I like wine, but I’m no connoisseur. I read the descriptions on the labels; they make the wine-drinking experience more exotic. A cabernet sauvignon “brings aromas of plums and cacao giving a good structure,” while a merlot has “aromas of black fruits and silkiness.” A chardonnay “showcases flavors of ripe pear and apple, complemented by notes of vanilla and a touch of oak.”
I don’t know what any of it means. I just know what I like.
In their new book, The Biased Mind, French economists Jérôme Boutang and Michel De Lara describe the process that expert wine tasters go through to assess the quality of a particular vintage. First, they do a visual inspection, noting the color and clarity of the wine and assigning it a visual score on a scale from 0-100.
After that, they smell the wine, rating it on characteristics such as “intensity,” “fruit,” “spices,” and “complexity,” using the same 0-100 scale. Averaging these numbers together forms a nose score.
Only then do the connoisseurs actually taste the wine, assessing it on qualities such as “soft tannins,” “concentration,” “sweetness,” “balance,” and “finish.” Again, the average of these individual ratings produces a mouth score.
While appearance, aroma, and palate are the three aspects that wine tasters consider, these do not earn equal importance. Instead, the experts multiply each rating by a percentage: 25% for the visual score, 25% for the nose score, and 50% for the mouth score. They then add together these weighted scores to yield an overall mark.
Using this process, connoisseurs arrive at very similar ratings for the same wine. You might think that such an intentional process helps guide the connoisseurs toward an accurate assessment. However, when expert wine tasters simply give their first impression of a particular wine on the same 0-100 scale, they also tend to agree with each other. Further, these spur-of-the-moment, or “hedonic," assessments are usually close to the scores resulting from careful analysis.
So why even bother with the score card?
The assessment process reflects the complex knowledge structure that connoisseurs have about wine. However, these findings suggest that experts really use their intuitions to judge the wine in either case. It’s just that they also have the vocabulary to rationalize those intuitions.
Still, you may argue, surely extensive experience has honed the experts’ intuitions. Don’t be too sure about that. When amateur wine drinkers are asked to rate a vintage on the same 0-100 scale, they give scores that are similar to the experts'.
If we tell the amateurs that experts rate wine on appearance, aroma, and palate, they too can assess these three characteristics. Yet they give different weights to each, typically about 10% for visual, 20% for nose, and 70% for mouth. Still, when they tally up the scores, they get a rating very similar to what an expert would have given.
So, you might ask, do the experts really know anything different than the rest of us?
In fact, the experts know a lot. For one thing, they understand that appearance and aroma influence our taste experience more so than an amateur would assume. This comes as no surprise to cognitive psychologists, who know that our perception of the world involves a melding of the senses, with each influencing the other to a significant degree. Experts can also identify many subtle aspects of a wine’s appearance, aroma, and mouth experience to which the amateur is oblivious.
Research in the area of judgment and decision-making has long shown that experts are just as susceptible to cognitive biases as non-experts. Regardless of the formal decision-making processes we put in place, in the end...we go with our gut. Whatever the math says, the decision still has to feel right.
Then do we even need to rely on experts to make decisions? As Boutang and De Lara point out, experts do know a lot about their field. In particular, they have a good sense of which variables are important and which are not in making a decision. For example, doctors generally know which symptoms to consider in a diagnosis.
However, experts aren’t very good at weighing multiple variables simultaneously, something that’s often necessary in complex decision-making. Thus, a fairly simple mathematical model—linear regression—will invariably point to a better decision than even a group of experts can make. In brief, we need the experts to point out the relevant variables, but then we'd be well-served relying on a computer to crunch the numbers on those variables.
As a case in point, Boutang and De Lara consider personnel selection. Intuition tells us that employers at least need to interview a short list of candidates. Yet they point to research that shows this “human touch” is flawed. As soon as we meet people face-to-face, all sorts of cognitive biases kick in. We’re much better at evaluating candidates according to a list of objectively quantifiable variables, such as years of education, number of projects completed, and so on.
We all like to think of ourselves as experts in one area or another. Yet more knowledge doesn’t mean better decision making; it only means better rationalization of our intuitions. We honed these intuitions during the eons spent as hunters and gatherers on the African savannah, but they don't always suit us well in our modern lifestyle.
We should never forget that the driver of our connoisseur mind is a caveman brain.
Boutang, J. & De Lara, M. (2016). The biased mind: How evolution shaped our psychology including anecdotes and tips for making sound decisions. Cham, Switzerland: Springer.
David Ludden is the author of The Psychology of Language: An Integrated Approach (SAGE Publications).