Designing a sensory trial for apples
Sensory trials of new apple varieties are carefully planned by the sensory team at Plant & Food Research (PFR). The number of apple samples each consumer tastes and the order they’re given to them can affect the consistency and accuracy of the results. Christina Bava of PFR explains important aspects of designing a sensory trial.
Questions to consider:
What are the steps mentioned to ensure consistent and reliable results?
What other procedures can you observe that contribute to reliability of the results?
How do these procedures help achieve greater reliability?
Transcript
Christina Bava (Plant & Food Research) When we design the sensory trials, there is quite a lot of thought that goes in to the process of how many samples of the control and the new variety apple that we give to them. So often we will give a number of samples of both the control and the new variety so that we are getting people to rate it more than once. That way we can see whether there is some consistency between their responses to make sure that they are giving us genuine responses.
So by control, we often use a product or, in this instance, a fruit variety that we know more about. So if we have something that is new and innovative and we don't know a lot about, we want to be able to compare it to a product that we know the market, you know, the sales for, we know how popular it is, we know what type of people tend to go for it. The control is always used as a baseline for each panellist so that we can understand their liking of a product that is readily available, and we use that as the control, and then when we give them the new sample, we can see how that compares.
We also give consumers the samples in a different order to make sure that there is no bias between the order that the samples are presented. It may be possible if we gave them all in the same order that the sweetness or the acidity of a sample was affecting the next sample. So we always try to ensure that our consumers cleanse their pallet between samples, so we often provide them with some water, some very plain unsalted crackers which they are able to eat between samples, and that tries to minimise any crossover effect between the different samples.
