But according to Cialidini's "Influence", there is a ten times higher chance of dying in a plane crash if you fly in the months immediately after a highly publicized suicide, so securing the plane against a suicidal pilot (as opposed to anyone wanting to carry out a terrorist attack) might very well be more worthwhile than securing it against terrorists.

(Cialdini cites research, which I'm too lazy to look up right now, that claims something as high as 40-50 excess deaths from car and plane crashes in the US per highly publicized suicide)

My claim wasn't due to the danger of a suicidal pilot, but rather the futility of securing a plane against a suicidal pilot. It seems like the problem is roughly equivalent to creating a plane that can operate entirely without a pilot.

I know. But it need not be that hard - being suicidal does not automatically mean someone is prepared to e.g. fight their co-pilot to kill themselves. In fact, most people who are suicidal are not strongly intent on going through with it - it often takes just a little nudge one way or the other to change the outcome.

A co-pilot at the helm, or an unlocked door through which someone may come in and question the pilots actions could very well make all the difference.

Monkey see, monkey do.

Discussing any behavior, especially in the media, makes it more likely.

https://en.wikipedia.org/wiki/Copycat_suicide (Werther effect)

The interesting bit (from the book IIRC) is that the copiers will often wait some time so it seems like they are not copying. This is why the risk to subsequent air travel increases after some delay from the original incident. People are partially unpredictable, so it's possible to have an immediate copy as well.