[Afraid] Untrue facts as "fallout option"?

Triggered by this thread, that I'm posing here rather than resurrecting (Forge rule about not resurrecting stuff a couple of pages down, right?).

http://www.indie-rpgs.com/forum/index.php?topic=22931.0

.... Part way through the conflict one of the players challenges with injecting the girl with sedatives so she can't hear the story.

Until he suggested that I'd not thought about whether access would work like that. It didn't feature in the image I had of the Monster in my head, but clearly there's a case that it might work, and it was thematic and we were all bought into the action. The player put the dice forward and the conflict continued. What makes sedating the girl work as a way of reducing access? It didn't work because I as GM decided it could work, it didn't work because the people round the table thought it was a good idea, it worked because the player had dice to invest in the action and thought it should happen in the story.

In that kind of scene, where the player announces a stake of accomplishing X by using this tactic Y and it's not an established fact yet that the tactic works (dousing victim with milk while chanting Egyptian prayers stops the pyschic access leading to the nightmares of this stage):

- win stakes, tactic works (players define outcomes and also factual elements of setting/story).

- lose stakes, it didn't work this time or it doesn't work at all (maybe it's river water, not milk)?

I would say that the rules about "differences for retries" should mean that adjusting tactics may be dicated anyway sometimes ~ is it all the time because of that?

If not all the time, could one possible fallout option (short term perhaps) be "tactic is false - doesn't work", similar to a "circumstance becomes true"?