I'm not sure about some of the inferences you derive in this post. In your rollercoaster example, you made a very reasonable assumption ("sections are empty because they’re not in use"), but it turned out to be incorrect. However, it seems to me it was very likely to be correct, and investigating whether it was false would in most cases be a waste of effort (contra "For half an hour, we hadn’t actually been trying.").
"But we didn't see anyone join us, even though they were within earshot."
Are you sure they were paying attention? Note, you didn't think to shout the information to people to make sure they were aware of it.
"What the hell was that?"
An assumption that would usually be true but just happened to be false here, hence the story? If it were true, there'd be no inspiring story.
The flaw here is that it costs too much time and energy to investigate *EVERY POSSIBLE* assumption, where the vast majority will be correct.
I've often thought the dog experiment is just a horrible moralizing distortion. Being defensive when in pain and escape hasn't worked before, is the *correct* *action* for an animal.
This is getting long, so let me just suggest taking into account more of the problems of "expected value" and being cautious of "survivorship bias".
I think that our assumptions were less reasonable than you think. When sections aren't in use, they're also usually cordoned off, but at the beginning of the line, both sections were completely open. There were also definitely employees at the entrance, so we totally could have asked someone. We just chose not to.
Earshot is maybe the wrong choice of words here. There were people who actively saw us slip under the fence (which I know because I saw them), and those people were also within listening range of the conversation (although you're right in that they may not have been paying attention). Those people chose not to join us OR ask what was going on.
Regarding the dog experiment, my thought is that escape should be attempted even if the pain *seems* unavoidable, because attempting escape and not attempting escape have the exact same negative outcome: being in pain. BUT, attempting escaping is the only action that has a potential positive outcome: potentially escaping. Even if the EVs of escaping are very low, they can't be any lower than the EVs of not escaping, unless you're talking about trying to conserve or save energy (which I addressed in the article). Additionally, we're not dogs: we usually have a wide array of actions available to us, and in my opinion it's very rare that the best option is maintaining the status quo. Learned helplessness might be the correct behavior for a dog in that particular experiment, but it doesn't make sense for most humans living their lives today.
Survivorship bias is one of my favorites! I'm not entirely sure how it's relevant here, though. Are you suggesting that I'm here telling the story because I had a single positive outcome, whereas everyone who had a negative outcome wouldn't tell their story? I think that makes sense. But in the future, if I see an empty line that isn't cordoned off, I am at the very least going to go ask someone about it. I suppose only time will tell what the EVs of that play are.
I'm not sure about some of the inferences you derive in this post. In your rollercoaster example, you made a very reasonable assumption ("sections are empty because they’re not in use"), but it turned out to be incorrect. However, it seems to me it was very likely to be correct, and investigating whether it was false would in most cases be a waste of effort (contra "For half an hour, we hadn’t actually been trying.").
"But we didn't see anyone join us, even though they were within earshot."
Are you sure they were paying attention? Note, you didn't think to shout the information to people to make sure they were aware of it.
"What the hell was that?"
An assumption that would usually be true but just happened to be false here, hence the story? If it were true, there'd be no inspiring story.
The flaw here is that it costs too much time and energy to investigate *EVERY POSSIBLE* assumption, where the vast majority will be correct.
I've often thought the dog experiment is just a horrible moralizing distortion. Being defensive when in pain and escape hasn't worked before, is the *correct* *action* for an animal.
This is getting long, so let me just suggest taking into account more of the problems of "expected value" and being cautious of "survivorship bias".
https://www.explainxkcd.com/wiki/index.php/1827:_Survivorship_Bias
I think that our assumptions were less reasonable than you think. When sections aren't in use, they're also usually cordoned off, but at the beginning of the line, both sections were completely open. There were also definitely employees at the entrance, so we totally could have asked someone. We just chose not to.
Earshot is maybe the wrong choice of words here. There were people who actively saw us slip under the fence (which I know because I saw them), and those people were also within listening range of the conversation (although you're right in that they may not have been paying attention). Those people chose not to join us OR ask what was going on.
Regarding the dog experiment, my thought is that escape should be attempted even if the pain *seems* unavoidable, because attempting escape and not attempting escape have the exact same negative outcome: being in pain. BUT, attempting escaping is the only action that has a potential positive outcome: potentially escaping. Even if the EVs of escaping are very low, they can't be any lower than the EVs of not escaping, unless you're talking about trying to conserve or save energy (which I addressed in the article). Additionally, we're not dogs: we usually have a wide array of actions available to us, and in my opinion it's very rare that the best option is maintaining the status quo. Learned helplessness might be the correct behavior for a dog in that particular experiment, but it doesn't make sense for most humans living their lives today.
Survivorship bias is one of my favorites! I'm not entirely sure how it's relevant here, though. Are you suggesting that I'm here telling the story because I had a single positive outcome, whereas everyone who had a negative outcome wouldn't tell their story? I think that makes sense. But in the future, if I see an empty line that isn't cordoned off, I am at the very least going to go ask someone about it. I suppose only time will tell what the EVs of that play are.