Given that we have a simulator of this worm right there (which includes it moving), can it really be up to debate whether it uses action potentials or not?
I'd think the simulation has to get it right, and so needs to simulate action potentials if the worm has them, or not simulate them (but whatever the worm has instead) if not, right? Or could the simulation still be incorrect and only based on current assumptions, but getting this wrong still allows some worm-like behavior?
I really wish the readme/FAQ would talk a bit more about the worm and the simulation, rather than have 80% of their content be about Docker, though, so that I could learn more what cells it actually simulates.
Not necessarily, because you could also simulate the worm without neurons at all. It's the closeness of the simulation to the real thing that demands that it is done right and the question effectively is: is this simulation close enough that if such a detail would be wrong that it would fail?
One way to answer that would be to add and remove such mechanisms to see if it would lead to different behavior.
That isn't immediately true, with enough fitting parameters you can capture the effect underlying behaviour without explicitly capturing it, or even without knowing it exists.
"With four parameters I can fit an elephant, and with five I can make him wiggle his trunk." - John von Neumann [0]
I'd think the simulation has to get it right, and so needs to simulate action potentials if the worm has them, or not simulate them (but whatever the worm has instead) if not, right? Or could the simulation still be incorrect and only based on current assumptions, but getting this wrong still allows some worm-like behavior?
I really wish the readme/FAQ would talk a bit more about the worm and the simulation, rather than have 80% of their content be about Docker, though, so that I could learn more what cells it actually simulates.