From the Daily Telegraph, wine tastes better in blue- or red-lit rooms. At least in Germany. [Original paper here is quite well-written and reveals some interesting details]
According to the newspaper reports, the psychologist who ran the study (Dr Daniel Oberfeld-Twistel) thinks the explanation is that the light puts people in a better mood. But the paper cites another study showing that in fMRI tests:
Whenever colors and odors were presented together that subjectively constituted a good match in the eyes of the observer, activity in the orbitofrontal cortex and in the insular cortex was observedNow we all know to be cynical about fMRI tests, but it seems reasonable to draw the very broad conclusion that there is some cognitive basis for this effect. I am inclined to believe that - like most behavioural wine effects - it originates in a self-fulfilling calculation of the expected "value" of the wine and is a cognitive rather than a mood effect. Indeed, the paper says a little later:
The complexity of the color-odor interaction may be founded in the fact that the process appears to be located at higher cognitive processing stages.However, my view is somewhat challenged by the fact that the experiment used opaque black wine glasses to eliminate the direct effect of the light on the wine's colour.
It's certainly possible that a cognitive "value" decision could be influenced by lighting, though I would expect the effect to be weaker than the impact of the colour of the wine itself.
One experiment was run in a winery and two others in a lab; the effect seemed to be clearer and stronger in the winery. If cognitive factors are involved this does not seem surprising, because value judgments would be influenced by assumptions about the quality of the winery as well as the product itself.
The authors mention a different cognitive effect, which is that blue light makes people more alert and thus affects cognitive processing. I guess that might be true too, but the paper is sceptical about it and so am I.
It would be interesting to test the emotional/mood hypothesis by affecting mood with sound instead of colour. I think the effect of sound on mood is much stronger than that of colour, but as far as I'm aware there is no known effect of sound on subjective perception of taste.