Absence of proof of difference is not the same as proof of no difference. Even the surgeon general was 10 years shy of scientific proof that smoking was harmful or that a low fat diet was likely beneficial when he recommended against smoking and high fat diets.
For the sake of argument, can I suggest that those are a little more complex and harder to study? By all means, let's do more controlled listening tests, but how many do we need before we're satisfied with the results? It doesn't seem like it should take nearly as much research to decide what sounds good as opposed to how certain substances and diets affect our long-term health.
