The other day a piece fell off of one side of my glasses (the part that touches the nose.)
The glasses stay on, but I've noticed a weird feeling of imbalance at times. I could be imagining it, I'm able to function apparently regularly. But I was thinking that the obvious analogy is to filmography: directors consciously adjust camera angles and framings in order to induce certain emotions or reactions to a scene. It's plausible that even a very slight asymmetry in your vision can affect you.
If this is true, might there be other low hanging fruit for adjusting your perception to increase focus?
Does the anti-p-zombie argument imply you can't simulate humans past some level of fidelity without producing qualia/consciousness?
Or is there a coherent position whereby p-zombies are impossible but arbitrarily accurate simulations that aren't conscious are possible?
Yes, it implies that. The exact level of fidelity required is less straightforward; it's clear that a perfect simulation must have qualia/consciousness, but small imperfections make the argument not hold, so to determine whether an imperfect simulation is conscious we'd have to grapple with the even-harder problem of neuroscience.
How does it imply that?
I have intuitions on both sides. The intuition against is that predicting the outcome of a process can be done without having anything isomorphic to individual steps in that process - it seems plausible (or at the very least, possible and coherent) for humans to be predictable, even perfectly, without having something isomorphic to a human. But a perfect predictor would count as an arbitrarily accurate simulation.
Which anti p zombie argument? You can argue that no one has qualia, that physical duplicates necessarily have non physical qualia, that physical duplicates necessarily have physical qualia, etc
The argument that qualia can't be epiphenomenal.
The argument might have been "if qualia it exists, then it probably has observable effects - you without qualia would be different from you with qualia".
But obviously you as a simulation is different in some aspects from you in reality. It's not obvious that the argument caries over.
2) What aspects?
1) You are assuming qualia exists?
Causality is different, for one. You in reality has a causal structure where future actions are caused by the state of you in the present + some inputs. You in the simulation has a causal structure where actions are caused by the simulator, to some extent.
I'm not really assuming that. My question is if there's a coherent position where humans are conscious, p-zombie humans are impossible, but simulations can be high fidelity yet not conscious.
I'm not asking if it's true, just whether the standard argument against p-zombies rules this out as well.
Well if qualia aren’t epiphenomenal then an accurate simulation must include them or deviate into errancy. Claiming that you could accuracy simulate a human but leave out consciousness is just the p-zombie argument in different robes