Reductionism does not work for biology, we need new maths

What does make the world turn?
User avatar
zeuzzz
Persistent Poster
Posts: 3859
Joined: Sun Jun 16, 2013 5:33 pm
Custom Title: Unicorn Herder

Reductionism does not work for biology, we need new maths

Post by zeuzzz » Mon Aug 05, 2013 9:10 pm

Been thinking about this a lot recently after reading a book about chaos theory, and even if reductionism works wonders for physics and chemistry, surely it will never work for biology? I'm pretty sure we live at the end of the legacy of applying reductionism to biology, the chaos is roving through the system and undermining linear deterministic materialist theories.

The first bad application of reductionism to biology was probably placing too much emphasis on genes, assuming that genes are responsible for any or every characteristic of animal life. Clearly now false.

And this train of thought can still be seen in all sorts of manner of scientific biological research and institutions, but less and less nowadays.

Where it breaks down is in the bifurcating systems inherent all over the body, of which the main characteristic is they are scale free/invarient. You can not reduce them using reductive ideologies. The complexity of the branching becomes scale free. Yet most of the important things in our bodies are bifurcating systems, all the branch points on neurons are bifurcating systems, and they are called bifurcating trees because they literally look like trees. Look at the cardiovascular system. Look at the pulmonary system, the trachea, bronchi, lungs, etc. If you look at a black and white image of a cross section of one of these at any scale they would look identical. A bit like the translationally invariant and scale invariant nature of Maxwells equations leading to scale invariance in the fourth state of matter plasmas, where chaotic EM forces dominate over gravity (on small scales).

You would see the same branching in a blood vessel composed of thousands of cells than you would in a neuron made of one, so it's independent of scale.

So this theme of fractal like bifurcating systems seems replete throughout the body, and in fact biology in general, so how does the body code for this? What is the algorithm for such bifurcating systems? Me's bufuzzled.

One thing is for sure it can not be reductive or deterministic. It's obviously not about genes, or set theory, or linear maths in general. A capillary with millions of ends can not be coded for by tens of thousands of genes. You can't code for recognizing your dog with simple reductive models of neurons, or many of the bifurcating systems that branch out to millions of end points, neurons as a reductive approach also break down with things like facial recognition. There are other ways that reductive models fail in biology but I think I've sufficiently slayed that dead horse above. But I've yet to find the answer to my above question. Anyone have any insight into this?
Always be you, unless you can be a unicorn; then be a unicorn.

User avatar
zeuzzz
Persistent Poster
Posts: 3859
Joined: Sun Jun 16, 2013 5:33 pm
Custom Title: Unicorn Herder

Re: Reductionism does not work for biology, we need new math

Post by zeuzzz » Fri Aug 09, 2013 1:52 am

To get a clue WTF I'm on about it might be nice to link to the book/talk that inspired this post: Why scientific reductionism does not work for biology (Chaos theory)

Comprende?

(Thanks Stanford uni)
Always be you, unless you can be a unicorn; then be a unicorn.

User avatar
SweetPea
Has No Life
Posts: 12924
Joined: Fri Aug 10, 2012 4:11 am
Custom Title: Too Cute

Re: Reductionism does not work for biology, we need new math

Post by SweetPea » Fri Aug 09, 2013 2:20 am

zeuzzz wrote:To get a clue WTF I'm on about it might be nice to link to the book/talk that inspired this post: Why scientific reductionism does not work for biology (Chaos theory)

Comprende?

(Thanks Stanford uni)
Heavy Duty Man. Congratulations on a very good find
How do the Deniers get so lucky?
http://www.skepticforum.com/viewtopic.php?f=16&t=24129" onclick="window.open(this.href);return false;

User avatar
OutOfBreath
Veteran Poster
Posts: 2218
Joined: Sat Jan 22, 2011 1:38 pm
Custom Title: Persistent ponderer
Location: Norway

Re: Reductionism does not work for biology, we need new math

Post by OutOfBreath » Fri Aug 09, 2013 10:29 am

zeuzzz wrote:One thing is for sure it can not be reductive or deterministic. It's obviously not about genes, or set theory, or linear maths in general. A capillary with millions of ends can not be coded for by tens of thousands of genes. You can't code for recognizing your dog with simple reductive models of neurons, or many of the bifurcating systems that branch out to millions of end points, neurons as a reductive approach also break down with things like facial recognition. There are other ways that reductive models fail in biology but I think I've sufficiently slayed that dead horse above. But I've yet to find the answer to my above question. Anyone have any insight into this?

Are any of these exclamations verified, or is it argument from awe/ignorance?

For instance, a basic simple code can easily be the source for "a capillary with millions of end points", since the simple code in no way have to specify each and every end point. The code can simply be that cells of that kind do A, until B occurs then they stop, or go on doing C. Then copy/paste basically. You don't need a blueprint, just contingency measures and general orders. Like an army commander who will not specify all the people who will be killed, buildings destroyed and civilians killed/evacuated by an offensive. He will say, "Advance until you reach the river, and eliminate all opposition encountered."

