The Super-Empirical (or Pragmatic) Virtues in Theory Choice

[StrangerDanger]

The single greatest obstacle to my acceptance of scientific realism is closely related to the underdetermination of theories by evidence.
Just so we're on the same page, here is a general statement of the epistemic thesis for scientific realism:

SR: Our mature and predictively successful scientific theories are at least approximately true of the world; furthermore, sufficient evidence justifies the existence of the unobservable entities these theories posit, i.e., genes, molecules, quarks, etc.

(there are attenuated iterations of SR in contrast to the one I have given, but let's stick to this one since it's the one I have difficulty accepting)

Now, a quick recap of underdetermination:

Empirical equivalency (EE): Virtually any theory has as its rival a host of competing theories whose claims do not differ at the observational level. How is this so? When we construct a theory to account for some data, this is usually done through an inferential process. Since theory construction is an inferential process, the theory must go beyond the data acquired through observation. Thus, since the empirical data does not entail the theory, there can be several different competing theories that each logically entails the same data.

Underdetermination (UD): Given EE, it follows that scientists cannot select one theory from a set of competing theories from the data alone. Rather, extra premises are required to select a theory in favor of its rivals. These are known as superempirical virtues.

Thus, since no body of evidence is sufficient to warrant the acceptance of one theory to the exclusion of all others (EE-->UD), theory choice must be made on non-evidential (i.e., pragmatic) grounds. It follows that our scientific theories are not chosen solely on an evidential basis.

But if it is true that theory choice is not based solely on the evidence, can we really say that our criteria for selection are demonstrably truth-conducive? Is scientific realism really plausible at that point?

So here's the central issue for me:

in cases of UD, when a theory choice must be made among a body of empirically equivalent rivals, are the superempirical virtues scientists praise--such as simplicity, explanatory power, comprehensiveness, elegance, conservatism, convenience, consistency, and abduction (inference to the best explanation)--to name of few, either a.) demonstrably truth-conducive, or b.) conceptual biases we single out in theories for their pragmatic value?

In sum, please answer the following: are the superempiricial virtues a.) pragmatic, or b.) truth-conducive? If b, then argue how they are demonstrably truth-conducive. If a, then do you happen to be an attractive single female who's into not so attractive constructive empiricists?

[Monocrow]

WOW!

A I say again 'A' philosophical view of your position might be that we descend from the trees to find 'THE TRUTH' Thus the pragmatic view itself is the search for 'the truth.' ART and Science are thus only true per se, as expressions of the search of the search. Her choice and my search are one, not as a science, but as an experiment-perhaps is a celebration of our temporal mutuality more than an exact science- PERHAPS!

[StrangerDanger]

Say what?