Re: Lilliputian Posthumians

Hal Finney (
Thu, 22 Oct 1998 11:23:51 -0700

Joe E. Dees, <>, writes, regarding the possibility of shrinking the brain:

> Obtaining the unpatterned materials is easy. Arranging them in a
> functional configuration is the hard part. Joe

We do have a model for how the neurons in the brain work. It is not complete, and it may not even be accurate, but it is based on the best information available.

Broadly speaking, neurons act as nonlinear weighted adders, with various points on the neuron being excited or inhibited by chemicals released by other neurons. The overall excitation level gets filtered by a nonlinear function, in many cases roughly a threshold filter, and this determines the degree of output excitation released by this neuron. At the same time, longer-range chemicals are floating around to modulate the overall sensitivity and excitation level of the neurons.

Based on this type of understanding, the actual information processing done by a neuron appears to be limited. You can approximate it crudely by a very simple mechanism. We don't know yet whether such a crude simulation would be adequate for intelligence, but given the inherent imprecision and associated fault tolerance of biological systems, it is plausible that this would work.

No one would claim that the matter is closed, but given our current understanding of neurobiology, it is reasonable to propose that the brain could be greatly miniaturized and accelerated while still supporting intelligence and consciousness.