biological epistemology
review-essay by reilly jones
Bright Air, Brilliant Fire:
On the Matter of the Mind
by Gerald Edelman
BasicBooks, New York, 1992.
252 pages, ISBN 0-465-00764-3
©1995 Reilly Jones - All Rights Reserved
Published in Extropy #14
Vol. 7, No. 1, 1st Quarter 1995
On the Matter of the Mind
by Gerald Edelman
BasicBooks, New York, 1992.
252 pages, ISBN 0-465-00764-3
©1995 Reilly Jones - All Rights Reserved
Published in Extropy #14
Vol. 7, No. 1, 1st Quarter 1995
This is a book about the coming neuroscientific revolution in which we will achieve a scientific understanding of what human nature really is, and not what philosophers of the past have reasoned it to be. Nobel laureate Gerald Edelman, of Scripps Research Institute, has produced a summary of his theory of consciousness intended for the non-scientific reader. The full theory is laid out in rigorous detail in a trilogy consisting of: Neural Darwinism: The Theory of Neuronal Group Selection (1987), Topobiology: An Introduction to Molecular Embryology (1988), and The Remembered Present: A Biological Theory of Consciousness (1989). He has added his own thoughts about the philosophical and social implications of this new view of humanity as well as submitting useful critiques of some of cognitive science’s conventional wisdom. Since these books were published, there has been mounting evidence substantially confirming his basic theories, as well as an array of books expanding on many of the facets of consciousness with strong parallels to his ideas.
The subject of the governing principles of human nature is the most important subject possible to us. How do we know our world? What is the purpose of rationality, imagination, emotions, intuition, spirituality, dreaming? How do these all work together? What is the secular basis of morality? What creates the sense of self? How do we categorize our perceptions? How do we remember? How do we learn? How do we decide to act, and then act? We’ve had philosophers tell us that we are political animals, sick animals, rational animals, feeling animals, social animals, power-seeking animals and lately, that we are machines. Erroneous theories of human nature lead inexorably to failed political systems. The most recent failures were those systems based on Locke’s rational animal, which produced the modern limited democracies committed to protect science in order to gain the benefits of technology, back when people still believed in purpose and progress. These systems have given way to Rousseau’s feeling animal, which produced today’s failed social welfare states committed to individual self-expression without moral responsibility, when people openly question the existence of purpose or progress.
Any new political system replacing the current failed systems must deal with human nature as revealed by the neuroscientific revolution. Edelman is in agreement with Ayn Rand’s assertion that, “It is with a new approach to epistemology that the rebirth of philosophy has to begin.” A biological epistemology is the prescription for more successful, meaning adaptive, philosophical and political worldviews. People are difficult. A person’s conscious life is boundedly rational, much of life takes place in diffused focus, emotional states where rules of logic and deduction are not entirely accessible.1 We cannot always escape superstitious beliefs, we do not always act in accordance with our purposes nor say what we are thinking even when we try to.2 This revolution’s most dramatic effects lie in our ability to biologically and mechanically enhance our consciousness, both individually and species-wide. The political ramifications of such changes are immense.
Edelman is strongest when explaining his scientific theory of consciousness and outlining some of the major implications this theory has for the methodology of brain science and the philosophy of mind. He is particularly strong when relying on his long research experience in somatic selectionism and adaptationism within immunology. He is weakest when he relies on work done primarily by others in the areas of language, the meaning of truth, emotional complexity and the teleology of consciousness. The book’s organization is difficult to follow, you have to read it in entirety to find all the relevant points. I believe this is due to the scope and complexity of the topic of consciousness, there is no good beginning or end to the topic, you have to view it as a whole. He makes cogent arguments against mainstream positions in cognitive science: the mind is not a computer, humans are not “intelligent machines,” the brain is not a “finite state machine,” no connectionist neural nets work like the brain, psychofunctionalism can never be coherent without a thorough understanding of the biological selectionism and morphology of the brain, no language acquisition device exists, no specified semantics exist, no formal grammar can be specified, no objectivism (classical categories or unequivocal descriptions of reality) fits with how we actually categorize experience, no “mentalese” exists to causally determine behavior.
