[ExI] Do digital computers feel?

Ben bbenzai at yahoo.com
Mon Dec 19 10:21:24 UTC 2016


I see this silly 'fundamental red' argument has been raised from the 
dead yet again. I thought we'd knocked it on the head ages ago, but 
apparently not.

Brent's 'fundamental red' concept has a glaring, fundamental flaw:
The idea that conscious perception of a colour (and presumably, by 
extension, any other conscious perception) is a 'fundamental quality'.

This simply cannot be. It's at odds with what we know about 
neurobiology, and it's logically infeasible. The sensation of 'yellow' 
is not a simple, single thing in our brains, it's a combination of many 
many things, varying over time and dependent on individual experiences, 
memories, associations of varying strengths and types, and variations in 
brain wiring. And probably other things we don't yet know about. It 
would be a very odd thing indeed if yellow, green, red, blue, orange, 
pink, that vaguely bluish-grey you often see in the clouds, etc., etc., 
not to mention literally billions of other sensations, were each 
'fundamental qualities'. That's stretching the concept of 'fundamental' 
to breaking point, for a start.

If a single neurotransmitter like glutamate embodied a single concept 
like 'red', how many neurotransmitters would we need? How many colours 
can we perceive? How many other sensations are we capable of perceiving? 
There's only one known way of representing such an enormous number of 
things, and that's by combining a smaller number of simpler (or more 
fundamental) things. We know how this works, are quite good at it, and 
see examples of it again and again in nature. Including in our brains. 
Biology is especially good at this, as anyone who's studied virtually 
any aspect of biology will know.

It's just as easy to say "purple" as it is to say "car", but neither 
thing is a 'fundamental' phenomenon. There are probably as many shades 
of purple as there are types of car, and I'd guess that the creation of 
the sensation of a specific colour purple in a person's mind is roughly 
comparable in complexity to the construction of a specific car.

Conscious sensation, at the very least, has to be a complex phenomenon, 
involving many neuronal events. The only 'fundamental' items here are 
spike trains (no, not trains belonging to Spike) or even individual 
action potentials.

An analogy with computers would be thinking that the characters in a 
video game were 'fundamental qualities' of the computer, while ignoring 
the fact that everything the computer does is a result of combinations 
of ones and zeros, which are themselves actually voltages on wires and 
in semiconductors. I know it's weird and a bit unintuitive that all of 
Halo, Minecraft, etc., boil down, fundamentally, to various combinations 
of 0 and +5 volts, but it's true.

Ben Zaiboc




More information about the extropy-chat mailing list