Random mutations cannot explain evolution of humans

[wogoga]

So whereas the genetic information of a human only constitutes a small fraction of the storage capacity of a DVD disc of 4.7 Gigabyte, in order to store all the synaptic connections of a three-year-old child, around a million DVD discs are needed.

Congratulations Wolfgang - you have just proved that learning is impossible since all of the synapse connections in the brain are hard-coded in DNA!

What I have shown is this: If all the synaptic connections of a three-year-old child were hard-coded (ignoring details such as location in the brain, length of axons, synaptic plasticity, etc.), in the order of 10^16 byte = 10,000,000 Gigabyte would be needed. Yet the whole active DNA of a human is less than 10^8 byte = 0.1 Gigabyte. And this information is assumed to be responsible also for many thousand enzyme species, differentiation into more than two hundred cell types, the highly complex anatomy of the human body at all levels, human learning capacity, instinctive behaviour and even talents.

Or take the brain of a newborn instead of a three-year-old. The complexity of this brain is the main precondition for learning after birth.

As any biologist can tell you, neural development starts from a mostly random network of synaptic connections. Connections are then reinforced by learning.

An extract from your link (emphasis mine):

"Some landmarks of embryonic neural development include the birth and differentiation of neurons from stem cell precursors, the migration of immature neurons from their birthplaces in the embryo to their final positions, outgrowth of axons from neurons and guidance of the motile growth cone through the embryo towards postsynaptic partners, the generation of synapses between these axons and their postsynaptic partners, and finally the lifelong changes in synapses which are thought to underlie learning and memory.

Typically, these neurodevelopmental processes can be broadly divided into two classes: activity-independent mechanisms and activity-dependent mechanisms. Activity-independent mechanisms are generally believed to occur as hardwired processes determined by genetic programs played out within individual neurons. These include differentiation, migration and axon guidance to their initial target areas. These processes are thought of as being independent of neural activity and sensory experience. Once axons reach their target areas, activity-dependent mechanisms come into play. Neural activity and sensory experience will mediate formation of new synapses, as well as synaptic plasticity, which will be responsible for refinement of the nascent neural circuits."​

Wikipedia on Axon guidance:

"Axon guidance (also called axon pathfinding) is a subfield of neural development concerning the process by which neurons send out axons to reach the correct targets. Axons often follow very precise paths in the nervous system, and how they manage to find their way so accurately remains a major puzzle."​

One thing is sure: the information cannot come from the DNA, simply because the DNA does not contain enough information. If you try to understand the following quote from Psychons and their Evolution, you can resolve this major puzzle yourself.

The maturation of a protein from the corresponding amino acid chain can happen in the following way:

In important (evolutionarily older) sequences of the chain, amino acids become active, that is they get animated by psychons. Because of environment continuity these psychons are the ones which have built up the same protein (or the same sequence of different proteins) innumerable times. These psychons build up protein parts which can be animated as a whole by other psychons which then build up the complete protein.​

So it also becomes comprehensible that RNA sequences (introns) are able to cut out themselves or that order is maintained during DNA recombination.​

Cheers, Wolfgang

Use as few hypotheses as possible, but not fewer!

 

[sol invictus]

One thing is sure: the information cannot come from the DNA, simply because the DNA does not contain enough information. If you try to understand the following quote from Psychons and their Evolution, you can resolve this major puzzle yourself.

What puzzle?

 

[Jimbo07]

I think you guys are missing the point.

Wogoga isn't saying that evolution didn't happen. His argument is that evolution by natural selection of mutations didn't happen because the probability is too low, so it must have been directed by an external agency.

I obviously missed it. Since it's cherry-picking, it seems to me that evolution happened by (insert favourite paranormal agent here) is the same as saying evolution didn't happen.

 

[Dancing David]

Or take the brain of a newborn instead of a three-year-old. The complexity of this brain is the main precondition for learning after birth.

More made up suff, wonderful.

An extract from your link (emphasis mine):

"Some landmarks of embryonic neural development include the birth and differentiation of neurons from stem cell precursors, the migration of immature neurons from their birthplaces in the embryo to their final positions, outgrowth of axons from neurons and guidance of the motile growth cone through the embryo towards postsynaptic partners, the generation of synapses between these axons and their postsynaptic partners, and finally the lifelong changes in synapses which are thought to underlie learning and memory.

Typically, these neurodevelopmental processes can be broadly divided into two classes: activity-independent mechanisms and activity-dependent mechanisms. Activity-independent mechanisms are generally believed to occur as hardwired processes determined by genetic programs played out within individual neurons. These include differentiation, migration and axon guidance to their initial target areas. These processes are thought of as being independent of neural activity and sensory experience. Once axons reach their target areas, activity-dependent mechanisms come into play. Neural activity and sensory experience will mediate formation of new synapses, as well as synaptic plasticity, which will be responsible for refinement of the nascent neural circuits."​

Oh boy, more loaded language, target, find a neurology source for that?

