Re: Your definitions.

You appear to be conflating ontological views (physicalism and dualism usually refer to these sorts of views, views about what kinds of things exist) with epistemological views. There is nothing in the definition of physicalism that requires us to have knowledge of the external world and nothing in dualism that requires us to give up rationality or science. You can be a physicalist and still think someone is deceiving your senses, for example. Also, this might just be me but 'materialism' should be jettisoned as outdated. "Materialism" means that you believe everything that exists is matter. But there is no reason to think that word is even meaningful in our fundamental physics. Thus I prefer "physicalism" the belief that what exists is what physics tells us exists.

Re: the relation to the simulation hypothesis

If you haven't you ought to read "Brains in a vat" by Hilary Putnam. It's just twenty pages or so. He argues that we cannot claim to be brains in vats (or in any kind of extreme skeptical scenario) because our language does not have the ability to refer to vats and computers outside our level of reality. When a brain in a vat says "vat" he is referring to some feature of the computer program that is being run for his brain. Thus he cannot refer to what we call the vat (the thing that holds his brain). I can explain further if that isn't clear. But one thing I got from the article is that we can understand the bizarre, muddled writings of substance dualists as trying to describe the vat! If you don't have any language that lets you refer to the vats you're going to sound pretty confusing. I find this pretty funny because the way Descarte's supposedly gets out of extreme skepticism is partly by trying to prove substance dualism! Irony!

Anyway, I'm a little confused by the invocation of the simulation hypothesis because while I'm willing to look at it as kind of metaphysical dualist hypothesis I can't see how our tools for learning the answer to this question would be in anyway different from our general scientific tools. Metaphysics, such as we can say anything at all about it, is just an extension of science.

(a) by observing the incompleteness of X (for example, the inexplicable deus ex machina appearance of random numbers) (b) by observing temporal, spatial or logical inconsistencies in X c) Privileged information given to them directly about X', built into the simulation in ways that don't need to be consistent with other rules in X

Why not just assume these were features of X to begin with? If I see an temporal, spatial or logical inconsistency I'm going to revise my understanding of space, time and logic in X. Not posit X'.

But what if we could point to just one thing that could not be explained in X? Just one thing that could not even be explained in theory because to do so would result in some contradiction in X?

We would revise our theory of X to remove the contradiction. I know you know this happens all the time in science.

I'm having a hard time dealing with the rest given the conflation between epistemology and ontology. Yes, if there are properties (like value and consciousness) that cannot be reduced to the fundamental entities of physics, then physicalism is wrong. However, it does not follow that Bayesianism is wrong, that empiricism is wrong or that the scientific method is invalid in certain magesteria.