitive factors like gender labelling highlight how cultural and biological factors interact in human chemical communication.

These conclusions emerge from a methodologically diverse approach combining chemical analysis, perceptual evaluation, and simulated social interactions, contributing to our understanding of human chemical communication in contemporary contexts.

Team
Monday 27 January 2025 13:30–17:30

CRNL | CH Le Vinatier | Bâtiment 462 Neurocampus Michel Jouvet | Amphithéâtre | 95 Boulevard Pinel | 69500 Bron