Is Iodine useful? maybe

[Gwern Branwen]

I've put an iodine post up: http://www.gwern.net/Nootropics#iodine

I've been working on this off and on for months. I think it's one of
my better entries on that page, and I imagine some of the citations
there will greatly interest readers - eg. not just the general IQ
impacts, but that iodization causes voters to vote more liberally.

I also include subsections for a power analysis as guidance to
designing any iodine experiment, and a section on value of
information, tying all the information together. My general conclusion
is that it looks like I should take some iodine, but currently
self-experimentation is just too hard to do for iodine.

--
gwern

[Gwern Branwen]

On Tue, May 29, 2012 at 7:05 PM, Gwern Branwen <gwe...@gmail.com> wrote:
> My general conclusion
> is that it looks like I should take some iodine, but currently
> self-experimentation is just too hard to do for iodine.

I've just posted my more-or-less complete (still hoping for Fitzgerald
2012's IQ data) meta-analysis:
http://www.gwern.net/Nootropics#meta-analysis

The conclusion is _d_<=0.1, as expected. Still, a fun exercise, and
now my iodine section is kind of a model of what I'd like all
nootropics discussions to be like: in depth background information, a
meta-analysis, power calculations for experiment design based on said
meta-analysis, a value of information calculation, and if the VoI says
so, to wrap it all up an experiment. (Except no experiment in this
case, because it's too hard.)

--
gwern
http://www.gwern.net