Wikipedia on Axon guidance:

"Axon guidance (also called axon pathfinding) is a subfield of neural development concerning the process by which neurons send out axons to reach the correct targets. Axons often follow very precise paths in the nervous system, and how they manage to find their way so accurately remains a major puzzle."​

Funny how some questions have answers: PubMed 'axon guidance', what mystery?
Items 1 - 20 of 2707Page of 136Next
1: Related ArticlesChisholm AD.

Semaphorin signaling in morphogenesis: found in translation.
Genes Dev. 2008 Apr 15;22(8):955-9.
PMID: 18413708 [PubMed - in process]2: Related ArticlesSaarikangas J, Hakanen J, Mattila PK, Grumet M, Salminen M, Lappalainen P.

ABBA regulates plasma-membrane and actin dynamics to promote radial glia extension.
J Cell Sci. 2008 Apr 14; [Epub ahead of print]
PMID: 18413296 [PubMed - as supplied by publisher]3: Related Articles, LinksMakita T, Sucov HM, Gariepy CE, Yanagisawa M, Ginty DD.

Endothelins are vascular-derived axonal guidance cues for developing sympathetic neurons.
Nature. 2008 Apr 10;452(7188):759-63.
PMID: 18401410 [PubMed - in process]4: Related Articles, LinksMoore SW, Sun KL, Xie F, Barker PA, Conti M, Kennedy TE.

Soluble adenylyl cyclase is not required for axon guidance to netrin-1.
J Neurosci. 2008 Apr 9;28(15):3920-4.
PMID: 18400890 [PubMed - in process]5: Related Articles, LinksGómez M, Hernández ML, Pazos MR, Tolón RM, Romero J, Fernández-Ruiz J.

Colocalization of CB(1) receptors with L1 and GAP-43 in forebrain white matter regions during fetal rat brain development: Evidence for a role of these receptors in axonal growth and guidance.
Neuroscience. 2008 Feb 29; [Epub ahead of print]
PMID: 18400407 [PubMed - as supplied by publisher]6: Related Articles, LinksCaunt M, Mak J, Liang WC, Stawicki S, Pan Q, Tong RK, Kowalski J, Ho C, Reslan HB, Ross J, Berry L, Kasman I, Zlot C, Cheng Z, Le Couter J, Filvaroff EH, Plowman G, Peale F, French D, Carano R, Koch AW, Wu Y, Watts RJ, Tessier-Lavigne M, Bagri A.

Blocking neuropilin-2 function inhibits tumor cell metastasis.
Cancer Cell. 2008 Apr;13(4):331-42.
PMID: 18394556 [PubMed - in process]7: Related Articles, LinksLi H, Kulkarni G, Wadsworth WG.

RPM-1, a Caenorhabditis elegans protein that functions in presynaptic differentiation, negatively regulates axon outgrowth by controlling SAX-3/robo and UNC-5/UNC5 activity.
J Neurosci. 2008 Apr 2;28(14):3595-603.
PMID: 18385318 [PubMed - in process]8: Related Articles, LinksCanty AJ, Murphy M.

Molecular mechanisms of axon guidance in the developing corticospinal tract.
Prog Neurobiol. 2008 Feb 15; [Epub ahead of print]
PMID: 18378059 [PubMed - as supplied by publisher]9: Related Articles, LinksLu YJ, Xu NW, Yang WQ.

Immunofluorescence laser confocal expression and localization study of rat nerve growth guidance cues Netrin-1 and Slit2 after spinal cord injury.
Chin J Traumatol. 2008 Apr;11(2):98-103.
PMID: 18377713 [PubMed - in process]10: Related Articles, LinksZhou Y, Gunput RA, Pasterkamp RJ.

Semaphorin signaling: progress made and promises ahead.
Trends Biochem Sci. 2008 Apr;33(4):161-70.
PMID: 18374575 [PubMed - in process]11: Related Articles, LinksWolf AM, Lyuksyutova AI, Fenstermaker AG, Shafer B, Lo CG, Zou Y.

Phosphatidylinositol-3-kinase-atypical protein kinase C signaling is required for Wnt attraction and anterior-posterior axon guidance.
J Neurosci. 2008 Mar 26;28(13):3456-67.
PMID: 18367611 [PubMed - in process]12: Related Articles, LinksScolnick JA, Cui K, Duggan CD, Xuan S, Yuan XB, Efstratiadis A, Ngai J.

