|
Author
|
Topic: Biological Information
|
warren_bergerson
Member
Member # 262
|
posted 21. November 2002 11:14
‘Information’ is an abstract mathematical concept which has obvious applications in the analysis of design and design processes. I am introducing this as a topic for discussion primarily because it illustrates the uses, and widespread misuses, of abstract mathematics in scientific analysis.
THE PROPER USE OF ABSTRACT MATHEMATICS
Abstract mathematics is very useful in evaluating ‘what if’ questions. The mathematical concept of information is useful in evaluating questions of the form "What are the implications if information in biological systems is defined as xxxx?’
In order to answer these questions, we start by identifying the key criteria which must be satisfied by biological information. Based on the key criteria we develop (propose) a formal mathematical definition of a new abstract concept called ‘biological information’. Starting with the proposed definition, we can then formulate and analyze questions such as ‘what logical mathematical processes produce changes in biological information?"
Such mathematical analysis produces results ‘in the context of the definitions used’. The definitions used can always be modified and refined if such modifications appear justified.
THE MISUSE OF ABSTRACT MATHEMATICS
Misuses of abstract mathematical analysis can and does take many different forms. Such abuses generally start with a failure to develop proper definitions and/or arbitrary insistence that definitions ‘must’ take a specific form. A second widespread type of misuse is to generalize analysis based on one definition of a concept of information to other very different definitions of information. A third type of misuse involves the failure to recognize the distinction between abstract mathematical conclusions and real world scientific conclusions.
THE MATHEMATICAL CONCEPT OF INFORMATION
In mathematics, the concept of information is defined in terms of rarity, complexity and/or improbability. A 3 digit base 10 number has a complexity of N/Nf or 1000/1 because the number can have 1000 different forms(N) and any specific 3 digit numbers is a member of a subset with 1(Nf) member. Inversely, a specific 3 digit base 10 number has an improbability or rarity of Nf/N or 1/1000.
It is important to note that in mathematics, information is not a stand alone concept. Information exists or is defined only in the context of rules which make it possible to measure/define complexity and improbability. [I will be happy to hear of any alternative mathematically precise definitions of the concept of information.]
INFORMATION AND SEQUENCES
Probably the most common uses of the concept of information involve sequences. A sequence of symbols such as numbers or numbers and letters is defined as containing information. In such applications it is fairly easy to define and quantify the number of possible forms ‘N’. In simple symbolic representations of information, each symbol is unique so Nf has a value of 1.
The concept of information defined in terms of sequences of symbols can be generalized to sequences or strings of objects such as strings of on/off switches or stings of DNA.
In defining the information is a string of symbols it is often convenient to start with the assumption the Nf =1 or that each combination of symbols conveys a unique piece of information. In many instances, stings of symbols involve redundancy or inefficiencies. In such instances the assumption Nf = 1 can be said to define the potential information in the string. The ‘actual’ information contained in the string is based on some value of Nf greater than 1.
To illustrate the point, consider a book such as a dictionary. It is easy to define the volume of potential information in such a long string of symbols using the assumption than Nf =1. The volume of ‘actual’ information is obviously much less than the potential volume.
As well as illustrating the difference between actual and potential information. The dictionary example also illustrate the practical problems with defining and measuring volumes of actual information. In practice, the analysis involving the concept of information is based on rough, one-sided estimates of volumes of information.
INFORMATION AND CHANGE PROCESSES
The mathematical concept of information by itself is of little or no analytical value. Information is only ‘interesting’ in the context of logical/mathematical processes or operation that act on information. Such processes would include but are not necessarily limited to processes to preserve, transfer, transform, create, and apply information.
In general, the abstract mathematical interest in information is an interest in processes of the form F(It)=It+1. Where It and It+1 are ‘information at point in time t and t+1 and F is some process responsible for the transformation from It to It+1.
The above, provides a quick and dirty overview, IMO, of the basic mathematical concepts associated with the mathematics of information. Comments are welcome.
BIOLOGICAL INFORMATION- THE CRITERIA TO BE SATISFIED
In order to perform mathematical analysis, it is useful to define a concept of information that can be useful in analyzing life forms. This specialized concept of information will be labeled biological information. As mentioned earlier, the first step in developing such a definition is to identify the criteria the definition should satisfy. As a starting point I suggest the following criteria:
CRITERIA:
1. The definition of biological information should be applicable to the analysis of all types of design and design processes associated with biological systems.
