It's hard to say, since in this case any kind of blind experimentation would require memory erasure and a conspiracy to prevent me from finding out that variable reinforcement schedules produce the strongest effects. But that same info I'd need to be prevented from knowing makes my prior probability for a placebo effect rather low.

I'll also add to my results that the technique doesn't work where I have pre-established Ugh Fields. (In those cases, I couldn't even get started, so I have no information about what the success rate would be over time.) It's too late at night for me to try to work through whether this has implications for the plausibility of placebo effects.