First of all, we've seen that low-level physical models of our universe don't gain explanatory power by trying to take "consciousness" as an ontological primiti

First of all, we've seen that low-level physical models of our universe don't gain explanatory power by trying to take "consciousness" as an ontological primitive. It's clear that consciousness is a high-level property of patterns of tinier things. So the algorithm you use to decide that a clump of neurons satisfies the "consciousness" predicate will almost certainly work by observing high-level properties of the neuron configuration. Since the consciousness predicate abstracts away low-level details, it's hard to see why neurons should be better than other computational substrates at forming predicate-satisfying configurations. Reductionists can't be neuro-chauvinists.