Relationship Between Uncertainty and Emergence (neuroscience)?

[PhysiPhile]

A huge common misconception about brain function is that function can be localized to "very" specific locations. However, reductionism falls apart when speaking about emergent phenomenon such as higher order brain functions (e.g. feeling of knowing, consciousness, visio-spatial processing). Just as the intelligence of ants, bees, an colonies of bacteria doesn't exist at the individual level, all of most fundamental human attributes only exists on the group level (purely epiphenomenological).

Because of this, there is a certain uncertainty in both space and time for which these emergent phenomena exist. Time has only recently enter neuroscience which changes things a lot because anesthesia freezes the brain map in time (and space) so isn't a very good model (anymore) to understand emergent phenomena.

This concept that the more you try to localize some aspect of nature the more uncertain you are about another aspect resonates with me.

Can someone point me in the direction to where I can develop mathematical tools so I can maybe obtain some deeper understanding of these emergent phenomena by analyzing how this uncertainty between function, space, and time occurs?

 

[conquest]

I don't completely now how valid an answer this is but I remember when reading David Bohm's introductory book on quantum mechanics that one of those great quantum pioneers (Schrödinger, Planck, Bohr someone big at least) suggested thought was completely due to the uncertainty principle (or something of that sense).

So I think thath book would be a good read then, at the moment I have lent it out otherwise I could have had the quote for you , sorry.

 

[PhysiPhile]

I don't completely now how valid an answer this is but I remember when reading David Bohm's introductory book on quantum mechanics that one of those great quantum pioneers (Schrödinger, Planck, Bohr someone big at least) suggested thought was completely due to the uncertainty principle (or something of that sense).

So I think thath book would be a good read then, at the moment I have lent it out otherwise I could have had the quote for you , sorry.

Hey thanks for the reply.

Thought takes place far into the macroscopic world so Heisenberg's uncertainty or the probabilistic nature of particles isn't applicable. Unfortunately, there have been brilliant physicist who have gone out of their field and made an erroneous connection between brain function and the quantum world to their detriment.

I am referring to an inherent property of emergent phenomenon is the non-localization of it in space. Maybe graph theory can shine some light on this for me...time to start digging.

 

[Stephen Tashi]

Although you can find many papers on "emergent phenomena" etc., you are asking questions about a things about which the scientific world has not reached a consensus (even with your proclamation that thought takes place in the macroscopic world). There aren't any precise and standard definitions of things like "thought" or "emergent phenomena" in mathematics yet.

If you want to talk about things like this using mathematics, you can't take a liberal arts approach and frame your ideas in philosophical language. You have to come down to earth, mathematically speaking. I know of very few people who have been able to do this. One, who might be making a little progress is Luis Rocha. http://informatics.indiana.edu/rocha/

 

[PhysiPhile]

Although you can find many papers on "emergent phenomena" etc., you are asking questions about a things about which the scientific world has not reached a consensus (even with your proclamation that thought takes place in the macroscopic world).

Without getting into a long winded debate, I think it is false to say that a consensus (in neuroscience) has not been reached. We know a single neuron does not exhibit a brain function - claims to the contrary are mostly reductio ad absurdums. We know that brain functions exist inside the skull where there is an astronomically large number of connections between these neural units. These parameters put us far into the macroscopic world where I can use maxwells equations and other classical concepts. It's been half a century since Hodgkin and Huxley developed nonlinear ordinary differential equations to model the neuron using classical mechanics and we have made much progress since then.

There aren't any precise and standard definitions of things like "thought" or "emergent phenomena" in mathematics yet.

That's what I'm trying to do. Presently, I'm in medical school going into neurology (specializing in seizures). I want to work towards application of all the physics, programming, and neuroscience that I've learned in my past and try and make headway into the daunting arena of modulating or replacing brain function. I have no delusions that I will most likely die before seeing anything substantial come into fruition.

If you want to talk about things like this using mathematics, you can't take a liberal arts approach and frame your ideas in philosophical language. You have to come down to earth, mathematically speaking. I know of very few people who have been able to do this. One, who might be making a little progress is Luis Rocha. http://informatics.indiana.edu/rocha/

I am trying not to get caught up in all the philosophical masturbation about does consciousness exist or is it an illusion (that seems to be a hot topic for whatever reason), and just stick to solving the problem (must be the engineering side of me). Thank you for the link - I'll look into him.

 

[lavinia]

A huge common misconception about brain function is that function can be localized to "very" specific locations. However, reductionism falls apart when speaking about emergent phenomenon such as higher order brain functions (e.g. feeling of knowing, consciousness, visio-spatial processing). Just as the intelligence of ants, bees, an colonies of bacteria doesn't exist at the individual level, all of most fundamental human attributes only exists on the group level (purely epiphenomenological).

I don't know a thing about what you are asking but am interested to understand what you are talking about. What do you mean by localize a function to specific locations? Do you mean things like rational thought is in the forebrain typr of thing?

