Why Perfume Feels Like Color: Synesthesia, Scent & Fragrance

A perfume can be blue without containing anything blue.

It can feel velvet-soft, metallic, transparent, dusty, round or cold. A citrus can seem to sparkle; a resin can have a low, almost sonorous quality; orris can suggest the blur of powder on suede. None of these descriptions belongs, strictly speaking, to smell.

And yet they often tell us more than saying that a fragrance contains bergamot, jasmine and sandalwood.

Perfume has always encouraged this kind of sensory borrowing because smell is unusually difficult to describe in isolation. We reach for color, touch, temperature, shape, movement and sound because they give form to something we cannot see or hold.

There is science behind some of these associations, although synesthesia is not quite the right word for what most of us experience.

The distinction makes perfume more interesting.

Synesthesia is a perceptual phenomenon in which one kind of stimulus consistently and automatically produces an additional sensory experience. Someone might perceive particular letters as inherently colored, for example, or experience sounds together with specific colors.

Researchers distinguish this from crossmodal correspondences, the much more widespread tendency to associate qualities in one sensory domain with qualities in another. Certain sounds seem intuitively brighter or darker, shapes softer or sharper, odors more naturally suited to one color than another.

The two phenomena are related areas of multisensory research, but they should not be collapsed into the same thing. Psychologists Charles Spence and Ophelia Deroy have argued explicitly that the ordinary cross-sensory associations found across large groups of people differ in important ways from genuine synesthetic experience.

For perfume, this distinction is useful. You do not have to literally see a wash of violet every time you smell orris to experience a fragrance as cool, pale or soft-edged.

Most perfume language already assumes that we understand this.

We describe smells as bright and dark, warm and cool, soft and sharp. We talk about transparent florals, dry woods, creamy sandalwood, sparkling citrus, plush rose and mineral musk. Our language moves easily between senses because the experience itself encourages us to.

Odor and color have been studied together particularly closely.

In a 2006 experiment, M. Luisa Demattè, Daniel Sanabria and Charles Spence found that participants consistently paired certain odors with particular colors. Those correspondences appeared not only when people consciously selected a matching color, but also in an implicit association task designed to test how readily the pairings were processed.

Subsequent research has continued to find meaningful odor-color associations. They are not necessarily universal, nor do they mean that a particular ingredient possesses an objectively correct color. Familiarity, the color of an odor's source, emotional character and cultural learning can all contribute to the association. A lemon smell being connected with yellow is hardly mysterious; other pairings are less easily explained. A broad review of olfactory-color correspondence research describes these relationships as potentially statistical, semantic, structural or emotional rather than manifestations of synesthesia itself.

This helps explain a familiar experience in perfumery. A fragrance can seem golden before we know why. Another reads as grey. A particularly green perfume can feel green even when the liquid is perfectly clear.

Part of that impression may come from identifiable references to leaves, herbs or citrus. Part may arise from the emotional tone of the fragrance. Part may come from associations we have accumulated over years without noticing them.

Color becomes a way of organizing smell.

Texture is perhaps even more natural to the language of perfume.

Velvet. Suede. Silk. Powder. Wax. Smoke. Cream. Lacquer.

Perfumers and wearers use these words constantly because ingredients can create impressions that feel uncannily tactile. Orris may seem powder-soft; a dry vetiver can feel almost fibrous; vanilla can suggest smoothness; certain woods feel polished while others seem rough or splintered.

Research suggests that smell and touch can genuinely influence one another. In experiments examining olfactory and tactile perception, participants judged identical fabric swatches as softer when they were presented alongside one odor than when they were paired with another. The smell had altered the reported tactile experience of the material itself.

That does not mean perfume ingredients have intrinsic textures waiting to be decoded. It suggests something subtler: sensory perception is not assembled inside perfectly sealed compartments.

Our senses continually inform one another.

The velvet we perceive in a perfume may be metaphorical, but metaphor does not make the perception meaningless.

Sound seems like a more improbable partner for scent, yet researchers have found correspondences there too.

Anne-Sylvie Crisinel and Charles Spence asked participants to pair odors with musical notes and found systematic associations between smell and pitch. Other research has found correspondences between odors and geometric shapes, including preferences for rounder or more angular forms.