2. The definition must be useful/practical in modeling biological information in life forms.
3. The definition must make to practical to develop models of change processes.
There are two key features of the criteria proposed. First, it is proposed that a single definition must be developed which can be used for a)human design processes, b)genetic design processes, c)the design associated with nervous systems and animal intelligence, and d)the processes which created genetic processes(life from non-life). Second, the definition developed must make it ‘practical’ to actually measure/estimate volumes of information.
INADEQUACY OF EXISTING DEFINITIONS
Evolutionary biologist tend to ‘insist’ that biological information must be defined in terms of DNA strings (and in some cases in terms of protein structures). The rather vocal insistence on DNA as THE form of biological information, often makes it easy to overlook the rather obvious inadequacies of this definition. It is, or should be, obvious that 1)there are design processes associated with biological systems other than genetic design, and 2)none of other forms of design are associated with any known form of physical string or sequence. Even with respect to genetic design, it is highly questionable whether the DNA definition of information has not led to useful/workable models of change processes.
Since we are discussing ID here, it may be useful to note that Dembski and others have attempted to define biological information in terms of the complexity or improbability of ‘generating’ biological information. More specifically, they have tried to define biological information in terms of the improbability of ‘generating genetic information using RM&NS processes’. There are two obvious flaws with this definition. First, the definition can not be used in analyzing human design processes. Second, the definition incorrectly implies that RM&NS processes are the only processes which both transform genetic information and meet the requirements for materialistic/deterministic processes.
THE DESIGN SCIENCE DEFINITION OF BIOLOGICAL INFORMATION
As I have stated elsewhere, design science (my version) defines biological information in terms of adaptive or teleological or functional or purposeful complexity. This is based on the perspective that biological systems involve complex causal processes which can be modeled or simulated by logic machine. Biological information, from this perspective is defined in terms of the teleological complexity of the ‘programs’ controlling the relationships between ‘causes and effects’ or ‘input and output’. N, in this approach, is defined in terms of the number of possible input-output algorithms, and Nf or Nft is defined as the number of these possible algorithms that ‘increase the likelihood of achieving the goal or purpose of survival. This approach/definition of biological information meets all the criteria listed above.
SUMMARY
Although this post is lengthy, it still only a brief overview of the abstract mathematical analysis of biological information. Even with a brief overview, however, it should be apparent that there are serious questions about many existing applications of mathematical information concepts.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 23. November 2002 09:51
The current dismal state of misunderstanding and misconceptions with respect to biological information can be blamed on a number of factors. The authoritative but unproductive insistence that ‘analysis of biological information should be limited to the analysis of DNA’ and "biological information should be characterized as a form of energy" certainly contribute to our current failure to identify, define, measure and analyze even elementary forms of biological information such as that associated with neurons. Aside from arbitrary dictates, one of the fundamental problems in the study of biological information is the failure to distinguish between what are labeled here as 1)‘process information’ and 2) ‘transfer information’.
The basic structure of information processing is sometimes expressed as "input - processing - output’. A more complete description of information processing, however, recognizes the interaction of an information processing devise with the environment. The basic structure of information processing is thus more accurately described as an complex cyclical process involving "input - internal processing - output - external processing - input- ….".
The interactive model recognizes that information processing is an ongoing process operating over elapsed time. At intervals of elapsed time, information is either being ‘processed’ or ‘stored and/or transported’. As should be obvious, most observations and measures of biological information are based on information in a static format that is being stored or transmitted. It is difficult to observe and measure ‘information’ during processing.
One of the fundamental principles of complex life forms is parallel or coordinated processing. Parallel processing can not function without effective methods of transferring information between processing units. The easiest form of biological information to observe and measure is the static transfer/storage form of information associated with the external transmission of information between processing units.
Transfer/storage forms of biological information include language, impulses between neurons, chemicals transmitted between cells in a multi-cellular organism, DNA strings, and active/inactive gene states. Transfer/storage form information is relatively easy to observe measure and quantify. However, the study of transfer information does not, by itself provide very useful insights into the more complex processing aspects of biological information.
The teleological complexity definition provided above is designed to be compatible with the processing of information in biological systems, not just with the transmission of information.
IP: Logged
|
|
Daniel Edington
Member
Member # 421
|
posted 23. November 2002 21:47
It seems that nothing shuts down a list topic sround here faster that questions.
The question is then: Why are ID proponants so afraid of questions?
[ 23. November 2002, 21:48: Message edited by: Daniel Edington ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 23. November 2002 22:39
Dan,
Your question, I believed related to increases in information or complexity in proteins. The question is meaningless you define the mathematical concepts such as biological information and the mathematical operations which transform biological information.
I intend to address the question of processes which increase biological information and biological question. To understand the answers to these questions, however, you must first recognize that ‘information’ is not a simple unit like ‘energy’. Information is more accurately as a multi-dimensional phenomena. Increases in information can take a number or different forms.
