A short, interesting article on the information capacity of the brain was written by nanotechnologist Ralph Merkle; see
http://www.merkle.com/humanMemory.html He gives figures between 10^9 bits and 10^15 bits. -- Ben G > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On > Behalf Of Alan Grimes > Sent: Tuesday, December 03, 2002 7:14 PM > To: [EMAIL PROTECTED] > Subject: [agi] How wrong are these numbers? > > > The functional unit of the cerebral cortex is the cortical column. > > A cortical column is roughly .5-.6 mm in diameter. (lets say that 4 can > fit in a square mm). > > The cerebral cortex is around the size of four sheets of regular paper. > Lets say the paper is 216x280 = 60,480 square milimeters. > > The cortex is the size of four sheets giving: 241,920 square milimeters > of cortex. > > Counting four cortical columns/mm we get: 967,680 cortical columns. > > Now things get fuzzy: > > How much information is in each column? > We know that the primary output of the column is through the firing of > the pyrimidal cell which yields a timed pulse coded signal. > > As a guess, lets say that the column stored 64-bits of relatively stable > state and 64 bits of transient state, or 16 bytes... > > rounding up the number of cortical columns, this would seem to indicate > that the capacity of the cerebral cortex might be as small as 16mb! > > Maybe that's too small lets be generous and give each column 1kb of > stable and 1k of transient state. Even still, we are only talking about > 2GB... > > >From other work I have done towards figuring out the cortex, I have > concluded that the cortex achieves nearly logarithmic storage efficiency > for the things that it bothers to remember. This is how you seem to know > a hell of a lot more than the actual information capacity of the brain > allows. > > If this is true at all, it is _VERY_ good news for AI devels... =) > > -- > pain (n): see Linux. > http://users.rcn.com/alangrimes/ > > ------- > To unsubscribe, change your address, or temporarily deactivate > your subscription, > please go to http://v2.listbox.com/member/?[EMAIL PROTECTED] > ------- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]
