Development of a non-verbal method for capturing and characterising fragrance evoked emotion concepts using visual imagery

Across the fragrance industry, there is a widely recognised need for new fragrance development directions that extend beyond benchmarks based solely on mean liking ratings. Germane to this, is the idea of non-valence aspects of affect that exist beyond liking that influence consumers' feelings and behaviours. If fragrance developers could access these non-valence aspects of affect, this might provide new product development directions that have not been considered previously. However, when questioned verbally, consumers struggle to project beyond liking. It is reasoned that image-based research methodologies might provide a way forward. Stage 1 describes the development of an image-based methodology linking 110 images to 85 concept descriptors. Analyses produced a four-factor model where Factor 1 relates to positivity versus negativity (valence) and Factors 2, 3 and 4, which are orthogonal to Factor 1, are characteristically non-valence (by definition). Further analysis revealed that Factors 2, 3 & 4 comprised 22 non-valence concepts. There is no evidence across the extant literature of other research tools that separate valence from non-valence concepts. Stage 2 involved consumer profiling of 79 commercial fragrances using a representative subset of 30 images. These were mapped onto the image concept space developed in Stage 1 to generate a concept profile for each fragrance. From this it was revealed how each fragrance evoked both valence and non-valence concepts and how fragrances that were liked equally could otherwise evoke very different non-valence concepts. These methodologies are described in detail in the belief that they may provide alternatives to liking-based approaches.