21 replies to this topic

[Copasetic]

eight bits, on 22 January 2012 - 12:53 PM, said:

I think that the "rustling in the grass" isn't a good model of how people come to believe in God. A comprehensive theory of belief formation would account for both, but, in my opinion, would have different things to say about them.

One key difference is temporal persistence. Whatever I am going to do about the rustling in the grass, I will do it now. Five minutes from now, I'll either have other problems, or no problems at all anymore. Either way, this matter will be closed.

Belief in God, in contrast, is complicated, develops in the face of continuing exposure to experience and argumentation, and comes in a wide variety of flavors. Belief in Odin is a different thing from belief in Jesus Christ, which ironically is a different thing from believing in the God whom Jesus Christ would be expected to believe in, based on his religious affiliation.

The time it took me to compose just that paragraph is more time than I have in the other problem, yet that paragraph doesn't begin even to describe the question of God, much less say anything about its solution.

I believe that natural selection might favor a rapid, robust and non-deliberative "default" generator to dispose of transient concrete emergenices. To account for God-beliefs in the same way is an extraordinary claim, a kind which, I have been told, requires extraordinary evidence.

Just wanted to comment quickly on this part, I didn't watch the video or read the whole topic--So take it for what it's worth. I don't have time to do it real just with a depthy post on the brain anyway---Happy says you?  

It seems to me Eb, you are making an assumption here that the basis for a belief arising over a short temporal span and a long one must arise from. I know you are quite schooled in learning, at least machine learning, so why then would you expect that "short term" beliefs cannot be coded, codified and reinforced by dopaminergic systems like any other incentive-motive learning?

Reinforcement in the nucleus accumbens (NA) via those dopaminergic systems from the ventral tegmental area certainly could pattern short temporal responses into long-term behavior. In fact, this very thing happens all the time.

Also you are discounting the idea of beliefs acting as surrogates for the natural reward systems of the brain. It has been shown that stimuli, even surrogate stimuli, provide disinhibition of the dorsomedial nuclei of the thalamus from the ventral palladium. This codes and codifies responses resulting in consummatory behavior associated with stimuli--Or at least the stimuli which could elicit those beliefs.

I agree that natural selection would favor a "rapid, robust and non-deliberative "default" generator"--But don't make the mistake of thinking natural selection also didn't also favor other neural circuitry that would inevitably interact with that "rapid, robust and non-deliberative "default" generator" such as learning, decision making etc. Also remember that many neural systems work in opposition to each other when it comes to complex and patterned behaviors. A great example would be decision making and prefrontal cortex-limbic axis, would that I had more time to elaborate though!

I think Eb, and here is the rub of it, you would actually need the extraordinary evidence to support that supposition that we have independent little belief generators which create beliefs like god and are distinct from belief generators such as "when it rains, I catch more fish".

[cluey]

It's natural human tendancies to believe strange things!!!!!!!..........it's the degree of "strange".........that is more the issue

[Habitat]

Copasetic, on 23 January 2012 - 07:01 AM, said:

Reinforcement in the nucleus accumbens (NA) via those dopaminergic systems from the ventral tegmental area certainly could pattern short temporal responses into long-term behavior. In fact, this very thing happens all the time.

I know, I know.  

[libstaK]

Copasetic, on 23 January 2012 - 07:01 AM, said:

Reinforcement in the nucleus accumbens (NA) via those dopaminergic systems from the ventral tegmental area certainly could pattern short temporal responses into long-term behavior. In fact, this very thing happens all the time.

Habitat, on 23 January 2012 - 07:52 AM, said:

I know, I know.  

Are you saying you understood Sheldon's er, Copasetic's dissertation there Habitat?  Do tell.

[Habitat]

libstaK, on 23 January 2012 - 09:03 AM, said:

Are you saying you understood Sheldon's er, Copasetic's dissertation there Habitat?  Do tell.

I hope someone understands it. Even if he's the only one ! To me it was jargon on steroids.  

