[ExI] mersenne primes again

[Damien Broderick]

On 4/1/2010 1:36 AM, spike wrote:

> Do have a sense of humor, my published friend.  I was using the comment as a
> form of self-deprecating humor for scientific types to jab ourselves.  Yes
> it is preposterous: our current mathematical techniques encorage us to
> disregard findings we cannot explain.

Okay, and what I wrote (humorlessly) does indeed accord with your mild 
self-mockery, but goes further, I'd say. The way you use the word 
"explain" above is indeed typical of scientific thinking at any given 
time, but it's really used to mean "fits in with this current 
simplifying model". That's very sensible heuristically, and pays off a 
lot of the time, but it does risk excluding data that doesn't yet have a 
handy model. So I'm not persuaded statistics is the key to what's going 
wrong; psi research, for example, is replete with sophisticated stats. 
Such a phenomenon is excluded not because it only deviates from mean 
chance expectation at the 0.05 or even 0.001 level (say), but because it 
offends or exceeds the current model. The explanatory protocol's tail is 
wagging the dog.

Damien Broderick