Neither. The data needed to make a bike out of any material with the desired ride characteristics already exists. It is used all the time by the engineers and frame designers in the industry. Just as, say, in the auto industry. My point is that generating new data is not going to solve the mystery of frame material. No mystery exists. New data is not needed to convince the layperson that their misconceptions about frame materials are mistaken. There is already more than sufficient knowledge out in the world to convince an educated and objective audience. The issue is that the majority of the audience in these questions are neither of these things, and will reject data that refutes their beliefs.

My point is that any study such as that proposed by the OP would be redundant and ineffectual. There's nothing in those questions that warrants such attention, especially when it's unlikely that an experiment could be designed to isolate the variables of interest.

That's not to say that there aren't quirks of bicycle handling and ride that don't pop up and surprise us (that's why we do prototypes), and I should hope that the data on these are recorded, since they are useful for future design work. But the question of how frame material affects comfort is neither interesting nor unanswered.