<br><br><div><span class="gmail_quote">On 15/06/07, <b class="gmail_sendername">Eliezer S. Yudkowsky</b> <<a href="mailto:sentience@pobox.com">sentience@pobox.com</a>> wrote:<br><br></span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
I mention this to show that the question of what it feels like to have<br>a lot of copies of yourself - what kind of subjective outcome to<br>predict when you, yourself, run the experiment - is not at all<br>obvious. And the difficulty of imagining an experiment that would
<br>definitively settle the issue, especially if observed from the<br>outside, or what kind of state of reality could correspond to<br>different subjective experimental results, is such as to suggest that<br>I am just deeply confused about the whole issue.
<br></blockquote></div><br>Related conundrums:<br><br>In a duplication experiment, one copy of you is created intact, while the other copy of you is brain damaged and has only 1% of your memories. Is the probability that you will find yourself the brain-damaged copy closer to 1/2 or 1/100?
<br><br>In the first stage of an experiment a million copies of you are created. In the second stage, after being given an hour to contemplate their situation, one randomly chosen copy out of the million is copied a trillion times, and all of these trillion copies are tortured. At the start of the experiment can you expect that in an hour and a bit you will almost certainly find yourself being tortured or that you will almost certainly find yourself not being tortured? Does it make any difference if instead of an hour the interval between the two stages is a nanosecond?
<br><br><br>-- <br>Stathis Papaioannou