Stuart_Armstrong comments on Sleeping Beauty gets counterfactually mugged - Less Wrong

1 Post author: Stuart_Armstrong 26 March 2009 11:44AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (23)

You are viewing a single comment's thread. Show more comments above.

Comment author: Stuart_Armstrong 26 March 2009 04:53:59PM 1 point [-]

If he did that, it would no longer be rational to accept to pay (expected return 0.5x(£260) + 0.5x(-£150 -£150) = -£20). The fact that this case is NOT identical to the first one is due to the whole Sleeping Beauty set-up.

Comment author: AllanCrossman 26 March 2009 06:44:34PM *  1 point [-]

No: I meant, if you pay, you pay a total of £250.

i.e. a slightly clearer statement of my imagined setup is this:

Omega flips a coin. On tails, he asks you if you'll pay £125 now, knowing that if this is day 1 he'll wipe your memory and ask you again tomorrow.

On heads, he simulates you and sends you £260 if you pay.

There is never any money paid to non-payers.

(Basically, the only difference between this version and yours is that both paying and not paying have a return that is £25 lower than in your version. That surely shouldn't make a difference, but it makes the problem go away.)

  • Not paying always gives £0.
  • Precommitting gives a return of 0.5 x £260 + 0.5 x (-£250) = £5
  • By your logic, at the time of the decision, return is 1/3(£260-£125-£125) = £3.33

Isn't that correct? I admit I suck at maths.