I am not incredulous because "it seems obvious that 1 death is a relatively superior outcome to 5 deaths." I am incredulous because it seems obvious that 1 deat

I am not incredulous because "it seems obvious that 1 death is a relatively superior outcome to 5 deaths." I am incredulous because it seems obvious that 1 death and no responsibility is a relatively superior outcome to 5 deaths and the responsibility. You ask if I have "considered the possibility that doing nothing means the actor (...) is then not actively responsible for any death." You've gotten this all backwards. The way the trolley experiments are run by default already tests the friction involved with taking active moral responsibility. You are presented with either the option of taking moral responsibility and saving four lives, or not taking moral responsibility and not saving four lives. If you change the default to be a choice between either taking moral responsibility and not saving four lives or not taking moral responsibility and saving four lives, not only is everyone going to choose the second option, you haven't successfully distinguished whether this is because they don't want the moral responsibility or because they wanted to save four lives. Changing the default in this case actually does the opposite of what you want, it emphatically changes the experiment from one that would "test the friction involved with taking active moral responsibility" (friction that acts against the instinct to save four lives) to one that does not! In this new case the "friction involved" works to act against what? People's predisposed instinct to switch the train from the track with one person to the track with five? I don't quite see much friction here, and it turns what was once a task where the 'relatively superior' outcome required a difficult choice into a task where the 'relatively superior' outcome requires no choice at all. You don't actually end up testing anything! (except I suppose, whether the test taker is a psychopath)