Edelman begins with the brain as a self-organizing and selectional system. I believe the marriage of self-organizational and selectional systems within thermodynamically open environments is the primary direction biological and social sciences are going to take for this scientific revolution (see review of Kauffman’s Origins of Order in Extropy #13).3 It is hoped that the fruit of this marriage will be universal laws of living forms and functions for any open-ended environment. Edelman calls it “completing Darwin’s program,” the study of morphologic evolution. New scientific disciplines are speciating in this vibrant research environment: recognition science (adaptation to novelty), selectional systems (mapping of biological and conceptual fitness space) , applied molecular evolution (mapping of polymer shape and function space), noetics (artificial sapience and inorganic neuromorphology), complexity sciences (self-constructing far-from-equilibrium dynamical systems), artificial life (synthetic biology), distributed metabolic systems (immunological, genomic, sociotechnological), and more.
Edelman starts his biological theory of consciousness with a physics and evolution assumption. No laws of physics are violated, there are no ghosts and no consciousness existed prior to its evolutionary appearance as a phenotypic property. I would add that no spook physics or spook biology are needed: no holographic mind, no consciousness collapses the wave-function, no Many-Worlds ontology, no computational microtubules in the cellular cytoskeleton, etc. He proceeds with a discussion of how consciousness evolved, how neurons work, what topobiology is, and what selectional systems are (he terms them “recognition science”). He presents his Theory of Neuronal Group Selection, explaining homeostatic values, neuronal maps and memory. Then, the old distinctions of reptile brain, paleo-mammalian brain and neocortex are recast by Edelman into Primary Consciousness (the older brains) and Higher-Order Consciousness (the newer brain). Each of these are made up of a primary neuronal repertoire (self-organization processes in embryonic development or “nature”) and a secondary neuronal repertoire (experiential development or “nurture”). Distributed processing of self-generated and non-self-generated signals occurs throughout this four-way anatomical matrix: primary, higher-order, nature and nurture.4
Consciousness evolved in stages: first, systems of the interior to take care of the body; second, systems to categorize world events and to permit sophisticated motor behavior; and third, systems to handle time and space, succession in motion and memory.5 This last stage has led to a more complex selectional system than simple natural selection operating at the previous stages. Elliott Sober notes, “The mind is more than a device for generating the behaviors that biological selection has favored. It is the basis of a selection process of its own, defined by its own measures of fitness and heritability.”6 Sober agrees with Edelman on the contingent nature of which selectional system will govern in any given instance. As Edelman puts it: “Given the diversity of the repertoires of the brain, it is extremely unlikely that any two selective events, even apparently identical ones, would have identical consequences. These observations argue that, for systems that categorize in the manner that brains do, there is macroscopic indeterminacy.” An important new holistic view of the evolution of genetic intelligence is being formulated, outlining the intimate connection between self and non-self, how organism’s perception of the environment leads to changes in the environment through selectional systems, and how increased perceptual accuracy hastens the pace of evolution.7
Topobiology is the study of the mechanical events occurring at particular places and temporal sequences during cellular development. Cells divide, migrate, adhere, differentiate and die. Edelman explains how these processes lead to individual uniqueness, even in genetically identical twins. “...Genes specifying the shapes of proteins are not enough; individual cells, moving and dying in unpredictable ways, are the real driving forces. The principles governing these changes are epigenetic -; meaning that key events occur only if certain previous events have taken place.” Unfortunately, researcher Rae Nishi reports that, “Very little is known about the mechanisms of neuronal cell death or of the mechanisms of cell rescue by trophic factors.”8
Edelman uses the term “recognition sciences” to mean the study of selectional systems with a particular definition of ’recognition.’ “By ‘recognition,’ I mean the continual adaptive matching or fitting of elements in one physical domain to novelty occurring in elements of another, more or less independent physical domain, a matching that occurs without prior instruction.” Natural selection (science of evolution) has given rise to two somatic selectional systems: the science of immunity and brain science. Somatic, in this sense, means occurring within the life of an individual. This sense of recognition and selection are representative of Stuart Kauffman’s synthetic biological models of “knower-and-known” (or subject-object relation) and “map-and-interpretation” (or approximation and evaluation of truth). Both Edelman and Kauffman emphasize that recognition systems can only survive with a sufficient degree of stability, poised at the boundary of chaos.
