The way I see it, there's no evidence that these problems require additional experimentation to resolve, rather than find an obscure piece of experimentation that has already taken place and whose relevance may not be immediately obvious.

Sure, that more experimentation is needed is probable; but by no means certain.