> can you give me an example of an established physics theory that was discarded by experimental evidence? Not entirely sure what you mean by "established" her

> can you give me an example of an established physics theory that was discarded by experimental evidence? Not entirely sure what you mean by "established" here. Depending on quite what you mean, N-rays or Aristotelian gravity might meet these criteria. > disproven theory will be rebranded as “classic” and will continue to exist and taught as if it were never disproved. I'm not sure why you object to teaching previous models that are simpler and easier to understand, while having inaccuracies in some places. The students know that they aren't perfect models, so it isn't deceptive, and it is useful pedagogically as a stepping stone for understanding more accurate models. There is a quote that seems relevant here: "All models are wrong. Some models are useful." Everyone* understands that our current understanding is imperfect (i.e. somewhat wrong). The goal is to become less (and less) wrong. > and physicists look at the outcomes and choose the experiments that makes the best prediction. Or the experiment that matches the model best. That isn't quite what I meant, but maybe you accidentally put a different word than the word you meant in a place or two? What I meant was that (imagine, hypothetically) an expirament is scheduled, and people(not necessarily physicists. Anyone who wants to. Though, people who understand things would probably do better) [make bets / buy and sell shares in different predictions] on statements about what will be observed when the experiment is done. (The market price of the shares in the different predictions can be interpreted as market assigned probabilities of different outcomes). Then, at the scheduled time, the experiment is done, and observations are made, and the people who made accurate predictions about what would be observed profit, and the people who made bets that didn't pan out don't. The models which allow people to more often make predictions that turn out to be correct, and therefore allow people to profit by winning bets, are the models that are more useful than the other models being used. Note that this doesn't mean the models are absolutely correct, but, they are less wrong than the models that the people who would tend to lose money in this would be using. If anyone had a model that was absolutely correct, and could be used effectively (like, it doesn't take them too much computation time to use), they could use it to (in the long run) profit relative to people using worse models. So, in such a system, I would think that the best models which have been found would win out, regardless of any sort of cultural influence or whatever you are concerned about. So my question was, then, if such system were to be put in place now, what model do you imagine the winners would be using? > The experiment is set up to give one result and one result only and all the successive experimenters do is to try to lower the experimental error. This has not been my observation. There are many physicists looking closely to see if they can find any place where expiramental results differ from the predictions that current models make, and when even small potential discrepancies pop up, they spring on them in the hope that it is a real difference from what current models predict. For an experiment to not meet the description you describe, what would it have to be like? Would it have to have a number of obviously qualitatively different plausible outcomes? If current models predict one qualitative outcome with very high probability, would you recommend that it be done? I suspect that designing an experiment such that the outcome is expected to be qualitatively consistent, but currently quite uncertain under current models, would be rather difficult. (The reason I say qualitatively is because you seem to be objecting to increasingly precise measurements of the same quantities, so I imagine that you are wanting somewhat qualitatively observations) (Please note that I am not a physicist, nor an expert in epistemology or in anything else.)