I am not avoiding your question. I am simply providing some step by step background that makes it possible to formulate and address the question in a mathematically rigorous manner.
It is generally recognized that understanding information and more specifically biological information is key to understanding life forms. It is also generally recognized that the current understanding of biological information or woefully inadequate. We don’t even understand information and information processing in a seemingly simple mechanism like a neuron.
What is not generally recognized, and what I am trying to address here, is that much of the problem in understanding biological information, is a lack of a precise mathematical definition of both the mathematical concept of information and the mathematical processes operating on the complex phenomena called information.
I am not trying to avoid the question of increases in biological information or biological complexity. I am simply providing background material so the question you raise can be formulated in an appropriate manner.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 25. November 2002 12:09
[The following comment are intended to introduce in general terms some of the ideas associated with solution spaces and the progressive falsify-replace approach to developing scientific theories. As discussed, solution spaces and the progressive falsify-replace approach are key elements of the design science approach to analyzing information processing in biological systems. Solution spaces and the progressive falsify-replace approach represent radical departures from current ‘peer review’ practices. ]
Design science (my version) is based on mathematical modeling of complex causation. With respect to biological information the mathematics used makes it possible to:
1. DEFINE BIOLOGICAL INFORMATION-
2. EXPRESS THE COMPLEX PROCESSES or algorithms operating on biological information(model biological information processing) and
3. DEFINE THE SOLUTION SPACES needed to implement the ‘progressive falsify-replace’ used in design science to develop scientific theories of biological information processing.
The progressive falsify-replace approach to theory construction and refinement is one of the defining characteristics of engineering standards science and one of the characteristics distinguishing engineering standards science from the more widely practiced peer review standards science. The progressive or systematic falsify-replace approach to theory construction and refinement is make possible by a mathematical phenomena called a solution space.
THE PROGRESSIVE FALSIFY REPLACE APPROACH TO DEVELOPING SCIENTIFIC THEORIES
As ‘traditionally’ interpreted, a formal predictive scientific theory is expressed as a single mathematical algorithm. This algorithm is ‘discovered’ by one scientist and then tested, validated, and possibly replaced by the actions of other scientists. In general, the traditional interpretation assumes that formulating predictive, mathematical algorithm theories is rather difficult and replacing a major mathematical theory is relatively uncommon.
In engineering standards science, a formal predictive scientific theory is expressed as a set of algorithms representing a subset or partition a solution space. Given a solution space, it is relatively easy to formulate a predictive mathematical theory which fits an existing body of data(as well as other requirements to make the theory testable and compatible with general principles of deterministic causation). It is also relatively easy and common to modify, refine and replace the current ‘theory’ to reflect the results of validation, testing, and new discoveries.
In simplistic terms, engineering standards science involves a gradual/progressive development/refinement of major scientific theories. Theories construction and refinement in engineering standards science is a process involving active contributions from large numbers of different scientists.
The mathematical nature of solution sets is such that no matter how much or what knowledge is accumulated, it is always mathematically/logically possible to formulate a theory from the solution set that is compatible with the accumulated knowledge. [ It may eventually become impractical to develop theories from a particular solution set, in which case it may be beneficial to redefine the solution set. ]
THE BENEFITS OF PROGRESSIVE FALSIFY REPLACE
The gradual and progressive approach to refining scientific theories is beneficial when the relationships being analyzed are complex and existing knowledge of the processes and mechanisms involved is incomplete. Exactly the situation that exists with respect to biological information processing.
Solution sets make it possible to formulate rigorous testable theories, covering a broad complex subject area. A general theory constructed with solution sets make it possible to test the logical consistency of various subsidiary theories and hypothesis formulated in a general subject area. Solution sets makes it practical to adhere to the ‘one failure falsifies standard’ of testing. [It is difficult to visualize the benefits of this approach without considering a specific example. Such an example for biological information processing will be presented in the near future. ]
BACKGROUND
It is reasonable to claim that solution spaces and the progressive falsify-replace approach represents a radical ‘new’ approach to scientific analysis and theory construction in the life sciences. These approaches and techniques are not, however, a new form of science. Engineering applications such a bridge construction provide the basic paradigm for progressive approach to theory development.
Complex engineering applications begin with general structure or format for modeling the phenomena being engineered (whether a bridge or a trip to the moon). This general modeling structure is roughly equivalent to what is being referred to here as a solution space. Over a period of time this general structure develops into sophisticated methodologies for designing and constructing complex and sophisticated objects. As applied here, this basic progressive approach is used to develop complex models or theories of complex biological machinery.
