[ExI] few bits per second

Ben Zaiboc bbenzai at yahoo.com
Sun Jun 27 10:23:24 UTC 2010


Ross Evans <ross.evans11 at gmail.com> wrote:

> All this is essentially pure speculation. We don't know how
> the brain
> encodes information, and as such to even talk in terms of
> bits and bytes is
> to fall foul of the assumption, that the brain can be
> shoehorned into a
> conceptual framework, that sees it deemed to be little more
> than a digital
> computer.

You're right of course that we currently have to speculate, but your assertion that "to even talk in terms of bits and bytes is
to fall foul of the assumption, that the brain can be shoehorned into a conceptual framework, that sees it deemed to be little more than a digital computer." is missing the point, I feel.  

I doubt that anyone seriously claims that the brain is 'little more than a digital computer', although it does display some signs of digital encoding.  The point is that digital computers can emulate non-digital systems to an arbitrary degree of precision (given enough processing power and memory), and we don't currently know how much precision and how much memory will be needed to emulate a human brain.  The estimates being thrown about are 'digital-equivalents', because talking in terms of bits is a very convenient way to quantify information.

I expect that a working brain emulation will not just be 'a computer program', in the sense that a word-processor is.  It will probably be an ecology of models of brain modules, which may themselves be interacting sets of smaller models - cortical columns, maybe - and only down at that level, or maybe lower down (individual cells, or even synapses) will you see anything like a computer program that deals in bits and bytes.  The vast majority of a brain emulation will be at higher levels of abstraction.

Ben Zaiboc


      





More information about the extropy-chat mailing list