I don't have the exact studies and I'm too lazy to look for them now, but I recall seeing a controlled trial on placebo effect showing that people who are (a) informed about placebo effect as such (IIRC that group was all medical professionals) and (b) informed that the pills they're taking definitely are just a placebo with no active substance, they still showed a placebo response. It was lower than for "normal people" who were told that the sugar pill is a medicine for testing, but still significant.
Huh, interesting. Because for the longest time I've been going "I can cause the placebo effect on myself by believing that it works" whenever I feel bad, but I've never been able to find any studies that'd confirm the effectiveness. I guess that if the placebo effect sticks around when you tell people they're taking sugar pills, though, it should work with just the self-referential invocation.
