|
Author
|
Topic: Growth of complexity
|
Wade
Member
Member # 1713
|
posted 24. August 2005 06:48
I imagine this question has been asked, but I'm looking for references to the results.
Have the artificial-life people (or anyone else) ever put out some numerical scale that estimates the current total complexity/order/whatever of all life-on-earth, and shown that a compounded rate of growth sufficient to generate the current value over 5 billion years is within the range they can demonstrate in simulated worlds (Tiera,etc.?)
Or, more basic, is there an agreed on scale of complexity that shows whether, say, a termite is more or less complex than, say, a Volkswagon bus,
and by how much?
[ 24. August 2005, 07:07: Message edited by: Wade ]
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 24. August 2005 23:45
Since it's the IDists who purport to have the definitive measure of complexity -- Complex Specified Information, or Specified Complexity -- I was sort of waiting for them to get around to it.
RBH
IP: Logged
|
|
Christopher D. Beling
Member
Member # 723
|
posted 25. August 2005 23:39
The information content of proteins is a well established and quantitative aspect of Molecular Biology (See Hubert Yockey's books of 1992 and 2005). The quantification is undisputable (i.e. non-teleological) science - just dealing with the communication of requisite information from the genome to the proteome. It is not restricted to IDists - it is very clear in Yockey's last chapter 13 that he is not an IDist himself (although his remarks are rather cryptic and non-commital).
To give an example Yockey works out that the information content of the iso-1-cytochrome c protein (used in energy metabolism in all living cells) is 233bits (i.e. about 2.1bits of info / amino acid residue). To obtain this answer is quite involved. Indeed one has to take into account the large number of homologous proteins in this superfamily - using "Mutual Entropy" as the measure of information content for the protein.
In principle the information content of all the proteins in the proteome of a particular species could be calculated. For example if in the human genome there are 30,000 genes and each had say an information content of 250bits then the total information content of man is 7.5x10^6 bits. This may not seem particularly high - but we must remember that this is all specified information - i.e. specified to make man. [It is complex specified information - since according to Dembski the maximum number of bits we could conceivably assign to chance is 5x10^2 bits]
In principle one could proceed throughout the entire plant and animal world - catalogue all known protein families - compute the "mutual entropy" information content for each family - and thus find a number for the total complexity of life (in bits). This is a mammoth task but perhaps approximate estimates are on the horizon.
With regard to the Volkswagon bus: It should be quite possible to compute the information content - based on an analysis of the engineering instructions required to build the bus. I guess it is not the kind of thing engineers would like to spend their time doing, but it is non the less possible. Instructions would be of two sorts - information regarding sequencing of construction and information on the shape, size and materials of components. For example with regard to the size of a block of metal - its width, length and height need specifying. But their are tolerances on these - infinite accuracy is both impossible and unnecessary. The tolerances restrict the number of bits of information to a finite number. With regard to materials - different material alloys could be specified again with certain tolerances.
IP: Logged
|
|
yanniru
Member
Member # 1732
|
posted 30. August 2005 08:09
Christoper,
“30,000 genes still don't look like they have anywhere enough information to specify 10^15 synapses” This quote from Prof. Chris King on the Wedconsciouysness forum suggests to me that there most be an invisible companion medium to our physical world that would store the extra required information. Since you seem to know about information theory, could you comment on this possibility?
Richard
IP: Logged
|
|
Christopher D. Beling
Member
Member # 723
|
posted 30. August 2005 23:00
Hi Richard,
Distinguished cosmologist George Ellis has recently turned himself to this aspect of neuroscience - and based on the work of others: I refer you to this website
http://www.metanexus.net/metanexus_online/show_article.asp?8740
Basically, your supposition is correct. There is not enough information to get anywhere near producing the complexity of a brain. But Ellis actually uses this fact to tell us something about the plasticity of the mind. He points out that there is enough information in the genome to build the basic anatomy of the brain. The actual synaptic connections that are made in any individual, however, depend [i.e. if we can talk about the total the brain configuration] on the individuals experience - that reinforces or weakens connections (and George Ellis develops a neo-Darwinian mechanism here). But note that the incorporation of neo-Darwinian mechanism - even if true - is not an argument here against non-intelligent agents producing information (as in conventional biological neo-Darwinism). The human mind has inbuilt intelligence/consiousness that is an essential ingredient in the formation of the synaptic connections. A huge amount of information is continually coming into the mind -and is being processed. What do you think about Ellis's scheme? -- Christopher
[ 30. August 2005, 23:03: Message edited by: Christopher D. Beling ]
IP: Logged
|
|
Art
Member
Member # 179
|
posted 31. August 2005 08:27
quote:
Christoper,
“30,000 genes still don't look like they have anywhere enough information to specify 10^15 synapses” This quote from Prof. Chris King on the Wedconsciouysness forum suggests to me that there most be an invisible companion medium to our physical world that would store the extra required information.
I'm not Christopher, but...
This quote doesn't make much sense to me. If we ignore the likelihood that there is not expected to be a one-to-one correspondence between gene and synapse (why should't one set of, say, 10,000 neuron-specific genes enable the formation of countless synapses?), we can use a simple exercise to realize that 30,000 genes have in them possibilities far, far more numerous that 10^15 items.
Suppose that each synapse was defined genetically by a unique combination of genes - synapse 1=geneA, synapse 2=A+B, synapse 3=A+C, etc... It turns out that 50 such genes would suffice in their entirety to uniquely specify 10^15 synapses (in round numbers).
Factor in numerous biochemical and genetic nuances, and I would argue that 30,000 is plenty, far more than "enough" to account for the complicatedness of the human brain (or body, or any other biological feature of people).
[ 31. August 2005, 11:55: Message edited by: Art ]
IP: Logged
|
|
|
Printer-friendly view of this topic