We don't hear these words put together enough. A useful fiction is an idea with limited use. A lot of categorical type ideas apply here. Good/bad words, masculine/feminine, plant/animal... All of these ideas are useful in some situations, but when you try to apply them to all situations you make life difficult for the rest of us.
Aesop's fables, for example. They are good stories to tell. They are useful stories to tell. But I wouldn't want to associate myself with people who think a fox and a raccoon ever had a conversation about the right way to eat a tomato. But letting that slide for a moment can help you put some other stuff in order.
This is where it starts to get a little technical. Watching What the Bleep Do We Know may put things clearer. Well, it will basically have a bunch of people tell you about what I am about to say is bunk. I guess there are a few useful insights in there.
What I want to call bunk is this objective probability that has found a home in quantum mechanics and is starting to spray all over the place. If you understand that probability is subjective, something used to fill in as-of-yet unknown information you can stop reading. Like if you flip a coin into the air and you knew everything about the coin and the table it was going to land on, at this point it wouldn't be a matter of probability, you would know which side it was going to land on. It is only when you don't have all this information that probability becomes useful to predict what is going to happen.
Then in 1927 Heisenberg published his Uncertainty Principle. This stated that for any given particle the more precisely we knew its position the less precisely we could know its velocity. This would be like saying we could never know both the spin of a coin and its distance from the table. If you can't know both parts it becomes necessary to talk about it in terms of probability.
But back to this idea of useful fiction. These very smart scientists came up with very useful ways to think about the probability of the very small things they were studying. The only problem is that they started to believe them. They started to believe, not only do we have to treat these particles as probabilities, but that they actually were these blurs of probabilities.
This lead to a paradox. How come all observations show solid, stable, deterministic forms? They resolved this by saying that when two pieces of matter interact they collapse into one of the probable states. In the most advanced forms of the psychosis blur doesn't collapse until it interacts with a conscious mind. This of course makes us feel pretty special, but now it leads to another paradox of how consciousness developed without definite matter. And I'm sure they have their best people on it.
I wonder if it will occur to them to question the question. That this didn't happen. There is no such thing as probabilistic matter. It is just a useful fiction. It is a good way to talk about matter if you want to predict it's behavior. Other than that it's useless.