gwern comments on (Virtual) Employment Open Thread - Less Wrong

35 Post author: Will_Newsome 23 September 2010 04:25AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (276)

You are viewing a single comment's thread. Show more comments above.

Comment author: gwern 08 June 2014 01:02:06AM *  6 points [-]

I don't think that's a good argument against itsunder9000: I've seen plenty of supposedly sophisticated people who you would expect to be able to distinguish from placebo advocate things which did not do jack for me in my own blinded or randomized experiments.

For example, the people who praise LSD microdosing have generally taken more psychedelics than I've even heard of; and what happened when I did a blind randomized self-experiment? No effects (1 trend toward benefit, 1 trend towards harm). Seth Roberts had been self-experimenting for decades longer than I have, and believed that treadmill usage benefited his spaced repetition performance; what happened when I did a randomized self-experiment while skeptical? Statistically-significant harm to my spaced repetition performance, the opposite of what this sophisticate's introspection told him. Seth Roberts thought vitamin D at night hurt his going to sleep and taken in the morning, improved his awakening the next morning; what happened when I did a blind randomized self-experiment? I found that he was right about the first thing, and wrong about the second (no benefit). And so on.

I didn't believe that much in the placebo effect before I started blind self-experiments. But I sure as heck do now.

(That said, I don't necessarily agree with itsunder9000. Big clinical medical trials are in unhealthy old adults, yes, but the psychology trials are usually done with that white lab rat of psychology - college students. And there are a few nootropics with enough studies to have some confidence in the claims.)

Comment author: VAuroch 08 June 2014 07:26:29PM -1 points [-]

It's definitely not a knockdown argument, it's true. It's mostly a large pile of circumstantial evidence that makes me slow to doubt it's effectiveness.