Incentives to experimentally confirm theories after tons of experiments with various perspectives have already been run

Question

It appears throughout history that the glory goes towards who proposes a theory and outlines the math behind it, the first person to experimentally test it, but not all the hundreds of experiments afterwards.

For the first group, there is a great incentive. For the 2nd or 3rd groups, who may have started around the same time but weren't done when the 1st group published, the incentives are still there. Scientists continue coming up with different ways to test the theory and publishing papers, life goes on great. But at some point you have too much evidence and yet people keep adding to the pile.

For example, special relativity has been tested by hundreds of experiments and is continuously tested everyday by satellites orbiting Earth. Is there any reason for someone to have added the 201st paper on the topic? I guess not, but someone did. Why are those papers published then?

On a related note, the 2nd or 3rd confirmation is still doing a great service. However, does the scientist behind it ever feel bad no one (except their colleagues) remembers their name, only the person who invented the theory and perhaps the first tester?

Answer 1

In general, those "confirmatory" tests are performed over time as experimental techniques get better and enable greater accuracy- so the experimental team gets to narrow down the uncertainty in the original numbers, and perhaps add another significant digit to one of those measured numbers.

Furthermore, it is an indication of new physics at work when a number measured in an experiment begins to diverge from the one predicted by a model after a few significant digits. For example, although the gyromagnetic ratio was set by Dirac's electron model at exactly 2, the experimental value is 2.002319... with the difference being due to (tiny, but measurable) corrections arising from quantum field theory, which were not included in the Dirac picture. In another example, the Lamb shift (a tiny difference in energy between two electronic energy levels experimentally measured in the hydrogen atom) was not predicted by the Dirac equation, according to which those energy states should have been exactly equal. This difference was also an indication of new physics arising again from quantum field theory.

These examples furnish a strong motivation to push ahead with new, more accurate measurements of old numbers.

Finally, imagine that a completely new technique is invented by which one of these "old numbers" can be experimentally measured, and imagine further that the new test confirms the old number. This means that any shortcomings which might have been lurking in the original experiment can be ruled out, bolstering confidence in the original technique and the number it measured.

Answer 2

I agree with anna v in the comments, and I would also add the following.

I think it seems to you that there is a significant, wasted effort to reconfirm things that have already been shown. But how in-detail have you understood what was being tested? Maybe the experimentalists are just looking at the subject with a finer eye to detail. For example, you mentioned that special relativity has been tested often enough. But there is, in reality, no such thing as "testing special relativity" as a whole. One can only do very specific experiments to test particular phenomena. And if you would assume that, having tested (say) 50 phenomena across a moderate range of parameters is sufficient evidence that any phenomena across any regime should be "just fine" as well, I would beg to differ.

In truth, it has happened many times historically that a theory was thought of as more generally valid than it actually was - not because of some obvious failure but because of a special case that had not been (or not been able to be) tested. The fact that there is at times less motivation to test for less "exciting" results is actually unfortunate and I am very thankful for the reproducibility that physics usually has.

Of course, it can (and does) happen that fully duplicate work is done. This is not a tragedy either, and is often intentional - for example ATLAS and CMS both independently found the Higgs at the LHC. If this were re-done 100 times with the same level of rigor, I agree that it would be a bad use of funding, but I would be surprised to find many instances of this in the literature.