According to a report in Scientific American Mind ("Paying for Pleasure," Apr/May 2008), 20 volunteers were recently given 5 supposedly different wines to sample. Each wine was identified by a certain price. In reality, only 3 different wines were sampled--2 of the wines were presented twice with different price tags. As expected, the subjects gave the wine with the cheapest price tag their lowest rating, while the most expensive wine was everyone's favorite. Subjects gave the "most expensive" wine a lower rating when that same wine was presented with a lower price tag.
While the subjects were sampling the wines, their brains were scanned using functional magnetic resonance imaging technology. The scans revealed an increase in activity in the medial orbitofrontal cortex while subjects were tasting the more expensive wine. This area is believed to underwrite or encode for the pleasantness of various experiences. Changing subjects' expectations via doctoring the price tags had a measurable effect on pleasure-related brain activity.
In other words, the wine really did taste better with an inflated price tag. The subjects underwent qualitatively different experiences when the wine was labeled with a more expensive price than when the same wine was labeled with a lower price. Granted, it should not have tasted any better, but that's another story.