[Merged] Immortality & Bayesian Statistics

[Toontown]

Once you specify "you", you are drawing the target.

Was there some other possible "me" brain in that large-but-not-infinite set of possible unique brains Pixel42 mentioned?

The U-brain assumption says no. This is the one. No. There was no other. No. There will be no other.

But that's not a target specified by the u-brain assumption, because...well, dave's rules.

But "you" doesn't exist until after the fact.

So the target you deny isn't hit until the target you deny is hit. But after it's hit, you have no choice but to deny it was ever a target because it's already been hit. Or you're a Texas sharpshooter.

Lots of ways to fall into that TS trap of yours. What do you think when you wake up, presumably in the morning? "Hey, somebody woke up."?

?
But noticing that the assumed one and only brain exists is drawing a target around it?

Yes.

That is one wierd "yes", dave. How do you avoid noticing that you exist?

Because if you do notice you exist, you've committed the Texas sharpshooter fallacy. So every time you say "I" or "me", you commit the TSF.

 

[godless dave]

Only one possible target is you. All the other possible targets would be different people.

That does not mean that it is in any way amazing when one target is hit and others are missed. That's what the hypothesis predicts.

 

[Pixel42]

So what was all that about "this (very large but not infinite) number of potential targets"? Could only one of those potential targets have been 'you', or not?

As I said in the post to which you were replying:

The fact that only one of those potential targets is me did not draw a target around it in advance, no. I draw that target only because I exist, i.e. after that particular target has been hit.

It seems to me that you are making a fundamental logical error, and if you can't see it yourself by now then you probably never will. You, of course, will say the same thing about me. Obviously I can't completely rule out the possibility that I'm the one who's missing something, but I can't help feeling that if your argument was valid I would not be coming across it for the first time in a thread on the JREF forum, let alone one started by Jabba of all people.

 

[Toontown]

Only one possible target is you. All the other possible targets would be different people.

That does not mean that it is in any way amazing when one target is hit and others are missed. That's what the hypothesis predicts.

Nor does it mean anyone, least of all me, is amazed.

Nor does it mean a random event ceases to have an expected likelihood after it occurs.

Nor does it mean the expected likelihood of an observed event is not a property of the means by which the expected likelihood was calculated or estimated, thus is independent of the occurrence of the event.

Nor does it mean it is invalid to ask, "Given that this event occurred, how likely is it that the means by which it's expected likelihood was estimated is correct?"

It couldn't possibly mean any of the above, because otherwise it would be impossible to statistically test a hypothesis. Which I have done, many times, and it works.

Nor does it mean I haven't pointed all this out to you several times, including yesterday.

So What DOES your observation actually mean?

 

[godless dave]

Nor does it mean anyone, least of all me, is amazed.

Nor does it mean a random event ceases to have an expected likelihood after it occurs.

Nor does it mean the expected likelihood of an observed event is not a property of the means by which the expected likelihood was calculated or estimated, thus is independent of the occurrence of the event.

Nor does it mean it is invalid to ask, "Given that this event occurred, how likely is it that the means by which it's expected likelihood was estimated is correct?"

It couldn't possibly mean any of the above, because otherwise it would be impossible to statistically test a hypothesis. Which I have done, many times, and it works.

Nor does it mean I haven't pointed all this out to you several times, including yesterday.

So What DOES your observation actually mean?

It means that there's no reason to reject the "unique brain hypothesis".

 

[Spindrift]

I've just read this entire thread and it has been very interesting. I don't know anything about Bayesian or unique brains but I have come to the following conclusion:
The only thing immortal is this thread.

Back to your regularly scheduled programming.

 

[Toontown]

It means that there's no reason to reject the "unique brain hypothesis".

It can't mean that, because that's an invalid statement. You have no idea. You haven't tested it.

Saying some things happen and some things don't happen is not a test of anything. We always expect that some things will happen and some things will not happen. The only hypothesis that wouldn't predict that is the one that says the universe doesn't exist.

 

[Toontown]

The expected frequency is very small. And what we observe is that, out of all the people with brains who could possibly have been born, only a very small number are. And prior to that, out of all the possible ways the universe could have turned out, it turned out only one way.

"Very small" is what we expect and it's what we observe.

So I'm still not seeing any reason why we should reject hypothesis H₀.

Based on your statement here, I must conclude that you still did not know, as of yesterday, what hypothesis H₀ asserts, and I'm tired of repeating it.

Hint: It has nothing at all to do with the proportion of existing brains to potential brains.

 

[godless dave]

Based on your statement here, I must conclude that you still did not know, as of yesterday, what hypothesis H₀ asserts, and I'm tired of repeating it.

Hint: It has nothing at all to do with the proportion of existing brains to potential brains.

It's not stated in your hypothesis, but the proportion of existing brains to potential brains is intimately connected with that hypothesis.

You seem to be making the mistake that one particular unique brain is the only possible outcome. It isn't. It's one of a large number of possible outcomes.

 

[godless dave]

It can't mean that, because that's an invalid statement. You have no idea. You haven't tested it.

We haven't tested using Bayesian statistics. We can't because we don't have nearly enough data nor even the means to collect that data. The way you want it tested, we would have to have complete data of the initial state of the universe and enough knowledge of astrophysics to model all possible outcomes. The latter is theoretically attainable but not for quite some time; the former is inaccessible.

Fortunately we have other means of testing the hypothesis that human brains are unique and are the source of human consciousness.

