Thursday, October 12, 2017

Consciousness in transitions

We can think of a digital computer processor as doing two things: Transitioning between states and remaining in a constant state between the transitions. How long the processor remains in a constant state depends on the clock rate: after the processor has done a flurry of computation (“combinatorial logic”) in a clock cycle, it will stay in a constant state until it’s time for the next flurry. If the clock rate is low, it will be able to stay in that constant state for a significant portion of the time, which is great, because presumably then the processor will be cooling off.

Suppose the computer is conscious by virtue of computation (as opposed to, say, being conscious by virtue of the functioning of a soul that God creates for it). When is it conscious? Is it during the transitions between the states or while remaining in a state? Intuitively, it should be during the transitions. After all, while it was remaining in a state, we could suddenly lower the computer’s temperature to near absolute zero. That wouldn’t disturb the computer’s remaining-in-a-state.

(Granted, it would disturb the computer’s clock. But the clock seems something extrinsic to the conscious system. One could in principle run a processor—very slowly—on a clock signal produced by a human being tapping a telegraph key, and surely that wouldn’t make the human’s hand a part of the conscious system.)

But the frozen state is functionally very much like the processor’s regular holding of a state when it waits for the next clock pulse. So just as it is implausible to think that a physical system like a computer that is frozen near absolute zero would continue to be conscious, it is implausible to think that the computer would be conscious while simply holding a state.

Thus, if a digital computer is conscious by virtue of computation, that consciousness occurs in and through transitions between states.

So what? I don’t know. I’m just trying to figure out what the best functionalist view would be like.

And note the contrast between this picture of consciousness-in-transitions and classical theism, according to which consciousness occurs in a timeless state.


Michael Gonzalez said...

This is an interesting point. I do wonder though if functionalism can possibly be worth pursuing if it can (even conceivably) incline us to think that consciousness is due to computation.... I think any view that says the difference between a conscious animal and an animal rendered unconscious is whether it's engaged in computation (or, presumably, its brain is engaged in such... though that's even more problematic, since brains don't engage in computation at all) is a view that will transgress the bounds of sense. It will be nonsense.

Or, if you don't mean "conscious" as in "not asleep" or "not rendered unconscious", then perhaps you mean "to be conscious that X". But then that is clearly a matter of having one's attention caught and held by X or to factor X into one's deliberations or to be aware of X. None of these meanings have anything to do with computation, nor are they the sorts of things that brains can meaningfully be said to do.

I really don't see how any progress at all will be made on the matter of consciousness until we realize that it is animal organisms which are conscious (in both senses: "not unconscious" and "conscious of X"); not their brains or souls or minds or any other such thing. Once we get that right, then we won't wonder about computation (since it's rather clear that an elk cannot compute and yet it can become very conscious of a predator in the vicinity).

Michael Gonzalez said...

Correction: When I said "conscious that X", I meant "conscious OF X". An elk is conscious OF predators in the area.