[Merged] Immortality & Bayesian Statistics

[godless dave]

Dave,
- Above, you say how the perfect reproduction of a specific neurosystem would not share experiences with the original neurosystem and would therefore be a separate self.

What makes it separate is that it's made of different, separate neurons.

- In the case of one person, the existing self does share the experiences of all the preceding selves, and is -- in that sense -- the same self. This self has "changed" enormously over the years, but in that one sense, remains the same self.

- I suspect that the above won't help either, but I think it does get at the meaning of the "self" that I'm talking about. When someone says that he doesn't want to die, he's alluding to the discontinuation of his heretofore continuous self.

Yes. And?

 

[Loss Leader]

- In the case of one person, the existing self does share the experiences of all the preceding selves, and is -- in that sense -- the same self. This self has "changed" enormously over the years, but in that one sense, remains the same self.

Okay. Let's agree then that each separate biological neurosystem is a "self". I don't see how that helps you, though. The "self" is assigned to a biological entity. When the entity dies, the "self" dies, too.

Or, the "self" is a constantly changing aspect of the constantly changing process of life. In that case, there is no "self" that can survive life because there is no constant, continuous "self" to begin with.

In the course of this thread, you've actually argued both things. Neither one actually helps you in the least bit, because both are finite. Now you seem to want to attach your theory to the first concept - each individual biological entity. It's not particularly right, but what's the next step?

I suspect that the above won't help either, but I think it does get at the meaning of the "self" that I'm talking about. When someone says that he doesn't want to die, he's alluding to the discontinuation of his heretofore continuous self.

Okay. So with this incomplete notion of "self," what would you like to do next?

 

[jt512]

You don't think what means what I think it means? Likelihood? Are you equating likelihood to probability?

According to Likelihood function
"The likelihood of a set of parameter values, θ, given outcomes x, is equal to the probability of those observed outcomes given those parameter values, that is L(θ | x) = P(x | θ)."

If we set

x = me (Jabba currently exists, the outcome)
θ = A (Jabba exists for one finite time at most, the parameter value)

We get:

"The likelihood of a set of parameter values, A, given outcomes me, is equal to the probability of those observed outcomes given those parameter values, that is L(A|me) = P(me|A)."

Then P(me|A) is the probability that Jabba currently exists, given that Jabba exists for one finite time at most

And L(A|me) is the likelihood that Jabba exists for one finite time at most, given that Jabba currently exists

But you said:



likelihood instead of probability.

I don't think this is just me being picky. If you are speaking formally, there is a distinction between "likelihood" and "probability".

According to Likelihood function
"But in statistical usage, a distinction is made depending on the roles of the outcome or parameter. Probability is used when describing a function of the outcome given a fixed parameter value. For example, if a coin is flipped 10 times and it is a fair coin, what is the probability of it landing heads-up every time? Likelihood is used when describing a function of a parameter given an outcome. For example, if a coin is flipped 10 times and it has landed heads-up 10 times, what is the likelihood that the coin is fair?"

You've been understandably misled by the notation and language of that article.

The symbol L(θ|x) is just the notation for the likelihood function. It is confusing to think of it as "the likelihood of θ." In fact it is defined as P(x|θ), given that x is observed. Let me write the definition this way: given that a set of data x has been observed, the function of θ, denoted L(θ|x), is called the likelihood function, and is defined by

L(θ|x) ≡ P(x|θ).​

When we use the likelihood function to do a hypothesis test of H1 vs H0, we fix θ at its value under each hypothesis and compute the corresponding conditional probability of the data under that hypothesis. In the present case, θ under the two hypotheses aren't real numbers, but rather the statements "H1 is true" and "H0 is true," and so we write the likelihood functions as P(x|H1) and P(x|H0). Hopefully it is clearer now the likelihood terms in Jabba's Bayes computations are P(me|SM) and P(me|NSM), or however he's now writing them. The main point is that, since they are likelihoods, then by definition, the "data" term comes before the bar.

 

[Humots]

You've been understandably misled by the notation and language of that article.

The symbol L(θ|x) is just the notation for the likelihood function. It is confusing to think of it as "the likelihood of θ." In fact it is defined as P(x|θ), given that x is observed. Let me write the definition this way: given that a set of data x has been observed, the function of θ, denoted L(θ|x), is called the likelihood function, and is defined by

L(θ|x) ≡ P(x|θ).​

When we use the likelihood function to do a hypothesis test of H1 vs H0, we fix θ at its value under each hypothesis and compute the corresponding conditional probability of the data under that hypothesis. In the present case, θ under the two hypotheses aren't real numbers, but rather the statements "H1 is true" and "H0 is true," and so we write the likelihood functions as P(x|H1) and P(x|H0). Hopefully it is clearer now the likelihood terms in Jabba's Bayes computations are P(me|SM) and P(me|NSM), or however he's now writing them. The main point is that, since they are likelihoods, then by definition, the "data" term comes before the bar.

