Sorry if this one has been discussed before; if it has, I could not find it (although it is related to several previous discussions.
If we get a question like "Does homeopathy work?", we can point to the many studies that have been done on the question.
However, there are many obviously false claims about phenomena which seem to be blatantly pseudo-scientific, but have not been scientifically studied, e.g. the recent "Can manipulating a lucid dream change the future?"
How are such questions best answered? There are several possibilities that I see:
Changing the question (which was done in this case) to a more specific one: is the source quoted in this question evidence for the claim? In this case, the title of the question is still about the general claim, but the question itself (last sentence) and the (accepted) answer are about the particular evidence.
The problem I see with this type of answer is that we do not say anything about the claim itself. In principle, you could immediately come out with the same question using another source, and ask if that provided evidence for the claim in question.
Providing a detailed scientific explanation of why the purported mechanisms by which the process would work are unfeasible. The details given for the sample question might include the physiology of sleep and possible mind-body-interactions.
This might be overkill and not good enough at the same time. I am not sure most readers who consider such phenomena possible would be interested in having that much detail; but they also might point out that the absence of a known mechanism does not disprove the reality of an effect.
Giving an offhand answer. In this case, it might read simply: "There is neither scientific evidence of nor a conceivable mechanism for dreams changing the future."
This is, of course, contrary to the policy of every answer providing evidence.
No answer.
Since there is no direct scientific evidence bearing on the claim, we cannot answer it.
So - what is the way to go?