Sitting around a table with a friend or relative is probably the most enjoyable way to taste wine. I am sure no one will mind to taste wine for a living, but believe it or not, it can get a bit tedious if one needs to taste a large number wines at a time. There are different ways in which scientists get to evaluate flavours in food or wine. The key is to getting an objective tasting panel together and letting them taste the wines blind. I am sure you need very little motivation to taste wines (otherwise you will probably be now busy trying your hand at knitting, biltong making or the likes), but tasting wine blind might sound a strange. No worries, it only refers to the fact that the panel gets a glass with a unique code, with no further info supplied. The reason for this is that we are visual beings and our minds can play tricks on us by getting certain clues. Such pre-conceived ideas might thus influence your tasting and judging ability and to avoid this in our research a tasting panel always taste wines blind. Any communication during the tasting is also prevented by letting each panelist taste in a separate tasting booth.
When we ask someone in wine science to taste wine we require an objective, unemotional judgement that is repeatable over time. However, people’s perception/preference for a certain descriptor might differ quite dramatically. For instance, if say we want to assess the coffee intensity in Pinotage wine, one can get say 9 different answers from 9 different untrained panellists tasting the same wine. Now, I am not sure about you, but statistics were never one of my strong subjects at varsity, but even I know that any good statistician will tell you that such a situation will lead to you getting results that is a bit iffy. So, how do we avoid this from happening in wine science? We basically train a panel to taste wine like a machine (the line “I’ll be back” comes to mind), this technique is called descriptive analysis. In this case a panel of say around 10 persons will first be trained to recognize for instance the coffee aroma in wine and then given examples of wines with different intensities of the coffee aroma. Once they are all calibrated and agree that a strong intensity should be rated say around 8 out of 10 and a low intensity say 2 out of 10 one can test if they taste objectively and repeatable for the coffee aroma. (By the way, coffee is apparently one of the easiest aromas for people to identify).
We actually use a tasting sheet with a line (ranging from none to intense) next to each descriptor that needs to be rated by each panelist in descriptive analysis. By measuring where the panelist marked on the line one can then get an indication of the intensity of that specific descriptor. Now you might ask why not simply just ask them to give a mark out of 10? Apparently our human nature is rather to select by default a mark of 5/10, especially if the panelist is an introvert. By using a line scale without numerical values a bit of this psychological influence is removed.
Descriptive analysis thus yields data that can be correlated to the chemistry of wine for instance. This technique can be used for many food stuffs in food science research, be it wine, beer, fruit, meat, fish, sweets etc. This is a great technique and I do not mind sharing good wine with other panelists in the name of science, however I draw the line when it gets to sharing my Karoo lamb choppies!