Running thought experiments on the nature and function of consciousness is a fun pastime of mine, although I can't vouch for how productive it is a philosophical pursuit.
Imagine that we have a self-piloting car, of the type that is now being developed by google. Only this is a fully-functioning car that can take the "driver" from a back road in Aberdeen, Washington to the end of the Florida Keys without any input. But this car carries a passenger. The passenger, not needing to drive, merely looks forward at the windshield. But since this is a futuristic car, and the passenger has so very little to do, the windshield can also show a number of programs. In fact, since this is a future derivative of our google car, it probably has thousands of terabytes of programming available to put up on the windshield. Our passenger goes along merrily for the course of the ride, watching the windshield and never knowing (or thinking) if it is really what is outside the car or if it is just a projection to pass the time.
In case the analogy isn't obvious, the self-piloting car is a brain that is working with inputs and outputs, without some type of free will. The "driver" is consciousness as an epiphenomenon. "Epiphenomenon" in the philosophy of consciousness refers to consciousness that is an echo or function of brain states, without entering into their causal chain. "Epiphenomenon" can also be seen somewhat as the echo on a computer's boot screen. The messages that display on a terminal when a computer boots are not making the computer boot, and the computer could boot without them, but they do describe what the computer is doing. If our consciousness is merely an "observer" in a self-piloting car, one of the big questions is, why does it get a (mostly) truthful account of surroundings? There are two possibilities:
- First, that there is something about the way phenomena are processed by the brain or the mind into consciousness that is just universally "valid". When a photon hits the retina and the neural impulse is transmitted down the optic nerve and into the brain, the neuron "firing" just automatically and universally reflects what is "really going on".
- Our "windshield" isn't displaying what is "out there" but is instead displaying one of trillions of random programs. I know this is a stoner fantasy: "What if we are really like on some alien world and just imagining that we are on earth?" but its a fair philosophical question. If our computer is going to boot just fine without us, why do we need to read a screen of scrolling fixed width monochrome text, when we could just see a comforting image in earth tones?
If we do take the view that consciousness is an epiphenomenon, and that we are just a passenger in a self-piloting car, we have to accept one of those two answers for why our consciousness seems to be viewing a coherent world, even though that is not information it needs to know.
Of course, we can also decide against the notion that our consciousness is epiphenomenon, in which case it makes sense that it needs true, coherent sensory information as a way to make decisions about the world around it. If the driver is actually in control of the car, the windshield better not be showing a marathon of old Three's Company episodes.