Newcomb's problem splits people 50/50 in two camps, but the interesting thing is that both sides think the answer is obvious, and both sides think the other side is being silly. When I created a video criticizing Veritasium's video This Paradox Splits Smart People 50/50 I received a ton of feedback particularly from the two-box camp and I simply could not convince anyone of why they were wrong.
That lead me to believe there must be some cognitive trap at play: someone must be not seeing something clearly. After a ton of debates, reading the literature, considering similar problems, discussing with LLMs, and just thinking deeply, I believe the core of the problem is recursive thinking.
Some people are fluent in recursivity, and for them certain kind of problems are obvious, but not everyone thinks the same way.
My essay touches Newcomb's problem, but the real focus is on why some people are predisposed to a certain choice, and I contend free will, determinism, and the sense of self, all affect Newcomb's problem and recursivity fluency predisposes certain views, in particular a proper understanding of embedded agency must predispose a particular (correct) choice.
I do not see how any of this is not obvious, but that's part of the problem, because that's likely due to my prior commitments not being the same as the ones of people who pick two-boxes. But I would like to hear if any two-boxer can point out any flaw in my reasoning.
Jump in the discussion.
No email address required.
Notes -
Building your strategy on causal understanding is better than building it on correlation. Correlation is for when you don't know the causation, when you do have the causation, it's not needed and you should instead reason on causation.
No, the conclusion you stated is incorrect.
But your choice isn't information you can use to affect your choice. Correlation is not the same as causation, so it's only useful for observation.
Your ice cream example, as artificial as it is, works, because you can observe how much ice cream you ate and use that as a proxy for the time of year. But as soon as you're deciding how much ice cream to eat, it's no longer useful because that's not the way the causation runs. You can't make sun screen more effective by eating more icecream. And that's why taking the causation into account allows for superior results.
That's true, but neither stated rigorously (which you'd need for actual mathematical argument) nor very interesting, and it's not the claim that needed defending. Or at least it's not clear how what you said follows from this. Also, you still haven't clarified the stork example.
If causation is available, which in this case it's not. Even then I would argue that all the causation you think you know are nothing more than hypotheses directly reliant on observed correlation.
It is demonstrably the case that one-boxing leads the better outcomes. It can be demonstrated mathematically and empirically.
I can write a computer simulation and compare the outcomes of one-boxers with the outcomes of two-boxers. The outcomes of one-boxers are always superior.
You "argument" against that is ipse dixit. "Nuh uh, it does not". That's not serious.
Another ipse dixit. Yes it is.
Did you even read my essay? It starts with an example that shows the correct choice depends on your choice. I already proved you wrong.
No, it works because all anyone needs is correlation, which is evidence of causation.
Probability theory is stated rigorously. I don't need to prove what has already been proven.
Causation absolutely is available in Newcomb, because the statement of the though experiment describes the causation. In Ice Cream, because we know how ice cream and sun screen work.
No, you can't, because you have no satisfying way to operationalize Omega's prediction without breaking the causal structure of the thought experiment.
Your claims aren't. Hence, they aren't probability theory, and you can't appeal to "probability theory" without doing the work to establish that your questioned claims follow from probability theory. For which you'd need to state them rigorously.
That causation is irrelevant.
That's like saying saying two-boxing leads to better outcomes if we assume two-boxing leads to better outcomes. It's circular reasoning.
You assume we have to ignore the probability of the predictor. There's zero justification for that, other than the fact that you want the result to be two-box.
Yes I can. It is trivial.
No. Each example is standalone.
That example is not a paradox, and the answer depends on the choice.
My claims follow directly from probability theory.
If
p=0.99, then it follows that:(1/n) Σᵢ₌₁ⁿ Xᵢ → 0.99This is pretty much a tautology in probability theory.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link