Attribute list panel training

Saint-Eve et al. (2011) investigated the question of the quality of the attribute list by proposing to different panels with the same level of training either an optimal list of nine attributes (based on previous work on the same product area), or a reduced list of seven attributes with important missing attributes, or an extended list of 11 attributes with two additional but non-essential terms. Results showed that the use of a list missing dominant attributes led to a biased description of the products. Subjects missing an attribute have actually no other choice than reporting their selection on another term, which dramatically modifies the sequence of perceptions. On the other hand, it seems it does not harm to have some additional attributes since they were almost not used. 

Quantitative descriptive analysis (QDA ) is a method that is often used. A panel of tasters first assesses a set of training samples and scores the intensity of a number of sensory characteristics (attributes) of the samples. There are many different attributes, and there is no unique, exhaustive list, but some that are commonly used to describe the texture... 

The next step of the training program is to develop a terminology system a common list of terms for the whole panel and of which the panel share a common understanding of the meaning of each term. Sensory concepts are aligned and recognition of specific attributes can be re-enforced by the use of appropriate chemical or physical standards. 

In our experience with FP, attribute elicitation is usually not a problan when working with experienced subjects or with product experts. Such subjects are indeed used to desoibing their sensations using descriptive terms. The only pitfall we met is when conducting an FP with panellists from a conventional profile panel heavily trained to describe the same category of products. Those subjects indeed tended to stick to the list of attributes they had learned before, and did not really adapt to the product set under study. 

---