Because of this, there is a certain uncertainty in both space and time for which these emergent phenomena exist. Time has only recently enter neuroscience which changes things a lot because anesthesia freezes the brain map in time (and space) so isn't a very good model (anymore) to understand emergent phenomena.

What is meant by uncertaintly in time? Isn't time time?


[/QUOTE]
This concept that the more you try to localize some aspect of nature the more uncertain you are about another aspect resonates with me.
[/QUOTE]

Can you describe an example?

 

[SW VandeCarr]

I am referring to an inherent property of emergent phenomenon is the non-localization of it in space. Maybe graph theory can shine some light on this for me...time to start digging.

For the most part you are talking about complex mostly non linear systems. The following is just an outline of the breadth of this area of research.

http://www.calresco.org/lucas/quantify.htm

 

[lavinia]

For the most part you are talking about complex mostly non linear systems. The following is just an outline of the breadth of this area of research.

http://www.calresco.org/lucas/quantify.htm

so how is thought an emergent self evolving complex system? What is the model?

 

[PhysiPhile]

What do you mean by localize a function to specific locations?

Reducing multiple observations down to a single principle is what conventional science is all about. For example, connecting the motion of heavenly bodies and the downward motion of falling objects with a single attraction force between the object, F=G(m1*m2)/r^2. Or finding what atoms where made of and then figuring out what those subatomic particles are made of. You do this until you reach a fundamental thing from which all things emerge. We want to know where things are in space and in time. This is the first step into understanding it. For example, deducing the spatial and temperal extent of an electron has caused a revolution in science - See: Heinsenbergs Uncertaintity Principle or EPR-Paradox. So the goal to understanding brain function is to find out where it is in space and time (x,y,z,t). Once we know that, then we can manipulate it.

Do you mean things like rational thought is in the forebrain typr of thing?

Yes, but now we can't just say it's in a certain lobe or area (e.g. Werncike's or Broca's). because we know that to create the most basic functions requires the interaction between multiple area's in series and in parallel so it is intrinsically impossible to localize (x,y,z,t) thought to even a million neurons.

What is meant by uncertaintly in time? Isn't time time?

All brain functions are a function of x,y,z,t. Current brain maps try to localize brain function to a specific region of the brain but while this region may be important it does not mean brain function can be localized to that region. Not only that but also brain function changes dramatically with time. So the auditory cortex can not explain the perception of hearing.

This concept that the more you try to localize some aspect of nature the more uncertain you are about another aspect resonates with me.

Can you describe an example?

This is true of the uncertainty principle. The more we localize a particle in space (time) the less we know about it in time (space). However, this is just an analogy - I don't believe Heisenberg's uncertainty applies to brain function which some argue.

 

[SW VandeCarr]

so how is thought an emergent self evolving complex system? What is the model?

Model? We are talking about models for biological systems of which the brain is part. "Thought" is a subjective appreciation of some of what the brain is doing as it interacts with complex internal and external environments.

 

[lavinia]

Reducing multiple observations down to a single principle is what conventional science is all about. For example, connecting the motion of heavenly bodies and the downward motion of falling objects with a single attraction force between the object, F=G(m1*m2)/r^2. Or finding what atoms where made of and then figuring out what those subatomic particles are made of. You do this until you reach a fundamental thing from which all things emerge. We want to know where things are in space and in time. This is the first step into understanding it. For example, deducing the spatial and temperal extent of an electron has caused a revolution in science - See: Heinsenbergs Uncertaintity Principle or EPR-Paradox. So the goal to understanding brain function is to find out where it is in space and time (x,y,z,t). Once we know that, then we can manipulate it.



Yes, but now we can't just say it's in a certain lobe or area (e.g. Werncike's or Broca's). because we know that to create the most basic functions requires the interaction between multiple area's in series and in parallel so it is intrinsically impossible to localize (x,y,z,t) thought to even a million neurons.



All brain functions are a function of x,y,z,t. Current brain maps try to localize brain function to a specific region of the brain but while this region may be important it does not mean brain function can be localized to that region. Not only that but also brain function changes dramatically with time. So the auditory cortex can not explain the perception of hearing.



This is true of the uncertainty principle. The more we localize a particle in space (time) the less we know about it in time (space). However, this is just an analogy - I don't believe Heisenberg's uncertainty applies to brain function which some argue.

what about the idea that the brain may be a quantum computer?

 

[SW VandeCarr]

what about the idea that the brain may be a quantum computer?

No evidence for that. The warm wet environment of the brain is not conducive to that. In fact, this suggestion is considered borderline crackpottery by most neuroscientists today. Of course that doesn't mean it couldn't be true, but the onus is on those who'd propose such a model.

 

[lavinia]

No evidence for that. The warm wet environment of the brain is not conducive to that. In fact, this suggestion is considered borderline crackpottery by most neuroscientists today. Of course that doesn't mean it couldn't be true, but the onus is on those who'd propose such a model.

