In short
People are poor at identifying the parts of a smell mixture. Studies suggest that only a few components can be named, and that training does not change this very much. Other findings show that certain many-component mixtures converge, and that mixtures with nothing in common can be made to smell alike. For a perfumer, these research findings offer a scientific frame for what accords have always done, while their laboratory conditions set limits on how far they can be applied to perfume.
What you will learn
- Summarize key findings on perception of mixtures
- Connect masking, blending and convergence to composition
- Test a mixture effect on strips
Before this lesson: Predicting Odor from Structure
Three findings
| Study | Finding | Conditions to note |
|---|---|---|
| Laing and Francis, 1989, and later work | Limited ability to identify components, with a commonly cited limit of about three or four | Set odors; identification, rather than liking |
| Weiss et al., 2012 | Mixtures with many nonoverlapping, equal-intensity components became more similar; a value near 30 components is often quoted | Intensity-matched components spanning odor space |
| Ravia et al., 2020 | Pairs of different mixtures made to smell alike, using a physical measure | Designed to test a model |
Effects you know from the bench
- Blending. Components fuse into something new, which is the point of an accord.
- Masking. One odor lowers the perceived strength of another. This is one route by which an unpleasant trace can vanish, or a pleasant one can disappear.
- Enhancement or synergy. Sometimes reported, but it's harder to demonstrate than people think. Be careful with the word. Test a claim of synergy by comparing with each part alone at the same dose.
- Adaptation across materials. Exposure to one material can reduce sensitivity to similar ones, as you learned in olfactory fatigue.
- Time. The mixture changes as components evaporate, which most lab studies don't address.
What this means for composition
Listing notes. A marketing pyramid that lists a dozen notes isn't describing what wearers can pick out. Few people could name them unprompted. You could use the list to set a mood, but be honest with yourself about it.
Complexity. A long formula doesn't give a more complex perception. A complex effect comes from organized contrast over time, as you found in complex compositions.
Muddiness. It's tempting to link muddiness to olfactory white. A real perfume isn't made of equal-intensity components spread across odor space, so we would treat that link as a thinking aid and nothing more. Still, it suggests something you can test: when too many materials compete at similar strength, character tends to weaken, and adding dominance may restore it.
Reconstruction and substitution. Metamers suggest that different materials can make a similar impression, which perfumers have long relied on. But a match at one moment may separate over time, and for different noses. See your substitution lesson.
How careful to be
- Don't call a perfume "white" because it has more than 30 materials. That's not what the study showed.
- Don't say "people can only smell four things". They can only identify about that many under test conditions.
- Don't treat a lab result on set odorants as a rule for fine fragrance. Use it as a hypothesis.
Designing your own test
You can test simple claims. To test masking, compare a material alone against the same material plus a second one at a stated dose, and have someone rate that first material's strength blind. Make sure of your concentrations, and the amount of solvent. To test identification, prepare coded mixtures from a set of well-known materials, and see how many people can name correctly.
Practical exercise: Mixture identification and convergence tests
- Choose six well-known, dissimilar materials, and read their SDS. Dilute each to a stated percentage on a scale. Try to match them by strength on strips, roughly.
- Have someone prepare coded mixtures with one, two, three and four components, in random order, along with blanks.
- Try to identify the components on strips. Record your score at each size.
- Have at least one other person repeat the test.
- Now prepare two different mixtures of about ten materials, with no overlap, trying to match strengths. Compare them with two different mixtures of two materials. Which pairs seem more alike?
- Write what your small test does and doesn't say about the published findings.
You have done this well if mixtures were prepared and coded by somebody else, your scores are reported with their tiny sample size, and you don't claim to have replicated any study.
Common mistakes
- Extending lab conditions to all perfumes. Note the conditions.
- Calling every improvement "synergy". Test it against the parts.
- Listing notes no one can perceive as if they were facts. Separate story and perception.
Key terms
- Masking
- A reduction in the perceived intensity of one odor by the presence of another.
- Olfactory white
- The convergence reported by Weiss and colleagues, in which diverse mixtures with many equal-intensity components come to smell alike.
- Olfactory metamers
- Mixtures with no components in common that smell the same, as reported by Ravia and colleagues.
Knowledge check
Written and reviewed by the RUŌOD Lab team. This is general education, not legal, regulatory, medical or safety advice. Example dilutions and exercises are for learning, are not validated products, and have not been safety-assessed. Completing a lesson confers no qualification. How we check these lessons.