New physiology based training programmes for improving effectiveness of the sensory benchmarking method

S. Kemény (1), A. Hegyi  (2), T.  Kuti (2)

1 Technical University Budapest,

2 Campden and Chorleywood Food Industry Development Institute, Hungary

 campdencampdenkht.com

This research aims to study improving the reliability of the sensory benchmarking method, which is a wide spread in quality control. As a result of the target of the project we developed different training techniques based on psychological principles, because the previous experiments (ref.:   ) showed that the training of the assessors influences the repeatability and the reproducibility.

The experiment was carried out in four different ways firstly by using detailed description of the product attributes without and with visual aids, secondly self-made training software to acquire and study the recognition of product attributes and subliminal pshicodynamic activation, in which short stimuli about the product attributes were showed in documentary film.

The experiment was repeated with 12 trained and untrained assessors. The assessments were carried out with three replications in 4 different frozen pea samples.  The assessors examined 4 frozen pea samples from different bulk at a panel session. According to the experimental plan, every panel evaluated same bulk of samples.

After making sure that the data showed normal distribution we applied ANOVA to evaluate repeatability and reproducibility of the sensory panel and individual assessors in the panel.

In conclusion, there were no significant effects of the four different training techniques on the performance of the trained panellists. However the performance of the untrained panellist significantly improved by the usage of self-made training software and got closed to the trained ones, especially the ratio of colour defects. The subliminal pshicodynamic activation did not have effect on the reproducibility and reliability of results.

Using visual aids reduced the ambiguity of the results of the untrained panel compared to using   description of the product attributes.

Finally we draw the conclusion that these training techniques can be useful in the industry to acquire the relevant sensory knowledge about the product quickly.