Role of IGF signaling in olfactory sensory map formation and axon guidance.
Neuron. 2008 Mar 27;57(6):847-57.
PMID: 18367086 [PubMed - in process]13: Related Articles, LinksKaneko-Goto T, Yoshihara S, Miyazaki H, Yoshihara Y.

BIG-2 mediates olfactory axon convergence to target glomeruli.
Neuron. 2008 Mar 27;57(6):834-46.
PMID: 18367085 [PubMed - in process]14: Related Articles, LinksColley KJ.

Structural Basis for the Polysialylation of the Neural Cell Adhesion Molecule.
Neurochem Res. 2008 Mar 25; [Epub ahead of print]
PMID: 18363101 [PubMed - as supplied by publisher]15: Related Articles, LinksPocock R, Mione M, Hussain S, Maxwell S, Pontecorvi M, Aslam S, Gerrelli D, Sowden JC, Woollard A.

Neuronal function of Tbx20 conserved from nematodes to vertebrates.
Dev Biol. 2008 Feb 21; [Epub ahead of print]
PMID: 18358469 [PubMed - as supplied by publisher]16: Related Articles, LinksSchwarz Q, Vieira JM, Howard B, Eickholt BJ, Ruhrberg C.

Neuropilin 1 and 2 control cranial gangliogenesis and axon guidance through neural crest cells.
Development. 2008 May;135(9):1605-13. Epub 2008 Mar 20.
PMID: 18356247 [PubMed - in process]17: Related Articles, LinksKurusu M, Zinn K.

Receptor tyrosine phosphatases regulate birth order-dependent axonal fasciculation and midline repulsion during development of the Drosophila mushroom body.
Mol Cell Neurosci. 2008 Feb 13; [Epub ahead of print]
PMID: 18356078 [PubMed - as supplied by publisher]18: Related Articles, LinksMamiya PC, Hennesy Z, Zhou R, Wagner GC.

Changes in attack behavior and activity in EphA5 knockout mice.
Brain Res. 2008 Feb 29; [Epub ahead of print]
PMID: 18353288 [PubMed - as supplied by publisher]19: Related Articles, LinksNakakura T, Suzuki M, Watanabe Y, Tanaka S.

Possible involvement of brain-derived neurotrophic factor (BDNF) in the innervation of dopaminergic neurons from the rat periventricular nucleus to the pars intermedia.
Zoolog Sci. 2007 Nov;24(11):1086-93.
PMID: 18348609 [PubMed - in process]20: Related Articles, LinksBaeriswyl T, Stoeckli ET.

Axonin-1/TAG-1 is required for pathfinding of granule cell axons in the developing cerebellum.
Neural Develop. 2008 Mar 17;3(1):7 [Epub ahead of print]
PMID: 18346270 [PubMed - as supplied by publisher]

One thing is sure: the information cannot come from the DNA, simply because the DNA does not contain enough information.

Duh, and so if a cell release chemicals that cause groth of axons around it, then what is that?

Duh.

If you try to understand the following quote from Psychons and their Evolution, you can resolve this major puzzle yourself.

The maturation of a protein from the corresponding amino acid chain can happen in the following way:

In important (evolutionarily older) sequences of the chain, amino acids become active, that is they get animated by psychons. Because of environment continuity these psychons are the ones which have built up the same protein (or the same sequence of different proteins) innumerable times. These psychons build up protein parts which can be animated as a whole by other psychons which then build up the complete protein.​

So it also becomes comprehensible that RNA sequences (introns) are able to cut out themselves or that order is maintained during DNA recombination.​

Cheers, Wolfgang

Use as few hypotheses as possible, but not fewer!

Yeah I found psychons all over PubMed too!

 

[Dr Adequate]

What I have shown is this: If all the synaptic connections of a three-year-old child were hard-coded ...

Which they aren't

And this information is assumed to be responsible also for many thousand enzyme species, differentiation into more than two hundred cell types, the highly complex anatomy of the human body at all levels, human learning capacity, instinctive behaviour and even talents.

Substitute "known" for "assumed", and that's about right.

"Axon guidance (also called axon pathfinding) is a subfield of neural development concerning the process by which neurons send out axons to reach the correct targets. Axons often follow very precise paths in the nervous system, and how they manage to find their way so accurately remains a major puzzle."

Typical crank reasoning. "If I can't explain it, then I can explain it, psychons do it by magic". Usually it's not psychons, but God or space aliens or the Secret Consipracy of Evil Jews.

One thing is sure: the information cannot come from the DNA, simply because the DNA does not contain enough information.

If that was "sure", then wouldn't you have a scrap of a shred of evidence for it?