As for reductionism, it has it's purpose in scientific reasoning, in narrowing down and operationalizing what is studied. If we can't observe something, we assume it isn't there. There may in reality be something that we just don't see of course, but until we do, it's pretty practical to keep it out of any explanations.

Peace
Dan
What is perceived as real becomes real in its consequences.

"Every judgment teeters on the brink of error. To claim absolute knowledge is to become monstrous. Knowledge is an unending adventure at the edge of uncertainty." - Frank Herbert

User avatar
zeuzzz
Persistent Poster
Posts: 3859
Joined: Sun Jun 16, 2013 5:33 pm
Custom Title: Unicorn Herder

Re: Reductionism does not work for biology, we need new math

Post by zeuzzz » Fri Feb 28, 2014 2:43 pm

OutOfBreath wrote: Are any of these exclamations verified, or is it argument from awe/ignorance?

For instance, a basic simple code can easily be the source for "a capillary with millions of end points", since the simple code in no way have to specify each and every end point. The code can simply be that cells of that kind do A, until B occurs then they stop, or go on doing C. Then copy/paste basically. You don't need a blueprint, just contingency measures and general orders. Like an army commander who will not specify all the people who will be killed, buildings destroyed and civilians killed/evacuated by an offensive. He will say, "Advance until you reach the river, and eliminate all opposition encountered."

As for reductionism, it has it's purpose in scientific reasoning, in narrowing down and operationalizing what is studied. If we can't observe something, we assume it isn't there. There may in reality be something that we just don't see of course, but until we do, it's pretty practical to keep it out of any explanations.

Peace
Dan
I don't really disagree with this, reductionism has it's benefits sure but I think some areas have lost all focus by thinking deeper levels somehow amount to more truth ... did you watch the lecture I linked to? At some point (I've forgotten) he outlines why the current dominant paradigm of a neural correlate for every mental function was based on a very powerful initial theory by two research scientists (weavel?) where they found a very strong simple point for point reductive neural path for every piece of information taken in by the retina and eyes. Yet as they showed this amazingly powerful model they showed a lot of intuition by bailing on this idea and researching other things as all the other hundreds of neuroscientists that went to try to find other point for point reductive neural correlates of consciousness have failed miserably ever since pretty much. Maybe Koch et al have done the best work on neural correlates since, but it's far from a fully explanatory model and is far more chaotic than originally assumed.

He basically outlines why linear maths fails totally for biology as the chaos and fractal nature implicit in the system becomes infinitely complex extremely quickly depending on the dimensional assumptions of the way you model and measure it.
Always be you, unless you can be a unicorn; then be a unicorn.

User avatar
SweetPea
Has No Life
Posts: 12924
Joined: Fri Aug 10, 2012 4:11 am
Custom Title: Too Cute

Re: Reductionism does not work for biology, we need new math

Post by SweetPea » Fri Feb 28, 2014 3:05 pm

"Grandmother" neuron :)
"Grandmother's Favourite Symphony" neuron.
How do the Deniers get so lucky?
http://www.skepticforum.com/viewtopic.php?f=16&t=24129" onclick="window.open(this.href);return false;

User avatar
zeuzzz
Persistent Poster
Posts: 3859
Joined: Sun Jun 16, 2013 5:33 pm
Custom Title: Unicorn Herder

Re: Reductionism does not work for biology, we need new math

Post by zeuzzz » Fri Feb 28, 2014 3:48 pm

SweetPea wrote:"Grandmother" neuron :)
"Grandmother's Favourite Symphony" neuron.
Summed up in two phrases, now why didn't I think of that :lol:
Always be you, unless you can be a unicorn; then be a unicorn.

User avatar
SweetPea
Has No Life
Posts: 12924
Joined: Fri Aug 10, 2012 4:11 am
Custom Title: Too Cute

Re: Reductionism does not work for biology, we need new math

Post by SweetPea » Fri Feb 28, 2014 4:26 pm

The old hypothesis failed, but a reduced version of it still hangs in there


http://discovermagazine.com/2009/jun/15 ... xC38IWlGcM" onclick="window.open(this.href);return false;

The Halle Berry neuron is a more attractive idea

How do the Deniers get so lucky?
http://www.skepticforum.com/viewtopic.php?f=16&t=24129" onclick="window.open(this.href);return false;

User avatar
zeuzzz
Persistent Poster
Posts: 3859
Joined: Sun Jun 16, 2013 5:33 pm
Custom Title: Unicorn Herder

Re: Reductionism does not work for biology, we need new math

Post by zeuzzz » Fri Feb 28, 2014 5:45 pm

SweetPea wrote:The Halle Berry neuron is a more attractive idea

Image
Now THAT is a new neural correlate I could get used to :D
Always be you, unless you can be a unicorn; then be a unicorn.