Understanding the underlying dimensions in perfumers' odor perception space as a basis for developing meaningful odor...

Area of Science:

• Olfactory science
• Perceptual mapping
• Computational chemistry

Background:

• Numerous low-dimensional perceptual maps for fragrances exist.
• Sensory maps for perfumery odor descriptors are challenging to standardize.
• Previous attempts at consensus have yielded limited success.

Purpose of the Study:

• To apply principal components analysis (PCA) to fragrance databases.
• To investigate the potential for creating standardized sensory maps for perfumery.
• To explore consensus on odor descriptors for fragrance mapping.

Main Methods:

• Principal Components Analysis (PCA) applied to two distinct databases.
• Database 1: Numeric odor profiles of 309 compounds with 30 descriptors.
• Database 2: Semantic descriptions of 119 perfume materials with 66 descriptors.

Main Results:

• PCA loading plots from numeric data closely resembled Jellinek's (1951) odor effects diagram.
• Similar results were obtained analyzing the semantic description database.
• A commercial map of fragrances was discussed based on the analyses.

Conclusions:

• Standardized sensory maps for perfumery odor descriptors are achievable.
• Consensus on odorant representation for specific odor qualities is a prerequisite.
• This research provides a data-driven approach to fragrance mapping.