"Eliezer S. Yudkowsky" wrote:
> Anders Sandberg wrote:
> > Can we simplify neurons? Probably; this is just how nature made them,
> > and I think it is possible to create more computationally efficient
> > input-output mappings retaining the same properties. This is an
> > interesting problem, both from a research, a theoretical and an
> > uploading standpoint. What can we do away with without losing the
> > important? How much detail is needed? Could you upload yourself into
> > an integrate-and-fire model? Could this process be continued on higher
> > scales?
> Nature has had billions of years to thoroughly exploit and deepen every
> possible quirk that could be used for computing power. Roger Penrose
> thinks that the fundamental units of computation are microtubule dimers,
> which would put the brain at 10^26 ops/second. He furthermore
> speculates that the brain uses quantum computing. I'm not sure I
> believe (guess-that) either, but I wouldn't be the least bit surprised.
I've noticed that Penrose is long on guesses and short on actual data. Lets see some microtubule signal studies that show actual computation. We don't know whats going on in there because we aren't developing any kind of a database of fine grained neuron to neuron signaling (or microtubule, for that matter) from which to analyse that actual computation is taking place, in what format, etc. Our present approach on that scale is tantamount to moving rowboats with battleships, or using the proverbial nuke to swat a fly.
When I sit in front of my TV, when it is on, I know that 'something' is going on in there, but nothing I see on the screen is telling me exactly what. I cannot infer a video receiver circuit merely from seeing that a display exhibits information, nor can I trouble shoot the circuit components on such a basis. I can develop many ideas of how the pictures came to be, from accurate theories all the way to theories about little men being trapped in the box, or it being a window to god..... just because I put nice fancy catch phrases in my description doesn't make it so.