IMO, the historical/cultural split between theoretical and applied science resulted in theoretical science loosing some of the concepts and principles needed to analyze complex causal relationships. The approach proposed here is viewed as ‘a return to the fundamental principles of science’ rather than as a new approach to scientific analysis.
SUMMARY
As viewed here, the successful scientific analysis of biological information processing requires solutions to three key technical problems- 1)a precise workable definition of biological information, 2)a practical method of expressing the transformations associated with biological information processing and 3)a practical methodology for developing predictive mathematical theories of biological information processing. I have now introduced for discussion solutions to two of the three technical problems.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 27. November 2002 07:42
STRUCTURE
One final issue to address before considering how biological systems generate information. The issue of structure or physical structure address the issue of where the various types of biological information processing occur.
Traditional views might suggest that biological information processing occurs ‘in the mind’, in the brain, in the neuron, in the cell or in DNA. The mathematical analysis of closed systems shows us that closed systems can’t create information. The ability of biological information processing to ‘create designs’ suggests biological information processing creates information. [Strictly speaking, the definition of biological information incorporates the concept that the ‘volume’ of information can change in specific locations within a system. ] This in turn suggests that biological information processing is an open or multiple location phenomena.
In its simplest form, the assertion that biological information processing occurs in an open system means that biological information processing involves interaction between a biological unit and an external environment. At a minimum, interaction between a biological unit and an external environment involves the following four types of information processing:
1. INTERNAL- The processing occurring within the biological unit.
2. OUTPUT- The processing or transformation from internal information to environmental phenomena.
3. INPUT- The processing or transformation from environmental phenomena to internal information.
4. ENVIRONMENTAL- The processing or transformations by which output impacts future input.
The single cell organism fits this simple four component structure. When you consider biological information processing in multi-cellular organisms, you transmissions between units and you have to consider the processes and mechanisms that make it possible for multiple processing units to act as coordinated units. The discussion here will not consider multiple biological units, but it is important to note that information processing in all multi-cellular organisms involves complex multiple unit structures.
STRUCTURE AND ANALYSIS
As defined above, simple biological information processing involves four separate types of processes. It can be argued that the most ‘interesting’ part of biological information processing is internal processing. Unfortunately, internal processing is the part of biological information processing which is the most difficult to observe and analyze. There are even those who have argued that it is not possible to observe and directly analyze the processing going on ‘inside the biological black boxes’. For the discussion here, it should be recognized that 1)it is at least very difficult to directly observe internal processing, and 2)most analysis of internal processing is indirect analysis based on observing the input, output, and environmental processes.
COMPARISON TO THE GENETIC/EVOLUTIONARY STRUCTURE
It is important to note the difference between the ‘input-output’ structure of biological information processing and the mutate-select structure of biological information processing used in genetics and evolutionary theory. The mutate-select structure defines biological information processing as two physical operations- selection and mutation operating on a physical object- DNA. The genetic model assumes that biological design/information processing consists of 1)mutations operating to produce variations and 2)selection or differential survival determining which physical forms survive.
There is a widely held belief/dogma that the processes associated with the mutate-select genetic structure are the major ‘materialistic, deterministic’ processes 1)which generate biological information, 2)which increase biological complexity, and 3)which create biological design. In looking at information generating processes, it is interesting to compare the capacities of systems with 1)input-output structures and systems with 2)mutate-select structures. For the moment it is only necessary to note that the two structures are dramatically different.
SUMMARY
In design science, biological information processing is viewed and defined in terms of an ‘input-output structure’. This means information processing involves interactions between biological units and an external environment. This interaction iinvolves input and output operations.
The input-output structure of biological information processing is very different than the mutate-select structure of biological information processing implicit in genetic and evolutionary analysis.
IP: Logged
|
|
Daniel Edington
Member
Member # 421
|
posted 28. November 2002 07:56
quote:
I am not trying to avoid the question of increases in biological information or biological complexity. I am simply providing background material so the question you raise can be formulated in an appropriate manner.
OK, let me know when you think you have answered my questions.
IP: Logged
|
|
richfaussette
unregistered
|
posted 28. November 2002 10:41
SUMMARY
In design science, biological information processing is viewed and defined in terms of an ‘input-output structure’. This means information processing involves interactions between biological units and an external environment. This interaction iinvolves input and output operations.
The input-output structure of biological information processing is very different than the mutate-select structure of biological information processing implicit in genetic and evolutionary analysis.