[eight bits]

Copa

Quote

at least machine learning,

I don't claim to be "schooled" in anything on the 'netz. I think I may have at one time or another mentioned here that I graduated from high school. Whatever I argue must stand or fall on its own merits. Which is just how things are in real-life engineering, or so I'm told.

But fair enough, things like ROC and comprehensive modeling of belief are sometimes discussed under the rubric of "machine learning." However, as is typical of engineering, they rest on a foundation of mathematics, which includes models of learning which are platform independent.

Quote

so why then would you expect that "short term" beliefs cannot be coded, codified and reinforced by dopaminergic systems like any other incentive-motive learning

I don't know that I do expect that. In what you quoted, I was making a distinction between the "short term beliefs" (note that I moved the closing quotation mark on you) and deliberation-durable beliefs (to coin a phrase), like religious acceptance.

To state my belief more affirmatively, what I would firmcode if I were Mother Nature is an action propensity, rather than a "belief engine." Forming an opinion about what's rustling is a waste of precious time. I'd code for "jump in," or "jump out," and cut the deliberative faculties out of the game. They can reboot when the meat substrate is safely distant for further inquiry, or has finished whatever other adventure the propensity has gotten it into.

That's why I moved your closing quotation mark. I'm unsure that that hominid forms a belief about a predator. I think the hominid just does whatever she does, because whatever it is, it's probably better than standing there - like standing there until she can work out an "optimal"  belief about what's rustling. There's no difficulty for rationality in that. Herbert Simon's Nobel Prize was in part for that, explaining how computational burden will be the death of us.

Also, as we both know, if we asked Lucy afterwards about what she did, she might refer to a belief to explain her action. That doesn't mean that she actually formed such a belief at the time. Even her distant descendants will conflate "recognizing a possibility" with "holding a belief."

And yes, I am sure the architecture of organ-meat computers is endlessly fascinating. I own one myself. I did mention, however, that an account of rational belief formation and change would best be platform-independent, IMO.

Quote

Also you are discounting the idea of beliefs acting as surrogates for the natural reward systems of the brain.

Really? Where? If the model which I admire is platform-independent, then why would I "discount" any physical realization of it?

Quote

But don't make the mistake of thinking natural selection also didn't also favor other neural circuitry that would inevitably interact with that "rapid, robust and non-deliberative "default" generator" such as learning, decision making etc

I'm not sure I made that mistake, either. If memory serves, I have a much more expansive view of what sorts of physical arrangements can learn, make decisions, etc. than you do. It would be odd, then, if I disbelieved that reciprocally irritating wetware could rise to the challenge.

Quote

I think Eb, and here is the rub of it, you would actually need the extraordinary evidence to support that supposition that we have independent little belief generators which create beliefs like god and are distinct from belief generators such as "when it rains, I catch more fish".

Perhaps I would, if I were arguing such a thing. The issue in the thread is not independence, but rather identity. I don't believe they are independent. I want the "jump in or jump out NOW" generator to interrupt deliberation. That is a dependency. I also disbelieve they are identical.

It is Shermer who claims that his "whatever NOW" generator produces a mediating belief. He claims further that that mediating belief is in useful ways similar to delibertaion-durable beliefs, and he gives several examples of beliefs that persist for decades. Claims are resolved in the order made. My burden, at most, is to state a reason to doubt.

From an engineering perspective, the burden is light for justifying doubt  that a strategy which is effective over one time scale is also effective on a time scale... well how many orders of magnitude are we talking about? - seconds  vs decades - 8 or 9 orders of magnitude? I'm golden.
-

Edited by eight bits, 23 January 2012 - 10:22 AM.

[Damrod]

In general, I like most of Shermer's works.  He gets a bit more skeptical (cynical?) and ornery as he ages I guess.  Here are a couple of very well written and insightful articles of his from several years ago...I like to post them in the middle of alternative history and paranormal discussions...just to keep people balanced...well....to try anyway...check them out.

http://www.skeptical...reads/id15.html

http://www.skeptical...hreads/id2.html