This stability is crucial to our search for truth. Edelman points to science as being “studies of stable relations among things,” mathematics as being studies “of stable relations among stable mental objects,” and logic as being studies “of stable relations between sentences that are applicable to things and to mental objects.” Robert Nozick characterizes this search: “Enhancement of inclusive fitness yields selection for approximate truth rather than strict truth. Knowing this, we can sharpen our goal and its procedures.”9 There are mathematical and computer models available now, dealing with how the continuous flow of external reality becomes mapped or compressed internally, simultaneously with interpretation or evaluation of such chunks of information.10 Our visual recognition system is unlimited in capacity and the selection of visual information occurs “early” in the course of processing (prior to recognition).11 The brain attempts to maximize perceptual accuracy by reducing the uncertainty in a variable input, thus gaining information about the precursor to the continuous stimulus distribution, while processing all perception through recategorical memory influenced by dynamically changing values.12 The importance of simultaneous map-and-interpretation is due to the critical problem in sensorimotor integration: the selection of single targets for movement from the continuous stimulus distribution.13 Friedrich Nietzsche showed great foresight when he wrote in The Will to Power: “We can comprehend only a world that we ourselves have made. It cannot be doubted that all sense perceptions are permeated with value judgments... The organic process constantly presupposes interpretations.”
A very brief summary of the Theory of Neuronal Group Selection (TNGS) in Edelman’s words:
The subject of the governing principles of human nature is the most important subject possible to us. How do we know our world? What is the purpose of rationality, imagination, emotions, intuition, spirituality, dreaming? How do these all work together? What is the secular basis of morality? What creates the sense of self? How do we categorize our perceptions? How do we remember? How do we learn? How do we decide to act, and then act? We’ve had philosophers tell us that we are political animals, sick animals, rational animals, feeling animals, social animals, power-seeking animals and lately, that we are machines. Erroneous theories of human nature lead inexorably to failed political systems. The most recent failures were those systems based on Locke’s rational animal, which produced the modern limited democracies committed to protect science in order to gain the benefits of technology, back when people still believed in purpose and progress. These systems have given way to Rousseau’s feeling animal, which produced today’s failed social welfare states committed to individual self-expression without moral responsibility, when people openly question the existence of purpose or progress.
Any new political system replacing the current failed systems must deal with human nature as revealed by the neuroscientific revolution. Edelman is in agreement with Ayn Rand’s assertion that, “It is with a new approach to epistemology that the rebirth of philosophy has to begin.” A biological epistemology is the prescription for more successful, meaning adaptive, philosophical and political worldviews. People are difficult. A person’s conscious life is boundedly rational, much of life takes place in diffused focus, emotional states where rules of logic and deduction are not entirely accessible.1 We cannot always escape superstitious beliefs, we do not always act in accordance with our purposes nor say what we are thinking even when we try to.2 This revolution’s most dramatic effects lie in our ability to biologically and mechanically enhance our consciousness, both individually and species-wide. The political ramifications of such changes are immense.
