Investigating Colour–Odour–Emotion Associations to Create a Framework for Fragrance Development

Humans are born with the biological capacity to smell, but odour perception does not have a fixed meaning. Instead, individual experience shapes how scents are interpreted, remembered and emotionally associated. Consequently, olfactory perception is subjective. Cis-3-hexen-1-ol, for example, may evoke freshly cut grass for one individual and green apples for another. However, research suggests that certain cross-modal correspondences between odour, colour and emotion are shared across individuals. This study investigated colour–odour–emotion correspondences in sensory perception and assessed their potential as a framework for fragrance development.

A three-stage, single-blind experimental design was employed. The colour palette was adapted from the Munsell colour system, comprising eight hues with four values and constant chroma, resulting in 32 colour segments. The emotional lexicon was derived from the Liverpool Emotion and Odour Scales and comprised 37 emotional descriptors grouped into seven categories.

Stage 1 examined fragrance-colour, fragrance-emotion and colour-emotion relationships using 24 fragrances and 212 participants. Chi-square analyses identified significant associations across categories, while post hoc analyses using standardised residuals revealed the direction and magnitude of these effects through strongly congruent and incongruent pairings.

Stage 2 evaluated strongly congruent fragrance–colour combinations identified in Stage 1 and their emotional associations using a seven-point rating scale with 191 participants. Significant associations (p < 0.05) from Stage 1 were selected for fragrance development.

In Stage 3, participants described a setting for which a fragrance was developed using one emotional category and three colour segments. Strong associations from stage 1 and significant associations from Stages 2 guided ingredient selection and concentration. Formulations were evaluated against participant expectations using a seven-point rating scale. All participants assigned ratings of, or close to, 7 for at least one fragrance developed using the framework.

These findings suggest that colour-odour-emotion correspondences can be used to create a framework for fragrance development.