When I first heard of this from a friend, he said that if you consider that part of the brain that is a computing machine, it must be an incarnation of one of the possible types of computing machines. The physical form doesn't matter. He said that people originally only thought that there was only one possible type of computing machine, the finite state machine. But then a physicist thought of the quantum computer. Then there were two known possible types of computing machine. Are any others known?

 

[PhysiPhile]

what about the idea that the brain may be a quantum computer?

If you watch this lecture from Walter Lewin at MIT (), you'll understand why quantum mechanics does not apply to neurons.

Classical mechanics is sufficient to explain brain function.

 

[lavinia]

If you watch this lecture from Walter Lewin at MIT (), you'll understand why quantum mechanics does not apply to neurons.

Classical mechanics is sufficient to explain brain function.

ok. So the brain is a Turing machine?

 

[SW VandeCarr]

When I first heard of this from a friend, he said that if you consider that part of the brain that is a computing machine then whatever its physical form, it still must be an incarnation of one of the possible machines. The physical form doesn't matter. He said that people originally only thought that there was only one possible type of computing machine, the finite state machine. But then a physicist thought of the quantum computer. Then there were two known possible types of computing machine. Are any others known?

"Neural" network models perhaps, but I'm not a computer scientist. The massively parallel architecture of the brain is not well modeled by the Turing-Von Neumann computer model. However, at least according to the the current paradigm, brain function is describable in terms of classical physics/chemistry, but the level of complexity means that classical linear mathematics doesn't work very well. Also the connections between specific psychological factors and specific physiologic factors remains largely unknown. However, studies with functional MRIs (fMRIs) have made considerable progress in correlating mental states with brain locations.

 

[PhysiPhile]

ok. So the brain is a Turing machine?

I wouldn't say that...What do you mean by brain? There are certain properties of the brain that are like a turing machine. For example, regulating blood pressure through the baroreflex. But, consciousness does not emerge from this simple reflex arc. More complicated brain functions require a mathematically complex description with enormous amounts of variables that can not be decoupled. So I would say currently brains are not the equivalent to turing machines because turing machines don't even come close to simulating perception.

 

[lavinia]

Not to my knowledge, but I'm not a computer scientist. The massively parallel architecture of the brain is not well modeled by the Turing-Von Newman computer model. However, at least according to the the current paradigm, brain function is describable in terms of classical physics, but the level of complexity means that classical linear mathematics doesn't work very well. Also the connections between specific psychological factors and specific physiologic factors remains largely unknown. However, studies with functional MRIs (fMRIs) have made considerable progress in correlating mental states with brain locations.

- What do you mean by classical linear mathematics?

- What about the research on neural networks? My friend showed me an article where a clam's nervous system was completely described anatomically then modeled with a computer program. Interestingly, as I recall, the program sometimes reached what he called equilibrium states. But... if one thinks of a nerve as a simple switch, then the brain no matter how complex is in theory modelable with a computer program, that is if no other effects make it into a quantum machine or some other machine that has not been discovered yet.

 

[PhysiPhile]

- What do you mean by classical linear mathematics?

The whole brain is not the sum of it's parts (emergence). Classical linear mathematics = sum of its parts.

http://en.wikipedia.org/wiki/Nonlinear_system

- What about the research on neural networks? My friend showed me an article where a clam's nervous system was completely described anatomically then modeled with a computer program. Interestingly, as I recall, the program sometimes reached what he called equilibrium states. But... if one thins of a nerve as a simple switch, then the brain no matter how complex is in theory modelable with a computer program, that is if no other effects make it into a quantum machine or some other machine that has not bee discovered yet.

That claim just isn't true. We are a very far away from modelling the brain because of the many boundary conditions needed that are unknown (see link above).

 

[lavinia]

The whole brain is not the sum of it's parts (emergence). Classical linear mathematics = sum of its parts.

http://en.wikipedia.org/wiki/Nonlinear_system
That claim just isn't true. We are a very far away from modelling the brain because of the many boundary conditions needed that are unknown (see link above).

Instead of referring me to a link why not explain to me why it isn't true.

I know what a non-linear system is but just can't imagine how this has to do with summing parts. Oh well. I will do some reading rather than clogging this thread with naive questions.
Thanks.

 

[PhysiPhile]

Instead of referring me to a link why not explain to me why it isn't true.

I know what a non-linear system is but just can't imagine how this has to do with summing parts. Oh well. I will do some reading rather than clogging this thread with naive questions.
Thanks.

Yeah, I suggest doing some reading if you're interested in this topic. Then if you still have questions once you understand the basics don't hesitate to ask me! I'm learning more all the time about these concepts.

 

[Bacle2]

If you can get a hold of papers by this guy:

http://en.wikipedia.org/wiki/Arnold_J._Mandell

I took a baby Chaos/Math Biology independent study class with him; brain function and chaos is all he talked about .

Sorry I could not find closer links.

 

[PhysiPhile]

Interesting update!

So at the risk of looking very foolish I brought up my idea to a neurologist who specializes in dementia at my medical school.

He told me that he actually published a paper on that topic that I brought up in my initial post. He approached it by using graph theory. I've attached the paper.