Edelman is strongest when explaining his scientific theory of consciousness and outlining some of the major implications this theory has for the methodology of brain science and the philosophy of mind. He is particularly strong when relying on his long research experience in somatic selectionism and adaptationism within immunology. He is weakest when he relies on work done primarily by others in the areas of language, the meaning of truth, emotional complexity and the teleology of consciousness. The book’s organization is difficult to follow, you have to read it in entirety to find all the relevant points. I believe this is due to the scope and complexity of the topic of consciousness, there is no good beginning or end to the topic, you have to view it as a whole. He makes cogent arguments against mainstream positions in cognitive science: the mind is not a computer, humans are not “intelligent machines,” the brain is not a “finite state machine,” no connectionist neural nets work like the brain, psychofunctionalism can never be coherent without a thorough understanding of the biological selectionism and morphology of the brain, no language acquisition device exists, no specified semantics exist, no formal grammar can be specified, no objectivism (classical categories or unequivocal descriptions of reality) fits with how we actually categorize experience, no “mentalese” exists to causally determine behavior.
Edelman begins with the brain as a self-organizing and selectional system. I believe the marriage of self-organizational and selectional systems within thermodynamically open environments is the primary direction biological and social sciences are going to take for this scientific revolution (see review of Kauffman’s Origins of Order in Extropy #13).3 It is hoped that the fruit of this marriage will be universal laws of living forms and functions for any open-ended environment. Edelman calls it “completing Darwin’s program,” the study of morphologic evolution. New scientific disciplines are speciating in this vibrant research environment: recognition science (adaptation to novelty), selectional systems (mapping of biological and conceptual fitness space) , applied molecular evolution (mapping of polymer shape and function space), noetics (artificial sapience and inorganic neuromorphology), complexity sciences (self-constructing far-from-equilibrium dynamical systems), artificial life (synthetic biology), distributed metabolic systems (immunological, genomic, sociotechnological), and more.
Edelman starts his biological theory of consciousness with a physics and evolution assumption. No laws of physics are violated, there are no ghosts and no consciousness existed prior to its evolutionary appearance as a phenotypic property. I would add that no spook physics or spook biology are needed: no holographic mind, no consciousness collapses the wave-function, no Many-Worlds ontology, no computational microtubules in the cellular cytoskeleton, etc. He proceeds with a discussion of how consciousness evolved, how neurons work, what topobiology is, and what selectional systems are (he terms them “recognition science”). He presents his Theory of Neuronal Group Selection, explaining homeostatic values, neuronal maps and memory. Then, the old distinctions of reptile brain, paleo-mammalian brain and neocortex are recast by Edelman into Primary Consciousness (the older brains) and Higher-Order Consciousness (the newer brain). Each of these are made up of a primary neuronal repertoire (self-organization processes in embryonic development or “nature”) and a secondary neuronal repertoire (experiential development or “nurture”). Distributed processing of self-generated and non-self-generated signals occurs throughout this four-way anatomical matrix: primary, higher-order, nature and nurture.4
Consciousness evolved in stages: first, systems of the interior to take care of the body; second, systems to categorize world events and to permit sophisticated motor behavior; and third, systems to handle time and space, succession in motion and memory.5 This last stage has led to a more complex selectional system than simple natural selection operating at the previous stages. Elliott Sober notes, “The mind is more than a device for generating the behaviors that biological selection has favored. It is the basis of a selection process of its own, defined by its own measures of fitness and heritability.”6 Sober agrees with Edelman on the contingent nature of which selectional system will govern in any given instance. As Edelman puts it: “Given the diversity of the repertoires of the brain, it is extremely unlikely that any two selective events, even apparently identical ones, would have identical consequences. These observations argue that, for systems that categorize in the manner that brains do, there is macroscopic indeterminacy.” An important new holistic view of the evolution of genetic intelligence is being formulated, outlining the intimate connection between self and non-self, how organism’s perception of the environment leads to changes in the environment through selectional systems, and how increased perceptual accuracy hastens the pace of evolution.7