All you need to do is find a protein the synthesis of which is not directed by DNA.

Of course, this would not confirm the "psychons" hypothesis and more than it would confirm the goddidit hypothesis, the "space aliens are telporting proteins into my body" hypothesis, the "magic gene fairy" hypothesis, or the hypothesis that it's synthesized by some non-magical biological process that we haven't found out about yet.

But it would be a start. So, please name one human protein that doesn't have a corresponding gene.

So it also becomes comprehensible that RNA sequences (introns) are able to cut out themselves or that order is maintained during DNA recombination.

Actually, that's comprehensible in terms of biochemistry, without dragging in imaginary invisible entities. If you want to know how self-splicing introns self-splice (and most introns, BTW, are not self-splicing) then you could start with that article on [swiki]splicing[/swiki] that I keep referring you to, or you could go directly to this link.

 

[zosima]

What I have shown is this: If all the synaptic connections of a three-year-old child were hard-coded (ignoring details such as location in the brain, length of axons, synaptic plasticity, etc.), in the order of 10^16 byte = 10,000,000 Gigabyte would be needed. Yet the whole active DNA of a human is less than 10^8 byte = 0.1 Gigabyte. And this information is assumed to be responsible also for many thousand enzyme species, differentiation into more than two hundred cell types, the highly complex anatomy of the human body at all levels, human learning capacity, instinctive behaviour and even talents.

I agree hard-coding all the connections of the adult would be quite tricky. Also, if our brains were hard-coded, we probably wouldn't have things like variation in memory from person to person, personalities determined by experience, accumulation of knowledge.

Of course no one would argue the connections are completely hard-coded.

As you said there is activity dependent and activity independent development, but its really more fair to characterize it in four stages. Neurogenesis, Migration, Axonal projection, Activity dependent tuning. Postnatally there is also another phase of activity dependent tuning and neural die off. Without explaining all the details, basically the brain gets the neurons roughly connected then uses test patterns and later real sensory stimuli to decide which of the connections to keep.
If you are curious about how this happens I would look up Netrins, or Bone Morphogenic protein or Sonic Hedge Hog(the axonal guidance cue that was named after the sega character), its not really that mysterious.

So I guess your point is well....pointless.

This means you've dropped your silliness about evolution,too. Ya wogoga?

 

[cyborg]

What I have shown is this: If all the synaptic connections of a three-year-old child were hard-coded (ignoring details such as location in the brain, length of axons, synaptic plasticity, etc.), in the order of 10^16 byte = 10,000,000 Gigabyte would be needed. Yet the whole active DNA of a human is less than 10^8 byte = 0.1 Gigabyte.

If all the items in the set of natural numbers were hard-coded data in the order of infinite Gb would needed. Yet the whole of the Peano Axioms can be written on the back of a cigarette packet.

Mathematics is clearly a fradulent enterprise.

 

[kjkent1]

What I have shown is this: If all the synaptic connections of a three-year-old child were hard-coded (ignoring details such as location in the brain, length of axons, synaptic plasticity, etc.), in the order of 10^16 byte = 10,000,000 Gigabyte would be needed. Yet the whole active DNA of a human is less than 10^8 byte = 0.1 Gigabyte. And this information is assumed to be responsible also for many thousand enzyme species, differentiation into more than two hundred cell types, the highly complex anatomy of the human body at all levels, human learning capacity, instinctive behavior and even talents.
(clip...)

Equations, formulae, computations, calculations, etc., are only models of reality. For any mathematical model to be useful, it must be capable of predicting real-world outcomes with better accuracy than mere guesswork.

If your mathematics is correct, you should be able to predict something about the future of human evolution with some amount of certainty -- and you should be able to demonstrate your predictions using your math.

So...show us how evolution actually works with a mathematical model -- use that model to predict future outcomes with reasonable certainty, and we will applaud you. Otherwise, all of your "math" is really just a talisman -- one that we've all read one hundred times before.

 

[zosima]

Equations, formulae, computations, calculations, etc., are only models of reality. For any mathematical model to be useful, it must be capable of predicting real-world outcomes with better accuracy than mere guesswork.

If your mathematics is correct, you should be able to predict something about the future of human evolution with some amount of certainty -- and you should be able to demonstrate your predictions using your math.

So...show us how evolution actually works with a mathematical model -- use that model to predict future outcomes with reasonable certainty, and we will applaud you. Otherwise, all of your "math" is really just a talisman -- one that we've all read one hundred times before.

Wogoga's math sucks. Wogoga's problem is that he has a hard time producing a mathematical model that is actually correct.

