Simulations represent possible realities, in preparation for possible eventualities. Our brains run simulations, the better to survive and reproduce. Simulations don't have to be rational; they just need to be possible. Or possibly possible. They operate in shades of darkness, from unknown unknowns to pretty sure.
When we imagine something, we are running a simulation. Simulations are not propositions about reality; they are play-throughs of possible realities. They test predictions (or guesses) and guide our actions. What happens after that is fed pack into the machine to inform further predictions and simulations.
Emotions bias the whole process. Emotional tendencies (e.g., excitement-seeking, approach, avoidance) create patterns. Prone to anxiety? The machinery will tilt towards the threat potential in its representations, systematically reviewing the possibilities.
Confident and optimistic? The machinery will tilt towards images of success and triumph, but not dwell on them because no preparation is required for what may come. We already know we can handle whatever is thrown our way, and it will be good.
Some simulations are more credible than others. This is the continuum of believability. Beliefs aren't either/or propositions; they have degrees of credibility.
This is the way I think. It feels closer to what is the case, although I realize it can obscure a deeper truth. Sometimes a "catastrophic neural stoppage" is simply cowardice.
Inspired by the short story, "The Poltroon Husband", by Joseph O'Neill in The New Yorker, March 12, 2018 Issue.
Note: This is a slightly revised post from 2018.