+++++
The differences you suggest do not pose a difficulty for the evolutionary analysis of human behavior. Inputs and outputs are implicit when ecologists talk about systems in which human beings behave. I am thinking specifically of Paul Colinvaux's Fates of Nations (1980)in which he discusses niche theory, an aspect of human reproductive strategies as operating within an ecological system of resources and behaviors. He couples resource acquisition and competition with group behaviors. In one example, he demonstrates the advantage of large populations and ample resources (Punic Wars) as the reason some civilizations win in resource competition and some die, but he also couples his discussion of resource inputs and outputs with Roman disciplina (group discipline) to demonstrate that it is the resource inputs and the mobilization of the resource inputs that empowered the Romans. I believe the Blalock equation (I=RM) also couples resources and mobilization. It is also important to note that given the evolutionary study of religion (suggested by EO Wilson, furthered by Colinvaux, championed by Kevin MacDonald, supported by David Sloan Wilson) we no longer are restricted to external forces impacting differential reproductive success. Religion evolved to raise selection stresses within cohesive human groups and the personal religious experience itself serves to raise selection stresses upon a willing individual effectively internalizing the formerly external environmental pressure of selection stresses.
I don't see any utility in suggesting that any aspect of biological information processing is not consonant with input/output processing. All biological information processing occurs within an ecological framework which is essentially an analysis of input/output processing in ecosystems.
Incidentally, with the advent of reflective self consciousness and the ability of humans to manage their own selection stresses individually and in groups, the effects of mutations which had always served to produce variety is diminished as humanity itself became aware of selection and human groups began intensively managing their own selection stresses.
rich
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 29. November 2002 11:00
FORMAT
As discussed above, the basic structure of biological information processing is ‘input-internal processing-output- environmental processing……’. The portion of this structure that is said to generate information in the internal processing segment. In design science, this internal processing is modeled, simulated and/or explained by a solution space. The purpose here is to introduce some of the design science notation and terminology used to express or model the complex causal relationship involved in internal processing.
To begin, internal processing in a biological unit can be viewed as a mathematical function F of the general form F(S)=R. In order to make it more practical to express and analyze, design science separates F into two levels. The lower level involves functions of the general form fxyt(s)=r which describe ‘simple’ input-output operation x performed by biological unit x at time t. The second level is a set of transformation operations T(fxyt)=fxyt+1 which change the simple input-output relationships.
The mathematical format used in design science to express dynamic multi-level algorithms involved with biological information processing is the logic machine program or computer program. Such programs can accept multiple inputs of different forms, perform different types of processing and produce different types of output. The idea that ‘biological information processing’ can be modeled or simulated by mathematical logic machine programs is widely accepted in modeling and simulating nervous system processing and human intelligence. This approach does not appear to be as widely accepted in modeling ‘genetic information processing’.
The logic machines used to model/simulate biological information processing have three basic types of operations. For each type of operation, analysis of biological information processing involves a)determining how the process or operation is represented in mathematical logical notation, b)determining how the process or operation would be manifest in a computer simulation, and c)determining the physical operations responsible for the process or operation in an actual biological systems. The three basic types of operation involved
1. RECORD INPUT- The ‘logic machine program’ indicates the input or stimuli received from the environment as mathematical/logical symbols representing real world stimuli. Computer simulations and real world biological units have physical mechanisms called receptors.
2. GENERATE OUTPUT- The logic machine program indicates the output or responses to be generated into the environment as mathematical/logical symbols which represent or denote real world responses. Computer simulations and real world biological unites have physical mechanisms called effectors.
3. PROCESS INFORMATION- The logic machine program involves logical operations expressed in much the same way as that computer programs are expressed. The physical processes or mechanisms by which computers perform these operations are the mechanical features of computer hardware. The physical manifestation of processing in biological units can take many different forms.
The discussion here is focused primarily on the abstract mathematical definition of biological information and the mathematical operations by which biological information is processed. Translating from the operations of abstract logic machines to either artificial/man made manifestations or to information processing in real world biological systems involves complex technical issues. However, there exist relatively ‘standard’ techniques for resolving the technical issues.
[Analyzing or ‘reverse engineering’ complex machines and complex biological systems is relatively easy if you know either 1)the logical structure of the processes involved or 2)the physical mechanisms/processes involved. Reverse engineering is extremely difficult if you understand neither the logic nor the physical mechanisms. Design science offers a description of the logic underlying biological information processing. The claim/prediction is then made that it will be relatively easy to identify the physical processes responsible for the identified logical operations. The original ‘source’ of the proposed logic is a subject for a separate discussion.]
