Decoding odor responses: universal patterns and individual signatures in psychophysiology using nonlinear models.

Area of Science:

• Neuroscience
• Sensory Science
• Computational Biology

Background:

• Olfactory perception is influenced by odorant chemistry and individual factors, making prediction difficult.
• Existing models often lack integration of psychophysiological data and individual response variability.

Purpose of the Study:

• To develop a unified model for olfactory perception by integrating odorant properties, individual rating tendencies, and psychophysiological responses.
• To investigate the relative contributions of stimulus characteristics, individual biases, and physiological signatures to odor perception.

Main Methods:

• Collected time-series psychophysiological data (respiration, heart rate, EMG) from 41 participants exposed to six odorants.
• Employed nonlinear modeling to analyze perceptual dimensions (valence, temperature, intensity) alongside odorant properties and physiological signals.
• Accounted for both group-level and individual-specific physiological effect patterns.

Main Results:

• Odorant identity and individual rating tendencies were primary drivers of perception, with varying contributions across dimensions.
• Physiological signals significantly enhanced predictive models, highlighting both generalizable and individualized response patterns.
• Respiration and EMG corrugator were key group-level predictors for valence and intensity; individual patterns varied significantly.

Conclusions:

• A comprehensive understanding of olfactory perception necessitates considering the interplay between stimulus features, individual biases, and physiological signatures.
• Both universal and person-specific physiological responses are crucial for nuanced odor perception modeling.
• This integrated approach offers a more detailed insight into the complexities of human olfactory experience.