How do crossmodal correspondences between color, sound, and aroma vary across cultures?

The pairings people make run through brightness and mood

The most consistent pattern across these studies is that people match sensory qualities by their emotional tone rather than by any fixed sensory rule. When listeners heard instruments playing scales and picked colors to match, their ratings of the timbres on words like bright, dark, and warm predicted how light and how saturated the chosen colors were [3]. In a much larger study, 278 participants heard excerpts crossing joy, sadness, and nostalgia with five musical genres; they mapped bright colors to joy and darker, more saturated colors to sadness, with lightness doing most of the work and hue depending more on the genre [4].

Sound and smell connect in a similar indirect way. In three experiments, participants sorted sounds and olfactory ingredients into emotional categories, then matched sound groups directly to odor groups; emotions such as stimulation and relaxation appeared to predict some of those matches, while others, like apathy and happiness, stayed ambiguous [2]. The researchers describe emotion as a possible indirect link between sound and smell rather than a proven cause — the matches were predicted by emotional categories, not shown to be produced by them.

Why meaning, not just the senses, drives the match

A color–sound study gives the clearest picture of the mechanism. Participants had to respond to a color or a sound while both were presented, and certain pairings — blue with a "drop" sound, yellow with a "shiny" sound — sped up or slowed down responses in both directions [1]. Crucially, the effect appeared even when people saw color words rather than actual colors, which suggests the link runs through the color category in the mind, not through the raw appearance of the light [1].

A follow-up experiment sharpened this. When participants had to label an ambiguous color somewhere between two hues while hearing a sound, the interference appeared only when they classified that color as belonging to the category tied to the sound [1]. In other words, the pairing bites when a person files the stimulus under a meaningful label. That is a strong hint about culture: labels and categories are learned, so where a culture draws its color boundaries could change when a pairing takes hold.

Where culture would enter — and why the evidence is thin

None of these studies set out to compare cultures. The color–sound work tested participants in one setting without cross-cultural comparison [1]; the sound–odor experiments likewise report no cultural comparison [2]; the instrument-timbre study notes that timbre–color matching is diverse and that its trends may only hold for certain instruments, without testing listeners from different backgrounds [3]; and the music–color study is explicit that its results describe the specific excerpts sampled, not broader categories [4]. A systematic review of sound and aroma associations in food and drink catalogues many experiments and their methods but does not address cultural variation either [5].

That gap matters because the mechanism these studies point to is semantic and emotional. If pairings are carried by shared categories and shared feelings, then they should be stable wherever those categories and feelings are shared, and should drift where they are not. The music–color results fit this: lightness tracked emotion consistently, while hue — the dimension most tied to naming and convention — depended on genre [4]. Genre is a cultural category, so hue is where cultural learning would most plausibly show up, though these studies did not test that directly.

What you can reasonably conclude

Treat crossmodal correspondences as real but conditional. Within the groups studied, people reliably link bright sounds to bright colors and to upbeat feelings, and some sound–smell matches follow emotional categories [2][3][4]. The color–sound interference effect depends on how a person categorizes what they see or hear, which makes the pairing a product of interpretation as much as perception [1].

The cross-cultural question, then, is not whether these associations exist elsewhere but which layer travels. Emotional brightness seems the most portable layer, since it showed up across many musical genres and instruments [3][4]. Category boundaries and hue associations are the layers most likely to differ, because they rest on learned labels [1][4]. Until studies run the same matching tasks with participants from different cultures, claims that these correspondences are universal — or that they differ — outrun the evidence.

Sources used in this answer

Related research questions