Let's consider three theories:
Which of these theories enjoys greater theoretical virtues and should be preferred over the others and why?
Against (1) and (2): by collapsing everything into either subjectivity or objectivity, they variously nullify the subjective/objective distinction. If we have some good reason to think ourselves to know that this distinction matters to the nature of the truth in this world, then to retain the distinction is a reason to reject (1) and (2).
For (1): if we don't have a direct perception of non-perceived reality (how would we?), and if there is no discursive reasoning conclusively to the contrary (Kant or the private-language argument notwithstanding, perhaps).
For (2): if introspection is the surest kind of (empirical?) knowledge.
One might also claim that (1) and (2) are equally better than (3) for positing fewer kinds of things on the pertinent level (either only subjective or only objective representations), or that (3) is better than (1) for having an explanation at hand for why things don't always go according to our subjective plans (an our-will/not-our-will distinction, then, and with will being prioritized among our subjectivities), or that (2) is better than (1) because there are intentional objects of so-called hallucinations anyway. Or one might think that (2) and (3) are either better than (1) if one thinks that God has revealed Itself to us, or has at least revealed us to each other (then per either revelation there is something besides our one mind, that is).
So perhaps which of the three seems most theoretically virtuous depends on what one's epistemic goals are. For more on virtue epistemology, see this SEP article (or see Keas[18], in which it is claimed that there are at least twelve major theoretical virtues; testability is not clearly listed among those, so the testability-or-lack-thereof of (1), (2), and (3) might not tell much for or against any of them, though).