A broader review of this work concluded that olfactory experience is particularly rich in crossmodal associations, including relationships with pitch and shape that cannot always be explained simply by the appearance of the odor's source.

Perfumery language begins to make more sense in this context.

Citrus can feel high because its brightness, volatility and freshness seem to occupy an upper register. Dense resins can feel low, slow and resonant. A fragrance with great diffusion can seem expansive, while an intimate musk may barely rise above a whisper.

These descriptions are poetic, but they are not arbitrary in the way they might first appear.

A perfumer does not literally compose in musical notation. Still, describing one material as a high clear note and another as a low hum can convey something that the ingredient name alone cannot.

There is another reason we borrow so heavily from the other senses: in English, naming smells is difficult.

Ask someone to identify a familiar odor without any context and the experience can be surprisingly frustrating. We often resort to naming the thing we think produced it: lemon, rose, smoke, leather.

For a long time this difficulty was sometimes treated as a basic limitation of human language. Cross-cultural research has complicated that assumption.

Asifa Majid and colleagues have shown that some languages possess dedicated vocabularies for odor qualities and that speakers of those languages can describe smells with far greater ease than English-speaking participants. Among Jahai speakers in the Malay Peninsula, for example, odor naming does not depend on constantly referring back to source objects.

So the problem is not simply that smell cannot be spoken.

It is partly that English has not given us particularly elegant tools for speaking about it.

Perfume culture has invented some of its own.

“Powdery,” “green,” “mineral,” “creamy,” “dry” and “transparent” become a kind of working vocabulary. None belongs exclusively to olfaction, but together they allow increasingly precise conversations about fragrance.

The metaphors become useful because people begin to understand what they point toward.

Natural materials are especially rewarding subjects for this kind of language because a botanical extract is rarely a single aromatic idea.

Rose absolute does not smell only of rose. Jasmine can carry floral radiance alongside fruit, warmth and indolic darkness. Labdanum moves between resin, dried fruit, leather and smoke. Patchouli can contain cool greenery, earth, wood and a soft sweetness that emerges much later.

Those internal contrasts give the perfumer several directions to follow.

Placed against citrus, a natural rose may seem brighter and more translucent. Against patchouli or labdanum, the same material can become dark, saturated and almost velvety. Orris can feel powdery beside violet and more root-like beside dry woods.

That complexity can produce very vivid cross-sensory impressions, but it would be misleading to claim that natural perfume is uniquely capable of doing so.

Synthetic aroma materials can be extraordinarily textural, abstract and atmospheric. Modern perfumery has expanded the range of sensations a perfumer can suggest precisely because its palette includes both botanical extracts and an enormous variety of individual aroma molecules.

The distinction is not between “natural materials with dimension” and “synthetics without it.”

It is between what different materials allow the perfumer to construct.

For a house especially interested in natural perfume, the attraction of botanicals is their particular complexity and variability, not a claim that they possess a sensory depth unavailable anywhere else.

Niche fragrance is not inherently more multisensory than mainstream perfume, just as it is not inherently darker, stranger or more sophisticated.

It does, however, provide fertile territory for abstraction.

A perfume intended for a very broad audience often benefits from an immediately legible idea. Vanilla. Rose. Fresh citrus. Clean skin.

A more narrowly conceived fragrance has room to begin somewhere less literal.

Cold porcelain.

Pink silk.

Wet stone.

Sunlight against old wood.

These are not perfume notes. They are sensory directions.

A perfumer can then work backward, assembling materials capable of creating the impression without attempting to reproduce any one physical object exactly.

This is where the language of color and texture becomes useful rather than decorative. Instead of asking only whether the formula contains jasmine, a perfumer might be interested in making the jasmine feel almost white, or humid, or translucent. A wood can be made polished or rough. An amber can glow or smolder.

Niche fragrance is interesting when this freedom results in specificity, not merely unusual copywriting.

The perfume still has to make the world it describes believable.

There is an obvious danger in all of this.

Once fragrance language moves beyond ingredients, almost anything can be claimed.

A bottle arrives surrounded by photographs of brutalist architecture, wet pavement and silver-grey skies. The description speaks of steel and distance. By the time we smell the fragrance, we have already been told what sensory world we are supposed to find.

Suggestion is powerful.