Thank you for the clarification.

If I understand correctly, you are referring to a hypothesis test of H0 and H1 given data x:

P(H0|x) = P(x|H0) P(H0) / P(x)

P(H1|x) = P(x|H1) P(H1) / P(x)

 

[jt512]

Thank you for the clarification.

If I understand correctly, you are referring to a hypothesis test of H0 and H1 given data x:

P(H0|x) = P(x|H0) P(H0) / P(x)

P(H1|x) = P(x|H1) P(H1) / P(x)

Yes, we could calculate P(H0|x) and P(H1|x) to determine which hypothesis is more probable (and by how much) after data x are observed. But there is a shortcut. Watch what happens if we divide your first equation by your second. P(x) cancels out and we get

P(H0|x)/P(H1|x) = P(x|H0)/P(x|H1) × P(H0)/P(H1) , or

(posterior odds of H0 vs H1) = (Bayes factor) × (prior odds of H0 vs H1) ,

where the Bayes factor is the relative likelihood of the data under the two hypotheses. The Bayes factor tells us how much the data should change our opinion of the odds of H0 (vs H1). The Bayes factor has the important property of being independent of the prior odds. In Bayesian inference, two people may wildly disagree on the odds of a hypothesis a priori, but they should both be in agreement on how much their odds are shifted by newly observed data. The Bayes factor is, in this sense, objective, even though the prior probabilities are subjective.

 

[Humots]

Yes, we could calculate P(H0|x) and P(H1|x) to determine which hypothesis is more probable (and by how much) after data x are observed. But there is a shortcut. Watch what happens if we divide your first equation by your second. P(x) cancels out and we get

P(H0|x)/P(H1|x) = P(x|H0)/P(x|H1) × P(H0)/P(H1) , or

(posterior odds of H0 vs H1) = (Bayes factor) × (prior odds of H0 vs H1) ,

where the Bayes factor is the relative likelihood of the data under the two hypotheses. The Bayes factor tells us how much the data should change our opinion of the odds of H0 (vs H1). The Bayes factor has the important property of being independent of the prior odds. In Bayesian inference, two people may wildly disagree on the odds of a hypothesis a priori, but they should both be in agreement on how much their odds are shifted by newly observed data. The Bayes factor is, in this sense, objective, even though the prior probabilities are subjective.

Nice! Thanks!

 

[Giordano]

Dave,
- Above, you say how the perfect reproduction of a specific neurosystem would not share experiences with the original neurosystem and would therefore be a separate self.
- In the case of one person, the existing self does share the experiences of all the preceding selves, and is -- in that sense -- the same self. This self has "changed" enormously over the years, but in that one sense, remains the same self.

- I suspect that the above won't help either, but I think it does get at the meaning of the "self" that I'm talking about. When someone says that he doesn't want to die, he's alluding to the discontinuation of his heretofore continuous self.

Aren't you assuming what you have promised to prove, that we share the experience of "prior selves?" What evidence do you have?

Most people don't want to die so that they will still be able to have future experiences. I don't care if the 5 year old that was me dies; he is long gone anyway. I value my memories of him, but he is gone.

 

[Humots]

Dave,
- In the case of one person, the existing self does share the experiences of all the preceding selves, and is -- in that sense -- the same self. This self has "changed" enormously over the years, but in that one sense, remains the same self.

- I suspect that the above won't help either, but I think it does get at the meaning of the "self" that I'm talking about. When someone says that he doesn't want to die, he's alluding to the discontinuation of his heretofore continuous self.

Aren't you assuming what you have promised to prove, that we share the experience of "prior selves?" What evidence do you have?

Most people don't want to die so that they will still be able to have future experiences. I don't care if the 5 year old that was me dies; he is long gone anyway. I value my memories of him, but he is gone.

I suspect that by preceding selves, Jabba is saying that our memories of ourselves at previous ages (such as the 5 year old that was you) are previous “selves”. This is true in a sense, because we remember distinct previous periods when we thought and felt differently. And of course we share the experiences of these "prior selves".

Also, there is a kind of overall sense of “self” in the continuity of development from our earlier to our present memories.

The problem for Jabba is that in no way does this support his claim of “a part of the self (the "experiencer" itself) that does not change from its first memory to its last” or that exists independently of the brain.

Jabba is throwing half-baked and trivial ideas against the wall and hoping something will stick.

 

[Akhenaten]

I suspect that by preceding selves, Jabba is saying that our memories of ourselves at previous ages (such as the 5 year old that was you) are previous “selves”. This is true in a sense, because we remember distinct previous periods when we thought and felt differently. And of course we share the experiences of these "prior selves".