Also, why does he use bytes to describe the information in the genome? I'm not an expert in genetics, but I was under the impression that the preferred unit was bp, kbp,mbp,gbp. Basically orders of magnitude of base pairs. Where does he even go from 10^15 synapses to 10^16 bytes? Somehow he's decided you need 10*2^3 bits to code the connection of a synapse? Why that number. Also I'm not even sure he's using byte correctly.

For example 10^16 bytes =~ 9,000,000 GB = 8.8 PB(peta bytes)
A gigabyte is actually 2^30 bytes not 10^9 bytes That said all the rest of the assumptions are bunk, so there is no reason that this one is any worse than the rest, only demonstrative of them.

 

[rocketdodger]

"Activity-independent mechanisms are generally believed to occur as hardwired processes determined by genetic programs played out within individual neurons. These include differentiation, migration and axon guidance to their initial target areas. These processes are thought of as being independent of neural activity and sensory experience. Once axons reach their target areas, activity-dependent mechanisms come into play. Neural activity and sensory experience will mediate formation of new synapses, as well as synaptic plasticity, which will be responsible for refinement of the nascent neural circuits."​

You are under the mistaken impression that the brain of a newborn is complex -- it is not. It is more like solid block of stone before the sculpter has started working on it. Or an empty flash drive.

Wikipedia on Axon guidance:

"Axon guidance (also called axon pathfinding) is a subfield of neural development concerning the process by which neurons send out axons to reach the correct targets. Axons often follow very precise paths in the nervous system, and how they manage to find their way so accurately remains a major puzzle."​

The "puzzle" part here refers to a lack of data, not a lack of understanding of the data.

One thing is sure: the information cannot come from the DNA, simply because the DNA does not contain enough information.

Yes, it does. Try defining "information," then looking at how much of this "information" needs to come from DNA, and maybe you will learn something.

 

[wogoga]

Typical crank reasoning. "If I can't explain it, then I can explain it, psychons do it by magic".

Typical dogmatic reasoning: "Despite contradictions and lacking explanations, our beliefs are nevertheless correct."

Substantial scientific progress is only possible if we take seriously the problems of our current theories and explanations, and if we are able to play with other hypotheses. As a result of biological evolution, uncommon hypotheses seem to be in contradiction with reality. However we should not forget that what seems to us as reality, is also a construction of our souls and brains.

A typical sign of dogmatism is ignoring logical inconsistencies. For instance: On the one hand, the fact that computer-implemented algorithms cannot be improved by randomly changing bits or bytes, is declared irrelevant because "genes are not computer programs". On the other hand, the highly complex architecture of the human brain at birth is explained by the assumption that the necessary information is somehow generated by an algorithm. Do you have any idea, how enzymes could implement such an algorithm?

And by the way, panpsychism (or better: pandualism) is a fully legitimate hypothesis with a long tradition. I know that it is a long and difficult process to substantially change one's own world view, e.g. from pure materialism to the recognition of psychons. For me, this step was not so difficult, because already in former lives I had considered panpsychism as a reasonable scientific hypothesis. See also my posts #29, #51 and #270 from "Reincarnation as a trivial scientific fact".

Some further arguments suggesting information in the form of psychons from the psychon theory:

Not even the assumption that proteins are fully coded by the DNA is true. The genetic code includes twenty amino acids. Apart from these amino acids many proteins contain other amino acids and other components, which are not coded. The genetic code is not universally valid as it was initially assumed. Several exceptions have been found.

Genes of plants and animals regularly contain non-coding sequences. These introns must be cut out from the RNA copies of the genes. The information indicating which regions represent no code and must be removed is not coded. Some introns even cut out themselves. In several cases, RNA nucleotides are changed, deleted or inserted (RNA editing) before translation starts. In order to produce correct proteins, ribosomes sometimes skip nucleotides instead of translating them. Even from the translated sequences sometimes parts are cut out before protein folding starts. All this is not coded!

After transcription, many amino acid sequences efficiently take on a stable form. Biotechnologically produced, random sequences do not fold to a protein. The common explanation is that proteins have been selected during evolution to fold properly. Yet, if only a very small proportion of possible sequences take on a stable form, only this small proportion can undergo selection of protein function, and the probability that random mutations destroy stability is very high. Furthermore, it is improbable that a protein, selected for a stable form, also acts as a catalyst for complex functions. On the other hand, there are related proteins of similar form and function, whose amino acid sequences have drifted apart substantially. There are even cases where the completely different amino acid sequences, corresponding to different reading frames of a given RNA sequence (frameshift), result in correct proteins or parts of proteins.​

Cheers, Wolfgang

 

[drkitten]

A typical sign of dogmatism is ignoring logical inconsistencies. For instance: On the one hand, the fact that computer-implemented algorithms cannot be improved by randomly changing bits or bytes, is declared irrelevant because "genes are not computer programs".

