I do think we have the ability to design a workable study program that is scientifically defensible.
I suspect that one study would be insufficient, and if it was sufficient, would be so large as to be infeasible.
But perhaps we could come up with a design for a series of studies. I’m impressed with the t shirt study, for example, as a modestly scoped but productive experiment. And reasonable to duplicate the experiment to validate results.
The goal would be to design (and refine and polish) a very grounded, parsimonious, and economic study with high objectivity and scientific validity.
Of course, it might take us a while, heh!