Why winetasters can’t always see the forest for the trees
Professional wine evaluation is a fundamental example of perceptual learning, a process that relies on prior experience to improve our abilities, which results in long-lasting changes to our perceptual
For example, when an expert taster evaluates a wine made from Sémillon, their perceptual state includes not only the wine they are tasting but also previous wines they have tasted and their perceptions of those wines. In short, it involves far more than the immediate impressions the wine delivers to our senses; it is intrinsically bound to our prior experiences.
While perceptual learning plays an important role in evaluating wine, there’s another phenomenon related to perception that arises from the wine itself: perceptual interaction. When our olfactory system
is confronted with complex aromas, we often perceive them as a single aroma due to odor blending in a process known as configural perception (our perception of the smell of coffee as a single aroma is just one of many examples).
At the same time, we are able to isolate individual odors within complex mixtures, as in the case of detecting a fault in wine, through what is known as an elemental perception process. The rules that govern these processes are poorly understood, but the contradiction they present raises the question: How do these types of perception facilitate or complicate matters for wine tasters?
I was recently tasting a flight of eight young semi-sweet and sweet wines from Bordeaux, all of which were made from Sémillon, vinified with some percentage of botrytized fruit, and aged in oak.
While the wines were characteristically complex, they didn’t exhibit the classic markers of orange marmalade, candied orange zest, or overripe orange that are characteristic of the style and indicate the grapes were subject to noble rot.
As I compiled an aroma profile for the wines, the absence of this descriptor was puzzling to me. Recent studies concerning the aromas of noble rot–affected dessert wines have revealed the importance of a well-known phenomenon in perfumery, perceptual blending, that results in the perception of confected-orange aromas.
Researchers identified two lactones responsible in this case: One is a compound that’s associated with oak aging (3-methyl4-octanolide, a whiskey lactone that has coconut, celery, and fresh wood aromas), the other with Botrytis cinerea development (2-nonen-4-olide, a newly discovered lactone that’s oily, coconut-like, and rancid).
While it was evident that noble rot had contributed to the sweetness of the young wines I was evaluating, the two specific lactones that result in the perception of candied-orange aromas weren’t detectable in them. Yet it’s very likely they would emerge with age.
Configural perception can present a dilemma for olfactory experts of all kinds, as specific training and repeated exposure to odors mean that we are better at elemental perception of odor mixtures; we can be better at detecting the parts than we are at perceiving the whole.
This is where perceptual learning comes into play. Sensory experts are keenly aware of this adaptation and develop the ability to move fluidly back and forth between perceiving the individual elements of an aroma and perceiving the blend.