bits and brains

JoshCahoon joshcahoon at cs.com
Tue Jan 4 05:38:42 EST 2000


>Subject: Re: bits and brains
>From: Bill Skaggs skaggs at bns.pitt.edu 
>Date: Mon, 03 January 2000 09:46 AM EST
>Message-id: <m7ln67b4cg.fsf at skaggs.bns.pitt.edu>
>
>joshcahoon at cs.com (JoshCahoon) writes:
>
>> How do people go about estimating the informational capacity of the brain
>in
>> terms of bits? And how do they apply bits to the capacity of consciousness?
>I'm
>> reading The User Illusion by Tor Norretranders, and he keeps contending
>that
>> consciousness processes about 16 bits per second, while the entire brain
>> processes about 11-16 million bits per second. Do ya'll think there is any
>> basis for these estimates?
>
>Are you sure you're stating these numbers correctly?  16 bits per
>second seems much too low -- that's less than 3 ascii characters per
>second!  
>
>As for the brain, the number also seems too low.  There are around 10
>billion nerve cells in the human neocortex, and I think most people
>would say that each one processes at least a few bits per second
>(though it isn't clear how much of this processing is actually
>*used*).  This would give on the order of 100 billion bits per
>second.  
>
>If the numbers you give were in bits per *millisecond* rather than
>bits per second, they would be more reasonable.
>
>	-- Bill
>
>
>
>
>
>

Yep. That's what the claim is, more or less. Here's some  quote, from The User
Illusion, by Tor Norretranders (it has a lauditory Hofstadter blurb on the
cover, by the way): [citing Human Physiology, Zimmerman "We can therefore
conclude that ethe maximal infomration flow of the process of conscious sensory
perception is about 40/bits/sec....Not because Zimmerman's analysis is
original...it simply repeats a conclusion that was reached ath the end of the
'50s and has been repeatd sporadically in the literature of medical,
psychological, and information theory ever since, though without making much of
a mark eithe r on physiology and neurophysiology" later, "Helmusht Frank has
very elegant arguments in favor of a capacity of consciousness of
16/bits/sec".....Norretranders presents arguments for several differents
estimates, but they're all in less than 100/bits/sec.




More information about the Neur-sci mailing list