Goodness me. They can't?

You haven't heard of "genetic algorithms"? I spent something like five years of my life doing exactly that --- randomly changing bits and bytes in computer-implemented algorithms and looking to see which of the random changes resulted in performance iimprovements. I've personally published something like thirty papers based on findings generated by this technique.

More importantly, they're not papers about "genetic algorithms" per se. You can't get papers published about "genetic algorithms," any more, since the technique is so standard no program committee will consider it novel. GA are yet another optimization technique (like hill-climbing, but better at avoiding the problem of local maxima); any reearcher in any field of modelling, from aeronautics to zoology, is generally familiar with GA as a generic problem-solving technique if they're at all competent.

Should I return my Ph.D. on the grounds that what I did -- and what the committee saw me do, and accepted, and what thousands of researchers world-wide do routinely on a daily basis -- is impossible?

And by the way, panpsychism (or better: pandualism) is a fully legitimate hypothesis with a long tradition.

So is the four humour theory of disease; that dates back to Galen. Something else the two hypotheses have in common is that they're both completely and entirely wrong. Geocentrism? The flat Earth? Sympathetic magic? People have been wrong for a very long time.

I know that it is a long and difficult process to substantially change one's own world view,

Not at all. I'm a professional researcher. Show me some evidence that my worldview is wrong, and I'll change it. But you've got no evidence.

 

[Dancing David]

Typical dogmatic reasoning: "Despite contradictions and lacking explanations, our beliefs are nevertheless correct."

And the fact that you don't address specific point and just continue to repeat your self is what?

Care to comment about how natural selection works and counter the multiple explanations given?

No because you are dogmatic.

Substantial scientific progress is only possible if we take seriously the problems of our current theories and explanations, and if we are able to play with other hypotheses.

Considering you haven't a shred of evidence your hypothesis will remain speculation solely.

As a result of biological evolution, uncommon hypotheses seem to be in contradiction with reality. However we should not forget that what seems to us as reality, is also a construction of our souls and brains.

that is fantasy, not reality.

A typical sign of dogmatism is ignoring logical inconsistencies. For instance: On the one hand, the fact that computer-implemented algorithms cannot be improved by randomly changing bits or bytes, is declared irrelevant because "genes are not computer programs".

well they aren't nor is the nucleus and mitochondria like a CPU, nor is the gene like RAM and on and on.

Building upon bad analogies does not make science, it makes bad generalization.

On the other hand, the highly complex architecture of the human brain at birth is explained by the assumption that the necessary information is somehow generated by an algorithm.

You haven't demonstrated the first part yet, still ignoring the evidence?

Figures.

Do you have any idea, how enzymes could implement such an algorithm?

I linked to twent papers on PubMed on one small part of it.

But please act ignorant.

And by the way, panpsychism (or better: pandualism) is a fully legitimate hypothesis with a long tradition. I know that it is a long and difficult process to substantially change one's own world view, e.g. from pure materialism to the recognition of psychons. For me, this step was not so difficult, because already in former lives I had considered panpsychism as a reasonable scientific hypothesis. See also my posts #29, #51 and #270 from "Reincarnation as a trivial scientific fact".

Yeah and I was a giant clam and a tree, might as well say that a fairy told you.

 

[drkitten]

Also, why does he use bytes to describe the information in the genome? I'm not an expert in genetics, but I was under the impression that the preferred unit was bp,

The units for information are bits and bytes. The problem with base pairs as a unit is that we don't really know how much information a base pair contains (we have an upper bound of 2 bits/pair, but that's not that helpful....)

In order to make a synapse, you need to know which of the 10^11 (~= 2^40) or so neurons is the synapse target, yes? If you assume (dangerous word there) that the target distrubution is uniiform, then it requires forty bits per neuron to specify the target.

Of course, the information doesn't all need to come from the DNA (the environment is informative, too) and we have no reason to assume that the target distribution is uniform (which is a worst-case scenario).

 

[wollery]

A typical sign of dogmatism is ignoring logical inconsistencies. For instance: On the one hand, the fact that computer-implemented algorithms cannot be improved by randomly changing bits or bytes, is declared irrelevant because "genes are not computer programs". On the other hand, the highly complex architecture of the human brain at birth is explained by the assumption that the necessary information is somehow generated by an algorithm. Do you have any idea, how enzymes could implement such an algorithm?

What Dr Kitten said, genetic algorithms. I had a friend whose PhD was almost entirely based on implementation of a genetic algorithm. The entire point was that the algorithm adapted, keeping changes which improved its fitness, and rejecting changes which diminished its fitness.

