[ExI] how large is a human mind?

Sat Jan 5 16:42:52 UTC 2013

Overall a very reasonable estimate, Anders, some objections below:

On Tue, Jan 1, 2013 at 8:02 AM, Anders Sandberg <anders at aleph.se> wrote:

> Now if (say) 25% of cortical synapses are individually located, that means
> around 3.75e13 synapses.

### My hunch is that the number of truly individually located synapses
is much lower than 25%. Think about the structure of your memories:
try to remember the earliest memory of your life. In my case this is a
story about being scared of a pig while vacationing with parents on a
farm, at age 2. There is a vague impression of dread, a ghost of a
memory of brick farm buildings, impending doom, impression of mud
being around. This can be constructed from generic mental imagery.
Even our love and hate experiences are run on more or less generic
neural structures, with predictable behavioral correlates  - what is
individual are the triggers.

I remember reading the article about single neuron activity having a
behavioral correlate, it is excellent. But this single neuron is in
this case located in a very generic cortical area - one devoted to
analysis of data from whisker movements. I would guess that if you
grafted the barrel field from one rat brain to another, suitably
connecting the long range projections, the acceptor rat would not be
much changed in its overall behavior, even though it would still react
to single neurons in the graft.

Most of the bulk of our brain consists of areas that are devoted to
low-level analysis - I am talking about the primary and secondary
sensory cortices, motor and premotor cortices, the insula, the
cerebellum. I would contend that most if not all of the circuitry
there is used for representations of fragments of experience that are
almost identical among most humans. Right now I am sitting at a
computer in a log home in the forest, looking out on hills through a
large picture window, with the sun barely clearing the hilltops and
nary a cloud in sight -a nice view, but certainly something
experienced before by millions of people. This view is built of
millions of hierarchically organized pieces, from single bits of data
in the retina, through color conjured up by the fusiform gyrus, to
concepts recognized in the parietal cortex. The pattern recognizers
responsible for this experience react predictably to commonly detected
patterns, and we know that from a large body of neuroscience research.
Various humans asked about their subjective experiences after being
placed where I am now would converge on essentially identical
narratives. These identical narratives produced from different pattern
recognizers that predictably react to stimuli imply that the
recognizers are functionally equivalent.

Since there are hundreds of millions of humans who already had
experiences very similar to mine, you could most likely use
representative examples of their pattern recognizers to replicate my
subjective experience, with only a sprinkling of links to my life
history needed to modify these generic thoughts. Since most of my life
history is made of fragments identical to what others have lived
through, a reasonable facsimile of this history could be built of a
thin tracery of links between generic mental images.

It is likely that using data exclusively from an individual synapse
mapping project would produce the most accurate representation of an
individual's experience but I doubt that the accuracy gain over
generic imagery would be large. Certainly I would be willing to accept
some degradation of my memories to save a few orders of magnitude on
storage costs. Every day we forget about a day's worth of memories,
stretching an ever thinner veil of a story over an ever longer span of
time. Furthermore, we keep changing and embellishing the stories of
our lives, so the narrative is in a slow but constant flux. Using
generic recognizers would allow me to maintain much brighter, more
detailed mental images of my past at a lower storage cost, and the
mild distortions introduced by generic imagery would not be different
in size from the distortions my memory keeps introducing on its own.

So my guess is that I would be satisfied with being represented by
less than 1% of my individual synapse count, with the rest contributed
by commercially available, off the shelf memories/mental modules.
Maybe I could stuff myself onto the hard drive of my laptop, if I
really cut all fluff out. Of course, YMMV.

------------------------
> The cost of storing and running a mind does not depend on absolute size but
> relative size and demand. If a mind requires m units of resources and M
> units are available, and there are N minds, we should expect the cost to be
> some increasing function of mN/M, likely concave (prices go way up when the
> M-brain is crowded to capacity). So we can model it as price = (mN/M)^a,
> where a>1. If the value of a mind is on average V, we should expect more
> minds being made until (mN/M)^a=V, or N=(M/m)V^(1/a). So the big determinant
> might be V and a, M/m just sets the scale.

### As you say, the relative size would matter but I would claim the
main determinants of survival would be size and value (V/m) compared
to competing minds, not compared to total available resources (M). A
mind with a higher V/m (squeezing higher utility out of less
GBytes+Tflops) would be more likely, ceteris paribus, to be copied and
run than other minds.

Rafal

>
>
>
> --
> Anders Sandberg
> Future of Humanity Institute
> Oxford University
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat

--
Rafal Smigrodzki, MD-PhD
Senior Scientist,
Gencia Corporation
706 B Forest St.

Charlottesville, VA 22903

tel: (434) 295-4800

fax: (434) 295-4951

This electronic message transmission contains information from the
biotechnology firm of Gencia Corporation which may be confidential or
privileged. The information is intended to be for the use of the
individual or entity named above. If you are not the intended
recipient, be aware that any disclosure, copying, distribution or use
of the contents of this information is prohibited. If you have
received this electronic transmission in error, please notify us by
telephone (434-295-4800) or by electronic mail
(fportell at genciabiotech.com) immediately.