[Originally published in November 2007]
Both Greta and I are big wine fans. Despite Jonah’s recent extremely popular post, I, at least, believe that I can tell the difference between good and bad wines. I’m still convinced that a good wine is more than just an attractive label (though I’m a sucker for labels with Zinfandel puns like “Zen of Zin” or “Amazin”). That said, the research suggesting that labeling has a lot to do with wine preference is also quite convincing.
Several studies suggest that we expect to prefer wines from certain vineyards or regions, and in many cases wine drinkers will actually rate the identical wine higher when it’s presented in a fancier bottle. These results apply not only to wine, but to a variety of foods. Restaurateurs have known this for years, placing special emphasis on the presentation of the food in addition to the actual preparation and ingredients.
So if presentation matters, then perhaps the presentation of wine could actually affect the taste of the food it’s served with. This is the premise of a study by Brian Wansink, Collin Payne, and Jill North.
In their first experiment they served 49 graduate students cheese and one of two types of wine as they arrived at a reception. The wine — in both cases the identical cheap Cabernet — was served in bottles labeled as being from California or North Dakota. Prior to drinking the wine, they rated its expected tastiness on a scale of 1 to 9. After sampling both wine and cheese, they rated both of them for actual taste. Here are the results:
Both expected and actual tastiness ratings were significantly higher for the “California” wine — despite the fact that the wine bottles had identical, professionally-designed labels from the fictitious “Noah’s Winery.” What’s more, the tastiness ratings for the cheese, which was the same, unlabeled mild goat cheese for everyone, matched the wine ratings. When people thought they were drinking better wine, they also liked the cheese more.
In a second experiment, 51 patrons of the Spice Box, the white-tablecloth campus restaurant at the University of Illinois, were served identical prix-fixe meals. When they were seated, the waiter at each table made the following announcement:
Thank you for joining us tonight for this special meal at the Spice Box. Because this is the first meal of this new year, we are offering each person at the table a free glass of this new Cabernet from the state of California [or North Dakota].
The wines were labeled as before, but the patrons were all actually given an identical glass of Charles Shaw Cabernet Sauvignon (the to-my-view undrinkable “Two-Buck Chuck”). Their meal and drink portions were carefully weighed in the kitchen before and after consumption, so that the amount consumed could easily be calculated. Here are the results:
The patrons who thought they were drinking “California” wine consumed significantly more food (p = .02), and marginally more food and wine combined (p = .08). There was no difference in the amount of wine consumed, probably because everyone was limited to one glass.
One possible explanation for the result is social facilitation — diners might have felt compelled to eat more if their table mates were cleaning their plates. However, the researchers controlled for social facilitation and found that the labeling difference still explained the differences in amount of food consumed.
So apparently if we believe we’re drinking better wine — whether or not it’s actually better — we’ll not only think the accompanying food tastes better, we’ll eat more of it too. Maybe the next time I try to lose weight I should start drinking bad wine!
Wansink, B., Payne, C., & North, J. (2007). Fine as North Dakota wine: Sensory expectations and the intake of companion foods. Physiology & Behavior, 90 (5), 712-716 DOI: 10.1016/j.physbeh.2006.12.010