jonah at ugcs.caltech.edu (Jonah Michaud) writes:
If this is a faq I apoligize.. From the LA Times, Jan 11, _The Mystery of
Memory_, by Steve Emmons:
"The brain is an unimaginable jumble of electrical circuits. Each of 10
billion brain cells connects with 50,000 others. One square millimeter of
cortex, the crinkly surfaced dome of the brain, contains 80,000 brain cells,
making the cortex the most complex electronic circuit board on Earth.
This means the brain's memory storage capacity is effectively unlimited.
You'd need many more than one lifetime to fill it up."
Is this true? How many cells would be needed [ . . . ]
The numbers are pretty accurate, except that the 50,000 figure is a
bit high -- 10,000 would be more like it. But this certainly does not
mean that the brain's memory capacity is "effectively unlimited". A
couple of years ago I took a shot at estimating the amount of memory a
human brain can usefully store. The number I came up with (after
guessing values for a few parameters that have never been measured)
was on the order of a terabyte (that is, one million megabytes).
This may seem like a lot of memory, but let's look at it. Suppose we
wanted to keep a record of everything we see. To be concrete, suppose
we needed to store a 100K JPEG image every second. (This is only a
tiny fraction of the information that comes in from the retina.) It's
pretty straightforward to work out how long it would take to exhaust a
terabyte of storage: the time is on the order of one year, even if
storage only goes on during waking hours.
Thus it's apparent that even if I've underestimated the brain's
storage capacity by a factor of 100, it still sets limits on the kinds
of computation the brain can do.
Is the statement "we only use 5% of our brain" true? And does it refer to
an untapped memory capacity? Is it consistent with natural selection- is
there a reproductive advantage to having excess "brain" or memory capacity?
Or does natural selection simply have no say in brain capacity above that
needed for hunting/gathering?
Nobody knows enough about how the human brain works to say what
fraction of it we use, but I think there is a strong argument that
it's a lot more than 5%. The argument is that humans suffer a much
higher childbirth mortality rate than most other mammals, and the
reason is the very large size of the human head. The large head is of
course there to make room for a large brain. If most of the brain
were unused, it seems to me that natural selection would very rapidly
get rid of the redundant parts.
The story I've heard (I'd appreciate any confirmation or denial) is
that the 5% figure comes from Karl Lashley's studies, in which he
lesioned the brains of rats, and observed how the lesions affected
their performance in various tasks. For cortical lesions, he found
that they had to be very large to make the rat's unable to do the
tasks. The current consensus is that Lashley's results were correct
but misleading, because he used tasks that could be done in many
different ways.
Also, one of my textbooks quotes 100 billion brain cells and about 10,000
connections, which differs significantly from the above. Is this a margin
of error?
According to the most widely accepted measurements, there are about 10
billion pyramidal cells in the neocortex, each having about 10,000
connections. Actually the majority of neurons in the brain are
granule cells in the cerebellum. There are about *40 billion* of
them, but they only make about 100 connections each. Thus in terms of
total number of connections the neocortex is dominant, with on the
order of 10^14 synapses.
-- Bill