 

[AdMan]

You seem to be making the mistake that one particular unique brain is the only possible outcome. It isn't. It's one of a large number of possible outcomes.

He's had this explained to him numerous times in many ways, yet it appears his unique brain is unable to comprehend it.

 

[Squeegee Beckenheim]

There is only one lottery ticket that brings the money.

If the unique brain hypothesis is true, then there is only one brain that brings "you".

There's only one brain that brings me. Had a different sperm fertilised my mother's egg there would be someone else. Someone not me, but still someone.

Similarly, if all but one of my lottery numbers were called, then someone would still most likely win. Just not me.

You're acting as if the specific me who is typing this was an inevitable result of my parents having sex. That's not true. Someone being born was likely, and I'm just lucky it was me.

That's why you winning the lottery is lucky for you, yet almost every week someone wins it. You winning is unlikely, someone winning is almost inevitable.

It doesn't matter when either of those realizations come to you.

It does. Because it's easy for me to look back and say "wow, isn't it amazing that out of all the people who could have been born, I was born?", just as it's easy to look at the lottery numbers and say "wow, isn't it amazing that out of all the numbers that could have been draw, it was those specific 6?". In order for it to be significant, however, the fact that it would be the person sitting here with this exact DNA, body and personality would have to be predicted, just as for it to be significant that you can tell what the lottery numbers are, you have to say what they are in advance.

 

[Jabba]

Dave,
- Jut to make sure -- I assume that you do not consider a likelihood of one over infinity close enough to impossible to pretty much waive the difference. Is that correct?

You haven't shown that the likelihood of someone existing is one over infinity.

Dave,
- OK.
- But, if I were to do that, would it be sufficient to make you suspect that the scientific model is wrong? If not, would it make you wonder?

It might. But first I would check your math.

Dave,
- Good. I can't remember what I'm putting off answering otherwise, but if I can't find anything more pressing, I'll try to give it (the "math") my best shot next...

Dave,

- Finally -- the math behind my suggestion that the "likelihood" of my current existence is not just unimaginably small, but actually one over infinity.
- I've been away and busy.
- Here's how I see it, one baby step at a time. Please humor me -- I see this stuff as being especially difficult to communicate. Also, I want to show that I'm back -- and, trying to present the entire explanation at once would take much too much time.

1. The "me" in this issue is my individual consciousness or awareness. This me is simply my "observer." It has nothing to do with any of my "characteristics." To lessen the verbiage I'll just call it my "self."
2. So, the first "number" to determine is that of POTENTIAL selves. That is, how many potential selves are there?

- Any questions so far?

 

[godless dave]

I'm following you so far.

 

[Pixel42]

I'm following you so far.

Not sure I am. What difference does it make how many there are if (a) they all have exactly the same chance of existing and (b) the existence of some of them is inevitable? Even if there were an infinite number of possible selves there would still be nothing remarkable about any particular one existing.

 

[Akhenaten]

- Any questions so far?

Why would there be, beyond the obvious?

You haven't said anything more trivial than that you are the very first iteration of you . . . ever.

Well knock me down with a feather!


Anything on the immortality front?

 

[Jabba]

Jabba's "math" so far:
1. The "me" in this issue is my individual consciousness or awareness. This me is simply my "observer." It has nothing to do with any of my "characteristics." To lessen the verbiage I'll just call it my "self."
2. So, the first "number" to determine is that of POTENTIAL selves. That is, how many potential selves are there?

I'm following you so far.

Dave,

- Good.
- To continue:

3. Science would seem to require some sort of PHYSICAL DIFFERENCE between you and me -- from our very beginnings -- to cause my awareness and your awareness to be different/separate "beings." I don't see through both sets of eyes.
4. I can think of 2 different ways that physical difference could cause us to be consciously separate beings:
4.1. Each specific sperm cell and ovum combination creates a totally specific "self." Any other combination would produce a different specific self. -- however, if we could somehow create an identical sperm and identical ovum, and brought them together, the resulting observer would be the same observer as before (whether 'he' remembered his past life or not).
4.2. Whenever a certain physical situation occurs (life?), a TOTALY NEW SELF is created "out of thin air" as an emergent property. There was no pre-existing model to turn on.

- Any questions, revisions or reservations?

 

[godless dave]

Yes:

however, if we could somehow create an identical sperm and identical ovum, and brought them together, the resulting observer would be the same observer as before (whether 'he' remembered his past life or not).

Identical twins come from the same sperm and ovum, but they don't share a self. The reason they don't share a self is because there are two of them, not one. So even though their brains start out nearly identically, each has its own point of view.

Also, brains develop in response to stimuli, so two brains that started out as identical will develop differently because they will be exposed to different stimuli.

 

[Pixel42]

The existence of identical twins makes 4.1 obvious nonsense.

ETA: ninja'd.

 

[Jabba]

Not sure I am. What difference does it make how many there are if (a) they all have exactly the same chance of existing and (b) the existence of some of them is inevitable? Even if there were an infinite number of possible selves there would still be nothing remarkable about any particular one existing.

Pixel,
- That IS a question that might still need answering -- but Dave had said (in my understanding) that if I could show that the "likelihood" of my current existence under the scientific model is actually one over infinity, he might be willing to waive that question. I'm currently trying to show that such a probability really is one over infinity.
- Eventually, I probably will need to show either what sets me apart from other hypothetical selves, or with why one over infinity eliminates that need.