If that's the case then Jabba's immortal selves don't even last as long as the bodies that house them, let alone continuing forever without them.

Also, there is a kind of overall sense of “self” in the continuity of development from our earlier to our present memories.

I'd be surprised if any of us here would disagree with that.

The problem for Jabba is that in no way does this support his claim of “a part of the self (the "experiencer" itself) that does not change from its first memory to its last” or that exists independently of the brain.

Exactly. Or at least, that's one of his problems. Each new attempt at clarification seems to bring him a few more. I guess that's why starting with a conclusion and working backwards is considered to be such a bad idea.

Jabba is throwing half-baked and trivial ideas against the wall and hoping something will stick.

It didn't work in the holy tablecloth thread either.

 

[Loss Leader]

Jabba is throwing half-baked and trivial ideas against the wall and hoping something will stick.

He now appears to be simultaneously arguing for two different definitions of "self." I want to throw him some kind of lifeline, but he's so tied up in knots, it'd probably sink him.

 

[Jabba]

- So -- I’m saying that the “self,” or at least some aspect of the self, remains the same throughout one’s life time.
- And now, at least some of us seem to be on the same page in regard to that aspect of “self” in this discussion…
- Next, I say that reproducing the same neuro-system would probably not reproduce the same self.
- Who agrees; and, who disagrees?

 

[PixyMisa]

- So -- I’m saying that the “self,” or at least some aspect of the self, remains the same throughout one’s life time.

Yes, you're saying that. It's not true.

 

[John Jones]

- So -- I’m saying that the “self,” or at least some aspect of the self, remains the same throughout one’s life time. - And now, at least some of us seem to be on the same page in regard to that aspect of “self” in this discussion…
- Next, I say that reproducing the same neuro-system would probably not reproduce the same self.
- Who agrees; and, who disagrees?

Which aspect is that?

 

[Filippo Lippi]

Which aspect is that?

The magical bit that lives in the land beyond physics

 

[Slowvehicle]

- So -- I’m saying that the “self,” or at least some aspect of the self, remains the same throughout one’s life time.
- And now, at least some of us seem to be on the same page in regard to that aspect of “self” in this discussion…
- Next, I say that reproducing the same neuro-system would probably not reproduce the same self.
- Who agrees; and, who disagrees?

Good Morning, Mr. Savage!

What specific aspect of the emergent property of a specific neurosystem are you, personally, claiming "reamins the same throughout one's lifetime"?
-Have you forgotten Lincoln's Axe?

What specific "page" do you think at lest some of us are on? Please provide specific examples...

Given the observed fact that consciousness appears to be an emergent property of a neurosystem, I think it is safe to say that since every neurosystem is unique, the emergent property of consciousness will be unique to each neurosystem. And water is wet.

Are you planning to get back to "immortality"?

Would it really be harder to start with your practical, empirical, objective evidence for the existence of the "soul"?

 

[Mojo]

- So -- I’m saying that the “self,” or at least some aspect of the self, remains the same throughout one’s life time.
- And now, at least some of us seem to be on the same page in regard to that aspect of “self” in this discussion…

What "aspect" would that be?

 

[godless dave]

We are not on the same page, Jabba. Continuous and unchanged are not synonyms. Even if we accept that continuity of self is not entirely an illusion, that does not mean that the self remains unchanged.

There are neurons in my brain that were there when I was born. They haven't stayed unchanged, though.

 

[Jabba]

- So -- I’m saying that the “self,” or at least some aspect of the self, remains the same throughout one’s life time.
- And now, at least some of us seem to be on the same page in regard to that aspect of “self” in this discussion…
- Next, I say that reproducing the same neuro-system would probably not reproduce the same self.
- Who agrees; and, who disagrees?

We are not on the same page, Jabba. Continuous and unchanged are not synonyms. Even if we accept that continuity of self is not entirely an illusion, that does not mean that the self remains unchanged.

Dave,
- I certainly accept that the overall self, including all its numerous characteristics, changes extensively over it's lifetime. I just claim that one aspect of the self --really, its core aspect -- remains the same during its lifetime. Do you agree with that?

 

[Squeegee Beckenheim]

Dave,
- I certainly accept that the overall self, including all its numerous characteristics, changes extensively over it's lifetime. I just claim that one aspect of the self --really, its core aspect -- remains the same during its lifetime. Do you agree with that?

You need to define what you mean by "core aspect".

 

[godless dave]

Dave,
- I certainly accept that the overall self, including all its numerous characteristics, changes extensively over it's lifetime. I just claim that one aspect of the self --really, its core aspect -- remains the same during its lifetime. Do you agree with that?

No.