Products selection strategies to better identify sensory key-drivers in a preference mapping, case study on apples

Products selection strategies

to better identify sensory key-drivers in a preference mapping,

case study on apples

Charles M.abc_{, Maître I.}a*_{, Mehinagic E.}a_{, Prost C.}b_{, Symoneaux R.}a_{, Ouvrard C.}a_{, Vigneau E.}de

a _{LUNAM Université, Groupe ESA, UPSP GRAPPE, Angers, France}

b _{LUNAM Université, ONIRIS, GEPEA, Equipe Flaveur, Nantes, France}

c _{Department of Food Quality and Nutrition, Research and Innovation Centre, Fondazione Edmund}

Mach, Trento, Italy.

d _{LUNAM Université, ONIRIS, USC ‘‘Sensométrie et Chimiométrie’’, Nantes, France} e _{INRA, Nantes, France}

Corresponding author:

Isabelle MAITRE

Groupe ESA, laboratoire GRAPPE 55 rue Rabelais

49000 Angers 33(0)2 41 23 55 55

i.maitre@groupe-esa.com

Keywords: (3-6 preferably): preference mapping, apples, sensory analysis, preference key-driver,

Products selection strategies

to better identify sensory key-drivers in a preference mapping,

case study on apples

Abstract

A good knowledge of consumer expectations is a key point to drive fruit & vegetables cultivar

selection. Defining the ideal product for consumers requires to implement sensory descriptive analysis and hedonic consumer tests in parallel. Preference mapping, modeling consumer liking scores with sensory descriptive data, identifies different groups of consumers and their respective sensory key-drivers of liking, and gives a clear sensory target to drive selection. Both the appreciation of

consumers (liking scores) and the sensory descriptive data depend on the selection of samples. What is the impact of samples selection process on the results and what could be the strategy to detect secondary sensory key-drivers ?

First, 31 apple cultivars were precisely described with 15 descriptors using a conventional sensory profile. In parallel, product liking were evaluated via a hedonic test with 224 consumers.

Then, according to this sensory characterisation, three product-spaces of the same number of products were selected: a first one composed of cultivars, representative of the whole sensory variability of apples, a second one, focusing on cultivars with a similar texture and a third one including apples with same acidity level.

The largest product space and the three product-spaces were analysed with the same methodology. The different consumer segments and their corresponding sensory key-drivers were identified using a CLV integrated approach (1).

The results from the largest product space (31 products) showed that sweetness and aromatic richness were important key-drivers of liking for all the consumers’ clusters and the main attributes distinguishing the clusters were the texture attribute crispy, juicy and soft but also the sour taste and some aromatic attributes as green, pineapple and banana.

The preference analysis in the sub-space representative of the initial product-space gave similar results, as far as the sub-space is really representative of the 31 ones. When exploring each of the two other product-sub-spaces, we showed different consumer segmentations and preference key-drivers identifications. When consumers tasted products which were similar for one of the sensory dimensions, the impact of other ones on liking was highlighted and could be specified. Flavour attributes in particular were put forward in some segments.

This underlines that results depend on the product-space choice and that this choice is crucial in preference studies. In addition, it indicates that there is a need for a deeper insight in texture-taste and aroma interactions.

3

References:

(1) Vigneau, E., Qannari, E.M. (2002). Segmentation of consumers taking account of external data. A clustering of variables approach. Food Quality and Preference, Volume 13, Issues 7-8, 515-521.