Found this interesting tidbit:
This was a follow-up to an earlier, similar study testing whether intentionally “treated” chocolate would result in improved mood, also under double-blind conditions. Both studies showed that the treated substance resulted in better mood. The latest study also studied the role of expectation to see if it modulated this intentional effect. It did, to a highly significant degree.
The bottom line is that if you believe/expect that you are consuming a specially treated substance, that belief alone will strongly influence your mood. But if the substance is also intentionally “treated,” then it will influence you even more. And vice versa—if you don’t believe, you’re less likely to see any effect.
This is related to the sheep-goat effect, long observed in psi studies, and to placebo effects in medicine and to experimenter expectancy effects in a wide range of areas. These effects have not been warmly embraced in science or in medicine despite the evidence that they exist because of a core assumption that underlies much of scientific epistemology: objective measurements are supposed to be completely independent of observation or psychological factors. This assumption works well enough to be useful in many contexts, but it’s not universally true.