CONSUMERS' TEXTURE PERCEPTION OF MILK DESSERTS. II - COMPARISON WITH TRAINED ASSESSORS' DATA.

ABSTRACT Considering the importance of texture in consumers' acceptance of many dairy products, particularly milk desserts, it is important to rely on methodologies to understand consumers' texture perception. The aim of the present work was to study the relationship between consumers' texture perception of milk desserts and trained assessors' data. Six commercial milk desserts were evaluated by a panel of 100 consumers, divided into two groups of 50, which evaluated the texture of the desserts using intensity scales for five texture attributes or a check-all-that-apply question. Concurrently, seven texture attributes of the desserts were evaluated by a trained panel. Results showed that both consumer profiling methodologies were able to identify differences in the texture characteristics of the desserts, providing very similar sensory maps. High correlation between textural attributes evaluated by consumers and trained assessors was found, indicating that consumers were able to evaluate the attributes in a similar way as trained assessors did. PRACTICAL APPLICATIONS Intensity scales and check-all-that-apply (CATA) question seemed to be good alternatives to the use of trained assessors' panels when evaluating consumers' texture perception. Although both methodologies provided similar information, CATA questions would be preferred over intensity scales for being easier because consumers only have to check the terms from the question that are considered appropriate for each sample, getting rid of the need of asking consumers to score the intensity of sensory attributes using scales. The application of consumer profiling methodologies has the advantage of being based on consumers' perception of the products. However, it is important to highlight that these methodologies should be considered complementary to trained assessors data and that they might be particularly recommended for those food companies that do not enough time nor resources to train a panel for a specific application. [ABSTRACT FROM AUTHOR]