Recession Relief for Pâté Lovers

Is your pâté consumption wreaking havoc on your pocketbook? A food processor and some Newman’s Own dog food may solve your problem. Economists at The American Association of Wine Economists conducted double-blind taste tests of five unlabeled blended meat products, including dog food. Subjects were unable to identify the dog food. (HT: Marginal Revolution) [%comments]


As the authors suggested, the panelists were primed to believe that the dog food would not be the worst-tasting product. This is a case of giving panelists too much information about what they are evaluating - giving them ample opportunity to form biases and take cues from variables extrinsic to taste, like prior knowledge and textural expectation.

trader n

The reporting is misleading.

It says that people cannot distinguish the dog food from pate, perhaps because they don't recognize the taste, but the overwhelming majority thought it tasted the worst.


Actually, 72% of the participants identified the dog food as the worst of the five samples, although they apparently couldn't make the leap to the fact that dog food should taste worse than pate.

Which makes me wonder: who are these people who can clearly identify which "pate" is the worst but, when told that one of the things they ate was actually dog food, assume that it could have been one of the better-tasting samples?


This is a very misleading post, since the test subjects overwhelmingly identified the dog food as the worst tasting of the five.


But 72% of subjects thought that the dog food tasted the worst of the five - worse than the spam or the liverwurst. 89% thought it tasted worse than both types of pâté. They may not know what dog food tastes like (the most common guess was that the liverwurst was dog food), but they still don't like it when they try it.


The entire study was statistically insignificant simply because your little sister did not include deviled ham, my dog's favorite topping, even though she cuts her tongue on the can. My dog, not your sister.


A badly designed study. When told they would be tasting dog food, and that it wouldn't be disgusting, guess after the fact that it wasn't the worst tasting thing they tried. This is a great example of expectancy bias, not so great an example of anything else.


Most thought it tasted the worst, so it would probably not make a very good substitute.


if the subjects were unable to identify the dog food, than the study actually proved a point.