Wow! Good research. I've got a hard copy of it around here somewhere,
but after three moves, only God knows *where* around here.

You will find that the conditions used in the study were actually a
little extreme; the test aircraft was a high-perf retract, (Bonanza, i
think), and the did not give them an AI. The theory was to develop a
survival procedure that would work in any airplane, including those not
equipped with an AI.

I suspect that if the experiment were repeated using a 172 with an AI
you would get better results--but not that much better, maybe a 30%
survival rate rather than 0%. However, those results would, I think,
be unrealistically optimistic; the uncontrolled variable is the stress
factor. A private pilot in a study, under the hood, in VMC, with an
instructor next to him knows that if he really screws it up the worst
that is going to happen is he 'fails' the test. His life is not really
at risk, and he knows it.

Put that same private pilot in a 172, solo, or with non-pilot
passengers, in IMC, with a little turbulence and I think the stress
level will be *enormous.* If you could do a study where the pilot was
convinced that it was real IMC, with real rocks below, no instructor,
and no help available, then you might see some results that have
validity.

Listen to the guy's voice on the 17 ways tape and tell me what you
think.

Gene