Okay, interesting article. But it has a few points, especially in a society where most technology we use isn't understood by many of the people in it. To quote the article:
In another experiment, the researchers demonstrated that young men and women instructed on how to use a voodoo doll suspected that they might have put a curse on a study partner who feigned a headache.
Why is it 'magical thinking' and a wrong assumption to believe in the voodoo doll experiment above, but if you are told by a 'doctor' that when you press the button he gives you, you're shocking the other participant in the study, that's okay to believe? That's the basis of the justly famous Milgram experiment, and as far as I can see the only difference is that it's somehow okay to believe in a random bit of science, but not in voodoo. Okay, in the Milgram experiment, the participant is shocked to demonstrate the effect, but would the absence of that proof render it 'magical thinking?' I doubt that many would class it so.
It seems to me that, if we are to expect people not to be sheep, we must and should expect them to believe the evidence of their senses, even when they get it wrong. If you can prod a voodoo doll, and the target reports pain - it could be a trick, or the hoodoo could be working.
If you can flick a switch and the light comes on - the technology might be working. Or, just possibly, it's a trick. How do we decide between those cases, save by preferring the comfortable case that conforms to our expectations?