BASIC CONVENTIONS OF BIOLOGICAL INFORMATION PROCESSING PROGRAMS
As described earlier, one level of biological information processing involved execution of simple input-output or cause-effect or stimulus-response processing. The portion of the logic machine program representing this input-output processing would be expressed as follows:
INITIAL VALUES
1. Let S= s1,s2, …sn Note: This is the set of input or stimuli the program can recognize.
2. Let R= r1, r2,….,rm Note: This is the set of output or responses the program can generate.
3. Let F= (s1, rx1), (s2, rx2),….., (sn, rxn) where rx is a member of R Note: this is the input-output processing algorithm operating in the unit at a point in time.
OPERATION
1. LINE1A: READ ENVIRONMENTAL STIMULI and assign value from S to variable SC
2. IN F, find (SC, rSC)
3. Let RC = rSC
4. GENERATE OUTPUT RC
5. Go to LINE1A.
SUMMARY
The above is a brief overview of the conventions used in design science to express or model biological information processing. The conventions and notation outlined above are believed to be fairly easy to follow.
Daniel- I think we finally have the preliminaries out of the way. We can now define what it means, in design science terminology, to generate information, to increase biological complexity, and/or to generate creative ‘intelligent designs’. Once increases in information are defined, we can identify the set of materialistic/deterministic algorithms(the solution space) which are capable of producing the increases in information.
IP: Logged
|
|
richfaussette
unregistered
|
posted 29. November 2002 12:05
warren wrote:
SUMMARY
The above is a brief overview of the conventions used in design science to express or model biological information processing. The conventions and notation outlined above are believed to be fairly easy to follow.
=========
I was questioning the ability of design science to properly formulate questions about human behavior as you described and your response clearly indicates that you are operating WITHIN the very paradigm that is in question which you specifically label 'design science.'
How does one base a rationale for defending a paradigm (utilize paradigm dependent assertions)using terminology that requires you invoke the very paradigm that is in question? That is circular and not logically permissible.
rich
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 30. November 2002 10:54
DEFINING CHANGES IN THE COMPLEXITY
The ability of deterministic/materialistic systems to generate information, complexity, and/or intelligent design is one of the key issues in the design/evolution debate. Genetic/evolutionary theory suggests that biological complexity is generated by mutate-select type processes. Certain forms of ID suggest that mutate-select processes can not produce/explain the levels of complexity or intelligent design observed in biological systems.
Design science (my version), agrees that mutate-select processes can not account for the information generating ability, the complexity, or the ‘intelligent designs’ observed in life forms. Design science, however, goes beyond the standard ID conclusion and asserts that there exist identifiable, complex, materialistic/deterministic processes called ‘information generating machines 1) which are capable of explaining the information processing capacities of biological systems, 2)which are capable of explaining changes in the complexity of biological systems, and 3)which are capable of explaining and simulating the ability of biological systems to generate intelligent designs. [ Of more general interest, the class or set of information processing machines constitutes a solution set which can be used with engineering standards science to formulate progressively improving series of predictive scientific theories. ]
DEFINING INCREASES IN COMPLEXITY
Using the design science definition of biological information, the volume of information/complexity in a systems increases if 1)a causal relationship in the system moves from a non-teleological form to a teleological form, 2)the complexity of a teleological relationship increases, 3)the system generates a new teleological causal relationship, and/or 4)the system creates a new or ‘creative’ form of a teleological relationship.
DEFINING THE DESIGN SCIENCE PERSPECTIVE
Design science makes the assertions that:
1. Any and all increases in complexity, increases in information, or generation of intelligent design of any biological systems can be defined and quantified as some combination of the of the above four types of increase in information and/or complexity.
2. For any and all occurrences of increased information/complexity in a biological systems there exists an identifiable set of mathematical algorithms (a solution set) which can model or simulate the increase in information as a materialistic, deterministic, testable, scientifically analyzable, process.
[Note: These ‘assertions’ in effect define an approach or perspective. If, it is proposed, you view and analyze biological information processing in terms of information processing machines(a mathematical concept), then 1)you have the technical or mathematical ability to model and explain all aspects of biological information processing, and 2)you have a practical, useful, productive approach to performing scientific analysis of life forms. The approach or perspective proposed is not necessarily the only possible approach, but the effectiveness of the proposed perspective can be objectively compared to any alternative approach. ]
DEMONSTRATING MATHEMATICAL FEASIBILITY
One of the early steps in support of the design science approach to analyzing biological information processing is to demonstrate the existence of sets of mathematical algorithms (existence of solution sets) which can 1)generate all four types in information, complexity defined above, 2)satisfy the logical requirements for a deterministic materialistic process, and 3)satisfy the engineering standards science requirements for a ‘scientific’ model.
