According to Hilary Putnam, if we can think that we are brains in vats, we are not brains in vats because we cannot think about things like vats without having discriminating causal relationships with real vats (Reason, Truth and History, p.16) .
How might Putnam’s argument apply to the claim that we could be living in a computer simulation: in other words, that the entire universe – including our brains – is simulated on a computer in some under-universe with which we have no direct contact?
A simulation is (to a first approximation) an implementation of a software object. So if we are implementations of software we will interact somehow with other implementations within the simulation. In fact (on Putnam’s assumption) that’s all we” be able to interact with and thus all we’ll be able to think about.
But this doesn’t entail that we can think that we are in a simulation because understanding that requires that we can think about the other bits of the under-world in which the simulation occurs. And we can’t do that for the same reason that we can’t think about our vats. Yet it seems entirely possible that we are a) in a simulation and b) that we can think about it.