By far the most important location to insulate properly is the outside of
the slab. This will also help to prevent frost from penetrating under the
slab in the event that there are periods of no heating in the slab.

Why is this? I have been unable to find any empiracal evidence of a
university conducting tests, or a trade group.

I'm considering embedding a fair number of 1-wire sensors to make some
real-world data available. I might even try one section of wall with
the radiant barrier and the next one with normal insulation to put that
controversy to rest.

I can swap it back next year when it doesn't work as well as they claim
it will.

Alan