It may or may not be apparent, but it is relatively easy to define algorithms or programs that satisfy these criteria. In the next day or two I will present logic machine code that performs one of the information generating processes. If anyone has a preference which of the four types of information increasing processes they would like to see posted let me know.
Rich- I am not clear what you are refering to as impermissible circularity.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 01. December 2002 09:10
As discussed above, design science identifies four types of increases in biological information which can be modeled and/or simulated by information generating machines. These information generating capacities are compatible with 1)the requirements for materialistic and deterministic processes and 2)the requirements for engineering standards science theory construction.
As a starting point in demonstrating the ability of an information generating machines to generate information and satisfy scientific and deterministic requirements, consider the following logic machine process for performing one subtype of the first of the four identified types of increasing information (moving a causal relationship from a non-teleological to a teleological form). [I can provide code for an information generating machine that performs all the identified information generating processes.]
For the discussion here, consider a input-output relationship of the form (sk, rx) or ‘input sk causes output rx’. Consider the set of such input-output relationships where sk is constant and rx can be an member of the set R of responses r1, r2,…rn. The set of input-output relationships as defined as N possible member. Further assume that there is some subset Rf of R containing Nf members which ‘increase the likelihood of achieving some goal G. The following logic machine program component will transform a non-teleological causal relationship (sk, rx) where rx is a member of R but not a member of Rf.
FIND TELEOLOGICAL FORM
INITIALIZE
1. Let FC=(sk, rx)
2. Let K=0
PROCESS
3. LINEA: Let K=K+1
4. Let FN=(sk,rK) Note: this is a process to create variance
5. ASSIGN SELECTION VALUE OF FC TO GC Note: This is an input and/or interact with environment operation
6. ASSIGN SELECTION VALUE OF FN TO GN
7. If GN>GC, assign FN to FC Note: This is a selection operation
8. Go to LINEA
A number of features of this program should be noted. First, this, expressed in logic machine notation is Aristotle’s deterministic, variance-selection explanation of teleological causation. Second, the program defined above identifies a set or class(solution space) of possible deterministic models. Third, the solution space defined by the above program is not the same as Darwinian or neo-Darwinian models.
SUMMARY
It is useful to note teleological causation is a central feature of biological causation. It is also useful to note that despite the nonsensical prejudices of peer review science, teleological causation is a ‘standard’ form of causation compatible with both deterministic and scientific requirements.
Biological information and biological information processing are complex topics involving a number of mathematical and scientific issues. The design science approach to biological information processing provides a comprehensive and logically consistent set of solutions to both the mathematical and the scientific issues. No other current approach can make such claims.
Daniel,
In answer to your question of complexity and proteins. A protein is the product of an assembly process which is a set of input-output relationships. The creation or evolution of a new protein involves changes in assembly instructions. The design science approach to biological information processing provides techniques for measuring and analyzing this type of change process.
IP: Logged
|
|
Daniel Edington
Member
Member # 421
|
posted 01. December 2002 09:53
quote:
In answer to your question of complexity and proteins. A protein is the product of an assembly process which is a set of input-output relationships. The creation or evolution of a new protein involves changes in assembly instructions. The design science approach to biological information processing provides techniques for measuring and analyzing this type of change process.
You have already stated this before (worded differently of course) and I have stated that this is not the answer to the question that I asked.
Does cytochrome c as, a specific example of a protein, contain information? if so, how much?
or as a modification to the question: If I had a buffer solution, containing only cytochrome c. and say I wanted to determine if cytochrome c was designed. Could I do so based solely on the contents of what I had in the solution just described?
A rather good analogy of this would be the classic mousetrap. We can take each of the individual componants by themselves and ask if they show evidence of design (without reference to the complex assembly they where extracted from.)
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 01. December 2002 11:43
I have been reading Warren’s set of posts on his version of design science. A great deal of it is a formal description of a standard view of interactional systems - a stimulus is processed by the organism, the resulting response affects the environment, which again provides stimulus to the organism, and so on.
However, throughout all his definitional and formal descriptions one thing seems notably lacking: a description of what aspects of reality one would measure, and how those measurements would be taken, in order to apply all these formalisms.
For instance, Warren writes,
quote:
The set of input-output relationships as defined as N possible member. Further assume that there is some subset Rf of R containing Nf members which ‘increase the likelihood of achieving some goal G.
We need examples - we need something that show how these formalisms might be made concrete. For a particular situation (both Daniel and I have mentioned some), what is N (what does one measure and calculate to find the total number of members of a set of input-output relationships), what is G (what goal is being targeted), and, most importantly, how and what does one measure to find that subset of N (Nf) which corresponds to those input-outputs which move the organism towards the goal?
