Even people who feel they cannot really detect musk may notice a particular musk at a low concentration. In a 1996 study of 32 people tested with 11 musks, researchers found people who were relatively insensitive to many of them but sensitive to two particular musks.
The study’s main aim was not to reduce a person’s sense of smell to a single score, but to compare each person’s sensitivity across several fragrance materials and look for shared patterns.
Measuring the concentration at which 11 musks become detectable
The participants were 32 nonsmokers aged 19–40, recruited through newspaper advertisements: 16 men and 16 women. They reported no problems with their sense of smell or nasal breathing. They had not been selected on the basis of sensitivity to any particular musk.
The researchers tested 11 structurally different materials with musky odors, including muscone, civetone, ethylene brassylate, and Galaxolide. They also used the non-musk odorant phenylethyl alcohol for comparison. Detection thresholds were measured by having participants smell two bottles, one containing a diluted fragrance material and the other solvent alone, and identify which contained an odor.
The concentration was reduced after two consecutive correct answers and increased after an incorrect answer. This process was repeated, and thresholds were calculated from stable responses. Testing was divided into four sessions, with two sets of bottles used alternately so that the same bottle was not used consecutively. There was a break of at least 10 minutes between odorants. What the researchers measured was the concentration at which an odor’s presence could be detected, not preferences or rankings of odor intensity.
Most people were broadly sensitive or broadly insensitive
Because starting concentrations and volatilization behavior differed between materials, the researchers did not simply compare the numerical thresholds across odorants. Instead, they ranked the 32 participants’ sensitivity separately for each odorant, then used statistical methods to group people with similar patterns of rankings.
There were two large groups. Twelve participants clustered on the relatively sensitive side across the musks, while 16 clustered on the relatively insensitive side. This does not mean that everyone in a group had exactly the same sensitivity, but that their patterns were similar when considered across all 11 musks.
The researchers also reported a positive association between average sensitivity rank across the musks and sensitivity rank for the comparison odorant, phenylethyl alcohol. They suggested that differences in more general olfactory sensitivity, rather than sensitivity to musks alone, might also be involved. However, they used only one non-musk comparison odorant; they did not measure sensitivity to every kind of smell.
Two small groups sensitive to two particular musks
The remaining four participants showed two distinct patterns, with two people in each group. One pair was highly sensitive to cyclopentadecanone and musk xylol, while the other was highly sensitive to delta9-hexadecenolactone and Tonalid. For many of the other musks, they were on the relatively insensitive side.
These combinations cannot be captured simply by dividing people into those who can and cannot smell musk. Someone with low sensitivity to one musk may rank highly for another with a different structure. The researchers gave provisional names to these two patterns of specific high sensitivity to multiple musks.
Meanwhile, in 18 of the 352 musk-threshold determinations, no threshold could be obtained within the concentration range tested. Such cases were more common in the insensitive and mixed-sensitivity groups than in the broadly sensitive group. Being highly sensitive to a particular musk could therefore coexist with difficulty detecting another.
Links to genes and receptors remain hypotheses for future research
The authors aimed to define sensitivity characteristics quantitatively so that future studies could examine their relationship with differences in olfactory receptors and genes. They suggested that general sensitivity might involve nasal anatomy, the number of olfactory sensory cells, or receptor expression. For the patterns of sensitivity to particular musks, they proposed differences in receptor genes or their expression as hypotheses.
This experiment involved threshold measurements and statistical classification, not genetic testing or receptor measurements. Each distinctive group contained only two people, and the original paper discusses the possibility that the classifications arose by chance. The study also cannot be treated as a survey showing what percentage of the general population belongs to these types.
For the correlation concerning general sensitivity, the correlation coefficient, sample size, and p-value reported in the original paper are not consistent with one another, so the precise statistical significance requires caution. It is more appropriate to focus on the study’s approach of examining broad sensitivity patterns alongside exceptions for individual materials.
Summary
Most of the 32 participants showed a broadly sensitive or broadly insensitive pattern across the 11 musks. At the same time, the researchers found two pairs of people who were insensitive to many musks but sensitive to two particular ones. The association with the non-musk comparison odorant also suggested that general sensitivity might overlap with these patterns.
Even within the single odor category called “musk,” people differ in what they can readily detect. This study offers a reason not to assume that difficulty smelling one musk means all other musks will be the same.
Both authors’ affiliation at the time of the research is listed as Givaudan-Roure Fragrances, with a note identifying Synesthetics as the first author’s subsequent address at publication. No separate research-funding or conflict-of-interest statement was found in the original paper.
References
Gilbert AN, Kemp SE. Odor Perception Phenotypes: Multiple, Specific Hyperosmias to Musks. Chemical Senses. 1996;21(4):411–416. doi:10.1093/chemse/21.4.411
