[MURG] Corpus callosum bandwidth

[Eugen Leitl]

On Thu, Oct 23, 2003 at 10:09:53AM -0700, digfarenough wrote:
> I'd always thought the corpus callosum is just axons
> extending from various parts of the neocortex.
> Therefore, I would think that the coding in each fiber
> reflects the coding of whatever area the soma of that
> cell lies in. E.g., an axon from primary somatosensory
> cortex going through the corpus callosum would using
> whatever coding primary somatosensory cortex uses,
> while one going through from an association area would
> use whatever sort of coding the association area uses.

Thank you.
 
> Is that not what you're asking?

I was trying to gauge the bandwidth of the "cable". It seems to be dominant
information channel between both hemispheres, so that number is interesting.
It is of course pointless to just postulate one fiber=0.1 kBit/s, or some
other off-the-wall number without having the slightest clue about the coding
used. This task just got harder, as according to you there are many codings,
probably most of them uncharacterized yet.
 
> I'm also curious as to why you ask.

I've seen a TBit/s backplane 10 GbE switch today, and wanted to estimate how it
compares with the bisection bandwidth of the fattest cable in the human CNS.
The networking part is progressing nicely; the computation part is just
awful. About 3 GBytes/s best-case memory bandwidth in current high-end PC
mainstream, and a CPU weak enough so only a trivial number of trivial
operations can be achieved on a sliding cache-sized memory window at 
that bandwidth.

Just playing with numbers, in other words. No way to tie that to reality yet.

-- Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144 http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
Url : http://askja.bu.edu/pipermail/murg/attachments/20031023/bb51f448/attachment.pgp