Warren has taken the position that establishing the theory of design science in a formal and well-defined way is a vital first step, and this is a defensible position.
But until the various definitions and formal processes are tied to some empirical data, it is incorrect to call this design science - it may be design philosophy, but it is not science until it describes something that can be examined and measured in the real world. We need specific examples, and a discussion of real-world considerations about the measurement issues involved, before this philosophy can be said to be moving towards science.
SUMMARY: Warren repeatedly says that he is offering a way to define and measure biological information. He has offered definitions, but no hint of how to measure the quantities he has defined.
A good exercise at this point would be to merely pick a real biological example and describe how to measure N and Nf.
[ 01. December 2002, 12:21: Message edited by: Evan ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 02. December 2002 12:24
Dan,
Quote: You have already stated this before (worded differently of course) and I have stated that this is not the answer to the question that I asked.
Mousetraps and proteins are products of biological design processes. Techniques have been developed for determining whether or not an object is the product of a biological design process. These crude indirect measures of design are estimates of the improbability of occurrence which can in turn be interpreted as an estimate of the complexity or information content of the object.
I offer the outline of a method of directly measuring the improbability of generating a protein. This would be a direct measure of the improbability of occurrence. Clearly a direct measure of improbability is far superior to a crude estimate. I answered your question.
Evan,
The discussion here has focused on 1)mathematical concepts of biological information and biological information processing which, it is claimed, 2)provide a powerful tool for analyzing biological information processing within 3) an ‘engineering standards science’ called design science.
I fully agree with need to demonstrate how the proposed mathematical concepts are used in design science analysis. However, in providing such demonstrations, it is important to recognize the differences between ‘engineering standards science’ and the more widely recognized ‘peer review standards science’.
Recognizing the differences in ‘type’ of science is essential, because starting with 1)the same mathematics, and 2)the same set of observed facts, engineering standards science produces conclusions which are very different from and which directly contradict the conclusions produced by peer review science.
Consider as a simple example the treatment of teleological causation. Most ‘peer review type’ science suggests that teleological causation, if it exists at all, is not a legitimate or appropriate subject for scientific analysis. Many ID theorists(most ID theorists, as viewed here, are practitioners of peer review science) appear to suggest that while teleological causation is a feature of biological systems, it is something different from standard deterministic/materialistic causation.
In design science, by contrast, teleological causation is 1)the central and defining concept in biological information, 2)a ‘standard’ deterministic and materialistic causal relationship, 3)which is the basis for formulating scientific theories, and 4)which is central to the scientific analysis of life forms.
What accounts for this dramatic difference between conventional peer review science and design science? In answering this question, we begin by recognizing that peer review science imposes artificial and arbitrary constraints on subject matter and methodologies. The conventional objection to the analysis of teleological causation, as a prime example, is an arbitrary and capricious restriction imposed. The ‘rationale’ for the restriction is that if you exclude teleological causation, then all science will ‘look more like physics’. Given the peer review taboo on the use of teleological causation, the mathematical characteristics of teleological causation have never been properly analyzed. The lack of analysis has led to a number of misconceptions or faulty conclusions. [Consider as interesting examples:
1. RM&NS processes can simulate and explain simple evolutionary changes.
2. Biological systems either can not generate information (extract information from the environment) using materialistic deterministic processes or can generate only very limited ‘quantities of information’.
Both of the above issues are generally viewed as issues of fact or science. More accurately both issues should be viewed as mathematics issues. It appears that both issues could be resolved by analyzing the mathematics of teleological causation.]
The type of demonstration you suggest, ‘demonstrate a specific example of how biological information would be quantified’, illustrates another basic difference between design science and most forms of peer review science. In design science, phenomena such as purpose, information, and causation are mathematical concepts which are useful in scientific analysis. In many conventional scientific approaches it is believed or assumed that causation and/or information are in some sense ‘real’. Using the conventional approach, it might be expected that ‘biological information’ is an absolute phenomena for which there is a single correct measurement. From this perspective, ‘bits’ of biological information might be viewed as some type of unit or particle like a unit or particle of energy.
Design science recognizes general rules for measuring and quantifying biological information. These are general rules which can be agreed upon by groups of scientists and which can be used in the analysis of many different forms of biological information. Design science, however, does not view ‘volume of biological information’ as a real world phenomena but rather as a useful convention imposed to make analysis easier. I believe Chris Langan refers to this design science approach as instrumentalism.
I will be happy to demonstrate some of the design science approaches to quantifying real world biological information. The demonstrations will not, however, be particularly enlightening unless you understand the design science concept of ‘volume of biological information’.
IP: Logged
|
|
|
Printer-friendly view of this topic