Premium
Do panelists memorize products when performing descriptive analysis on few products?
Author(s) -
Lestringant Pauline,
Delarue Julien,
Heymann Hildegarde
Publication year - 2018
Publication title -
journal of sensory studies
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.61
H-Index - 53
eISSN - 1745-459X
pISSN - 0887-8250
DOI - 10.1111/joss.12305
Subject(s) - memorization , sample (material) , descriptive statistics , perception , context (archaeology) , product (mathematics) , computer science , set (abstract data type) , task (project management) , statistics , psychology , mathematics , cognitive psychology , engineering , paleontology , chemistry , geometry , systems engineering , chromatography , neuroscience , biology , programming language
Few studies have investigated the impact of the number of samples on the descriptive analysis (DA) process and results. One may hypothesize that when training on and evaluating small product sets, panelists memorize the products instead of actually evaluating the samples based on their perceptions of the attributes. In this study, we tested whether products were recognized during the training and evaluation phases of separate ketchup and lemonade panels. After performing a full DA task, panelists did a recognition test, in which they were asked to estimate the number of products they evaluated and which ones they recognized from a larger sample set. Panelists did not realize how many different samples they tasted, neither during training nor evaluation. Yet, they were able to recognize samples they had previously tasted in DA. Similarities among products led to mistakes. However, samples that were very different from others were not better recognized. Practical applications This study is a first attempt at evaluating the extent of product recognition in a realistic descriptive analysis context. The goal of sensory profiles is to measure human perception of products, which can be biased by memory effects. Since memorization may depend on the number of products panelists are trained on, we tested a “worst case scenario” with only four products during the training phase. This is relevant to many situations in which small sample sizes occur: a 2 x 2 experimental design, few innovative products present on the market, or limited number of prototypes available.