A fair number of topics jostling for attention lately: slime moulds outfitted with skittish cyborg exoskeletons, Jim Munroe’s microbudget megasavvy take on nanotech, even this recent research on free will in fruit flies (which I’m wary of, but am holding off commenting upon until I’ve thoroughly read the original paper). And I’m in bitten-off-more-than-I-can-chew mode at the moment, so I don’t have time to put all that stuff on the crawl right now. But there is one thing that struck me like a bolt from the blue (except it was actually a bolt from an e-mail server) late last night, as I was trying to clear away the e-mail backlog:
There’s this guy, allegedly named Nick Alcock, who seems to know way more than he admits to. He first ruined my morning back in March by pointing out that if vampires really needed to eat people because they couldn’t synthesise gamma-Protocadherin-Y on their own, and if they needed that protein because it was so damned critical for CNS development, then women shouldn’t have working brains because the gene that codes for it is located on the Y chromosome. It was a shot across the bow I could not resist; we’re still going at it two months later.
One of the things we’ve veered into lately is the classic philosopher-wank question: if you’ve got a nonconscious zombie that natural selection has nonetheless shaped to blend in — to behave as though it were conscious (we’re talking the classic philosopher zombie agent here, not the fast killer-zombies under discussion a few days ago) — how could you detect it? More fundamentally, why would you bother? After all, if it behaves exactly like the rest of us, then the fact that it’s nonconscious makes no real difference; and if it does behave differently, then consciousness must have some impact on the decision-making process, findings about after-the-fact volition notwithstanding. (The cast of Blindsight mumble about this dilemma near the end of the book; it’s basically a variant on the whole “I know I’m conscious but how do I know anyone else is” riff.)
So this Alcock dude points out that if I’m right in my (parroted) claims that consciousness is actually expensive, metabolically, then zombie brains will be firing fewer synapses and burning through less glucose than would a comparable baseline human performing the same mental tasks. And that reminded me of a paper I read a few years back which showed that fast thinkers, hi-IQ types, actually use less of their brains than the unwashed masses; their neural circuitry is more efficient, unnecessary synapses pared away.
Zombie brains run cooler than ours. Even if they mimic our behavior exactly, the computational expense behind that behavior will be lower. You can use an MRI to detect zombies!
Of course, now Nick has turned around and pointed out all the reasons that would never work, because it is his sacred mission in life to never be satisfied. He’s pointing out the huge variation in background processing, the miniscule signal one would have to read against that, the impossibility of finding a zombie and a sentry (trademark!) so metabolically identical that you could actually weed out the confounds. I say, fuck that. There are places where the conscious and subconscious minds interface: I say, look at the anterior cingulate gyrus (for example), and don’t bother with crude glucose-metabolism/gas-mileage measures. There’s gotta be some telltale pattern in there, some trademark spark of lightning that flickers when the pointy-haired boss sends a memo. That‘s what you look for. The signature of the ball and chain.
Of course, it won’t be enough for this Alcock guy. He’s bound to find some flaw in that response. He always does.
Maybe I just won’t tell him.