 

[zosima]

The units for information are bits and bytes. The problem with base pairs as a unit is that we don't really know how much information a base pair contains (we have an upper bound of 2 bits/pair, but that's not that helpful....)

So then there really isn't a way to go from bp to bits/bytes.

I don't think bits and bytes are units of information in the same sense that a meter is a unit of length. Its just a measure of the number of binary digits we can use to write down a number, so essentially just a range of numbers.

In order to make a synapse, you need to know which of the 10^11 (~= 2^40) or so neurons is the synapse target, yes? If you assume (dangerous word there) that the target distrubution is uniiform, then it requires forty bits per neuron to specify the target.

Ya, I was trying to draw attention to the weakness of this assumption. The idea that neural connections would somehow be specified by neuron name is tenuous at best. There are no molecules during development that could serve as a way to 'address' or 'target' neurons specifically. The best that might be able to happen is if you had a number of opposing chemical releasers around the brain that would set up a coordinate system you could use to do some sort of rough positional targeting.

Of course, the information doesn't all need to come from the DNA (the environment is informative, too) and we have no reason to assume that the target distribution is uniform (which is a worst-case scenario).

I would agree, the majority of the information from the genome only gives a rough outline.

I'm really not sure what you mean by uniform target distribution. How are you imagining that synaptic connections are set up?

 

[drkitten]

So then there really isn't a way to go from bp to bits/bytes.

Oh, there is, but we don't know it yet.

I don't think bits and bytes are units of information in the same sense that a meter is a unit of length. Its just a measure of the number of binary digits we can use to write down a number, so essentially just a range of numbers.

No, that's exactly what bits and bytes are. The underlying mathematics (and theory) is given in Claude Shannon's classic 1948 paper "A Mathematical Theory of Communications" and the sixty-odd years of work in "information theory" since then; it's a quite well-developed branch of mathematics, physics, and computer science. A "bit," technically speaking, is the amount of information contained in the tossing of a (fair) coin, and that can be generalized as a function over any probability distribution. The rolling of a fair die, for example, has between two and three bits of information; a loaded die will contain slightly less information because the die itself is more predictable. Rolling an 8-sided die (they look like octahedra) gives exactly three bits of information, while rolling
a 16-sided die gives you four. This isn't really an appropriate place for me to give you a thirty-page tutorial on information theory, but if you can't get satisfaction from Google, feel free to ask and I (or Complexity, for that matter) will be happy to answer.

Ya, I was trying to draw attention to the weakness of this assumption. The idea that neural connections would somehow be specified by neuron name is tenuous at best.

On the other hand, there is definitely something that guides the dvelopment of neural pathways, whether it be a chemical gradient, and there must be something that tells a neuron to grow here and not there. Since a neuron could in theory grow anywhere, but doesn't, there must be some process that tells it wher to go -- and since the set of possible paths is much much larger than the set of paths that we actually see, there must be something adjusting path-space. That "something," by definition, contains information in some form.

I'm really not sure what you mean by uniform target distribution. How are you imagining that synaptic connections are set up?

There are approximately 10^11 neurons in the human brain, and in theory, any particular synapse can be connected to any of those. If for some reason a particular neuron in the visual cortex never connects to any of the neurons in the hippocampus, then that "reason" must have a a biological specification detailing which neurons it will not connect to. The mechanism is irrelevant -- all that is relevant is that the specification exists and can dictate where the synapses can and cannot go.

Numerically, the worst possible case is where any neuron "could" in theory connect to any other one, with equal probability. In this case, determining which neuron will be connected to contains exactly as much information as in the rolling of a 10^11-sided die, which is about 40 bits or 8 bytes. If the probabilities are not equal, then it's the rolling of a loaded 10^11-sided die, which has less information (depending upon how loaded the die is).

 

[zosima]

Oh, there is, but we don't know it yet.

There might be an average amount of information, but there is no way to assume its not a variable mapping. We know it is a variable mapping, because some amino acids can be coded in more than one way. And that doesn't even take things like protein structure into account.

No, that's exactly what bits and bytes are. The underlying mathematics (and theory) is given in Claude Shannon's classic 1948 paper "A Mathematical Theory of Communications" and the sixty-odd years of work in "information theory" since then; it's a quite well-developed branch of mathematics, physics, and computer science. A "bit," technically speaking, is the amount of information contained in the tossing of a (fair) coin, and that can be generalized as a function over any probability distribution. The rolling of a fair die, for example, has between two and three bits of information; a loaded die will contain slightly less information because the die itself is more predictable. Rolling an 8-sided die (they look like octahedra) gives exactly three bits of information, while rolling
a 16-sided die gives you four. This isn't really an appropriate place for me to give you a thirty-page tutorial on information theory, but if you can't get satisfaction from Google, feel free to ask and I (or Complexity, for that matter) will be happy to answer.

