Good practices in sensory analysis

Written by admin1 April 15, 2020 0 comment

For every mistake there is a remedy

Having become aware of the physiological and psychological errors that can be committed during sensory evaluations, we can go onwards to describe some simple methods that will allow you to reduce their impact, and obtain an objective character to our sensory analysis.

Random coding: it is now common practice that the samples subjected to analysis are identified through a three-digit random code (e.g. 362). In this way, the judge is not able to gather any information to the sample and is prevented to associate the code to familiar numbers (e.g. happy or sad dates, favourite player’s shirt number, etc.) which may unconsciously affect the evaluation.

Order of presentation: all samples are served to the judges in a different order. For example, if we have three samples to serve (A, B, C), the first judge will receive them in a sequence (A, B, C), the second  will have another sequence (e.g. A, C, B) the third, still another different sequence (e.g. B, A, C) and so on. When the number of judges is equal or a to multiple of number of the possible combinations (in this case 6), we say that the sequence is perfectly balanced and the blocks are complete. Increasing the number of samples exponentially increases the number of combinations and therefore the number of judges necessary to balance and complete the blocks. For this reason, in some cases it is possible to operate with incomplete blocks or with dedicated experimental designs (which will be treated elsewhere).

Avoid judge fatigue: It is good to keep in mind that it is never appropriate to subject too many samples to analysis to avoid the strain on the judges’ sense organs, which will reduce their sharpness. There is no rule on how many samples we should present because the stress levels can vary between different products and ingredients (spices, alcohol, etc).

Stabilization: it is necessary that the sensory sessions are performed in controlled environments where the experimental context is kept constant. The ideal solution is the use of special evaluation booths (sensory booths) or to carry out the test in dedicated environments in which the judge is not “distracted” by external factors or influenced by other judges.

Calibration: Judges, like any other analysis instrument, must also be calibrated. Calibration applies both in qualitative and in qualitative terms. For instance, the sensory meaning of the attributes being assessed must be well defined (e.g. smell of straw, smell of hay and smell of stable). In quantitative terms, if the evaluations are performed on a scale, all the judges should evaluate a stimulus of equal intensity in a similar way.

Interpretation: it is a general evaluation that the panel leader must perform on the overall trend of the sensory session. Therefore, it is necessary for the panel leader to have a critical spirit that goes beyond the mere statistical result, to understand if and what external factors can influence the analysis.