Research on crossmodal correspondence makes this more interesting, not less, because it reminds us that perception can be shaped by information arriving through other senses. Color, texture, packaging, language and surroundings can influence how an odor is interpreted.

This does not make fragrance storytelling dishonest. Context has always been part of how art is encountered.

It does mean there is pleasure in occasionally smelling before reading.

A perfume strip without the note list.

A bottle without its campaign.

A few minutes in which the fragrance gets to suggest its own color.

Sometimes it agrees with the story. Sometimes it doesn't.

The disagreement can be more interesting.

There is no need for a three-step method.

The next time you smell a perfume, simply notice the language that arrives before you begin searching for ingredients.

Perhaps the first impression is temperature. It feels cool.

Perhaps it has a surface: polished, dusty, waxed, soft.

Perhaps a color appears without effort.

Perhaps the perfume feels expansive or confined, translucent or opaque. Maybe it seems to hum rather than sparkle.

None of those impressions needs to be correct.

What matters is that they describe your experience of the perfume more precisely than “floral woody musk” ever could.

Afterward, the note list can become interesting again. You may begin to understand why a particular combination produced the sensation, or discover that the material you expected to be responsible is not present at all.

This is how fragrance literacy becomes more pleasurable rather than more academic. Knowledge gives sensation more vocabulary.

It should not tell sensation what to say.

Most of us are not synesthetes.

We do not need to be.

Ordinary perception is already more entangled than the traditional division into five separate senses suggests. Research has repeatedly shown that odors can form systematic associations with colors, sounds, shapes and tactile qualities. Language makes use of those connections every time we describe a fragrance as bright, smooth, warm or sharp.

Perfume gives these associations an unusually rich place to play.

A natural perfume can take the smell of a root, flower or resin and make it feel like fabric, light or weather. A niche fragrance can begin with an abstract sensation and use scent to make it temporarily physical.

The experience does not need to be neurological synesthesia to feel wonderfully strange.

Sometimes a perfume really does seem blue.

The interesting part is discovering why.

Crisinel, A.-S., & Spence, C. (2012). “A Fruity Note: Crossmodal Associations Between Odors and Musical Notes.” Chemical Senses, 37(2), 151–158. doi:10.1093/chemse/bjr085.

Demattè, M. L., Sanabria, D., & Spence, C. (2006). “Cross-Modal Associations Between Odors and Colors.” Chemical Senses, 31(6), 531–538. doi:10.1093/chemse/bjj057.

Demattè, M. L., Sanabria, D., Sugarman, R., & Spence, C. (2006). “Cross-Modal Interactions Between Olfaction and Touch.” Chemical Senses, 31(4), 291–300. doi:10.1093/chemse/bjj031.

Deroy, O., Crisinel, A.-S., & Spence, C. (2013). “Crossmodal Correspondences Between Odors and Contingent Features: Odors, Musical Notes, and Geometrical Shapes.” Psychonomic Bulletin & Review, 20(5), 878–896. doi:10.3758/s13423-013-0397-0.

Hanson-Vaux, G., Crisinel, A.-S., & Spence, C. (2013). “Smelling Shapes: Crossmodal Correspondences Between Odors and Shapes.” Chemical Senses, 38(2), 161–166. doi:10.1093/chemse/bjs087.

Majid, A., Burenhult, N., Stensmyr, M., de Valk, J., & Hansson, B. S. (2018). “Olfactory Language and Abstraction Across Cultures.” Philosophical Transactions of the Royal Society B, 373(1752), 20170139. doi:10.1098/rstb.2017.0139.

Majid, A., & Kruspe, N. (2018). Research on odor vocabulary and cross-cultural olfactory language, contextualized in Majid et al.'s broader work on how linguistic and cultural environments shape odor description. See also Majid & Burenhult's work demonstrating dedicated odor vocabulary among Maniq and Jahai speakers.

Spence, C. (2020). “Olfactory-Colour Crossmodal Correspondences in Art, Science, and Design.” Cognitive Research: Principles and Implications, 5.

Spence, C., & Deroy, O. (2013). “How Do Crossmodal Correspondences and Multisensory Processes Relate to Synesthesia?” In The Oxford Handbook of Synesthesia. Oxford University Press.

Winter, B. (2019). Sensory Linguistics: Language, Perception and Metaphor. John Benjamins.