I understand how information theory works. I also understand the mathematics behind it, and I know that bit ends up being a bit because of the application of the base 2 logarithm in Shannon's theory. Logarithm's in other bases tell you just as much. Thats why I would say its a measurement but its not unit. In the sense that it has a no correspondence to some calibrated physical property. It simply distinguishes the number of states that a system or an information channel can take. In the same way that you can have physical quantities with units that cancel and become unit-less quantities. In the same way that probabilities, counts, and ratios don't have units. The log of a probability, or distribution, or number of states, also doesn't have units.

Or I could just take the much more conservative tack and say, it's not SI, it doesn't have an SI correspondence, and it certainly isn't derived from a combination of SI units. I mean its not like there is some ideally balanced coin stored in France somewhere that allows us to be sure just how much information is in a bit. I guess we flip that coin whenever we have to calibrate computer chips. LoL...just to make sure their bits are fair....

On the other hand, there is definitely something that guides the dvelopment of neural pathways, whether it be a chemical gradient, and there must be something that tells a neuron to grow here and not there. Since a neuron could in theory grow anywhere, but doesn't, there must be some process that tells it wher to go -- and since the set of possible paths is much much larger than the set of paths that we actually see, there must be something adjusting path-space. That "something," by definition, contains information in some form.

Yes that something is well understood. Axons follow chemical gradients, amongst other things. But they're generally just attractive or repulsive and there aren't even close to enough types to attract every single axon to every single axon. They're essentially very rough instructions, they're in no way exact. Of course the information has to come from somewhere, but in all likelihood, its coming from the environment in a pretty high entropy sort of way. It's easier to think of the specific connections in the human brain as underspecified by both the genome and the information we have about the womb. In other words that information is going to vary a lot between individuals. Thats why the brain uses a competitive algorithm with pruning and die-off. It maintains an equivalence of result but not of architecture specifics.

There are approximately 10^11 neurons in the human brain, and in theory, any particular synapse can be connected to any of those. If for some reason a particular neuron in the visual cortex never connects to any of the neurons in the hippocampus, then that "reason" must have a a biological specification detailing which neurons it will not connect to. The mechanism is irrelevant -- all that is relevant is that the specification exists and can dictate where the synapses can and cannot go.

Numerically, the worst possible case is where any neuron "could" in theory connect to any other one, with equal probability. In this case, determining which neuron will be connected to contains exactly as much information as in the rolling of a 10^11-sided die, which is about 40 bits or 8 bytes. If the probabilities are not equal, then it's the rolling of a loaded 10^11-sided die, which has less information (depending upon how loaded the die is).

Sure you can think of it that way, but as I've explained above its pretty unilluminating. Basically neurons in one tissue just project position to position to another tissue, with quite a bit of variance.

To put it in a more human understandable form the genome has instructions like:
Project: Retina->Thalamus
Project reverse layering: Thalamus->V1

So 40 bits is such a loose bound, insofar as it tells us about information thats conserved between individuals ,that its essentially useless.

 

[kjkent1]

A typical sign of dogmatism is ignoring logical inconsistencies.

A typical sign of an internet troll is to always avoid the argument of one's opponents, by:

1. Ignoring them.
2. Ridiculing them.

Followed by restating one's position again as if it were self evidently correct.

Anyone can poke a hole in a dam. Very few can build a dam. Your argument is just another attempt to poke a hole in evolutionary theory. The theory remains standing, despite occasional holes, because those who know how to build a dam, work on the holes until they are filled.

So, poke away -- it will avail you not. Only when you actually build your own "dam" theory and that theory holds "water," will anyone take your theory seriously.

...that is, anyone who matters.

 

[zosima]

A typical sign of an internet troll is to always avoid the argument of one's opponents, by:

1. Ignoring them.
2. Ridiculing them.

Followed by restating one's position again as if it were self evidently correct.

Anyone can poke a hole in a dam. Very few can build a dam. Your argument is just another attempt to poke a hole in evolutionary theory. The theory remains standing, despite occasional holes, because those who know how to build a dam, work on the holes until they are filled.

So, poke away -- it will avail you not. Only when you actually build your own "dam" theory and that theory holds "water," will anyone take your theory seriously.

...that is, anyone who matters.

Well to Wogoga's credit, he has written a lot on a theory of reincarnation via "psychons" . In the language of your analogy it is a little pile of dirt which makes mud better than it holds water.