Re: We can all think we are unbiased
Agree with the first part, but placebo in drug trials is a poor example of unconscious bias. People in blind trials with placebo generally do respond honestly - it's just that placebo frequently works as well as the drug on trial (that's precisely what the trials are designed to test for - a drug is not deemed effective unless it outperforms placebo). Placebo is powerful medicine - strange, but true.
OT: Homeopathic remedies are a fine example; for those people that believe they are effective, they frequently are. However, it's hard to perform satisfactory double-blind trials on homeopathic remedies (that's the remedies themselves, as opposed to the full treatment process) for the following reason: a plausible explanation for the apparent efficacy of homeopathic treatment is that it is down to a combination of the beliefs of the patient and practitioner, and in particular the level of personal attention involved (as compared, say, with typical medical practice). Therefore, to test whether the remedies themselves play any part whatsoever in the efficacy, the practitioner, as well as the patient (and, of course, testers) would need to be unaware whether they were using placebo (water), or a homeopathic remedy (water... sorry, my bias). It seems that most homeopathists object to this, on the (I guess kind of understandable) grounds that they would not be able to apply effective treatment if they were not confident they were prescribing a "real" remedy. The only way to perform a satisfactory trial, then, would be to dupe the practitioner (and patient) by sometimes replacing the remedies with placebo. This is not, however, considered ethical.