It's not a binary thing, it's a matter of probability, one can be more or less likely to overreact and training helps with that. You most likely can't fix the p

It's not a binary thing, it's a matter of probability, one can be more or less likely to overreact and training helps with that. You most likely can't fix the problem entirely, mistakes happen, but you can at least reduce the number of incidents.