By scientific, I mean a controlled experiment, changing one variable at a time. If a test ride is involved where the rider describes a subjective feeling, he should be made unaware of what he's riding until after he gives his feedback.

These types of tests are rare, which makes it hard to separate the folklore from the truth.

Also, there are other factors, such as unintended consequences. I've read that the sound of radial spokes is different from that of crossed spokes, but in a subtle way. This can lead people to claim that the ride quality is different, though they are not actually experiencing a different in stiffness or shock absorption.