Topobiology is the study of the mechanical events occurring at particular places and temporal sequences during cellular development. Cells divide, migrate, adhere, differentiate and die. Edelman explains how these processes lead to individual uniqueness, even in genetically identical twins. “...Genes specifying the shapes of proteins are not enough; individual cells, moving and dying in unpredictable ways, are the real driving forces. The principles governing these changes are epigenetic -; meaning that key events occur only if certain previous events have taken place.” Unfortunately, researcher Rae Nishi reports that, “Very little is known about the mechanisms of neuronal cell death or of the mechanisms of cell rescue by trophic factors.”8
Edelman uses the term “recognition sciences” to mean the study of selectional systems with a particular definition of ’recognition.’ “By ‘recognition,’ I mean the continual adaptive matching or fitting of elements in one physical domain to novelty occurring in elements of another, more or less independent physical domain, a matching that occurs without prior instruction.” Natural selection (science of evolution) has given rise to two somatic selectional systems: the science of immunity and brain science. Somatic, in this sense, means occurring within the life of an individual. This sense of recognition and selection are representative of Stuart Kauffman’s synthetic biological models of “knower-and-known” (or subject-object relation) and “map-and-interpretation” (or approximation and evaluation of truth). Both Edelman and Kauffman emphasize that recognition systems can only survive with a sufficient degree of stability, poised at the boundary of chaos.
This stability is crucial to our search for truth. Edelman points to science as being “studies of stable relations among things,” mathematics as being studies “of stable relations among stable mental objects,” and logic as being studies “of stable relations between sentences that are applicable to things and to mental objects.” Robert Nozick characterizes this search: “Enhancement of inclusive fitness yields selection for approximate truth rather than strict truth. Knowing this, we can sharpen our goal and its procedures.”9 There are mathematical and computer models available now, dealing with how the continuous flow of external reality becomes mapped or compressed internally, simultaneously with interpretation or evaluation of such chunks of information.10 Our visual recognition system is unlimited in capacity and the selection of visual information occurs “early” in the course of processing (prior to recognition).11 The brain attempts to maximize perceptual accuracy by reducing the uncertainty in a variable input, thus gaining information about the precursor to the continuous stimulus distribution, while processing all perception through recategorical memory influenced by dynamically changing values.12 The importance of simultaneous map-and-interpretation is due to the critical problem in sensorimotor integration: the selection of single targets for movement from the continuous stimulus distribution.13 Friedrich Nietzsche showed great foresight when he wrote in The Will to Power: “We can comprehend only a world that we ourselves have made. It cannot be doubted that all sense perceptions are permeated with value judgments... The organic process constantly presupposes interpretations.”
A very brief summary of the Theory of Neuronal Group Selection (TNGS) in Edelman’s words:
(1) Developmental Selection: This entire process is a selectional one, involving populations of neurons engaged in topobiological competition. A population of variant groups of neurons in a given brain region, comprising neural networks arising by processes of somatic selection, is known as a primary repertoire. The genetic code does not provide a specific wiring diagram for this repertoire. Rather, it imposes a set of constraints on the selectional process.
(2) Experiential Selection: Selective strengthening or weakening of populations of synapses as a result of behavior leads to the formation of various circuits, a secondary repertoire of neuronal groups.
(3) Reentrant Mapping: The linking of maps occurs in time through parallel selection and the correlation of the maps’ neuronal groups, which independently and disjunctively receive inputs. This process provides a basis for perceptual categorization. This is perhaps the most important of all the proposals of the theory, for it underlies how the brain areas that emerge in evolution coordinate with each other to yield new functions. To carry out such functions, primary and secondary repertoires must form maps. These maps are connected by massively parallel and reciprocal connections. A fundamental premise of the TNGS is that the selective coordination of the complex patterns of interconnection between neuronal groups by reentry is the basis of behavior.
The coordination of brain areas, or modularity, is one of the most promising paths to future enhancements of consciousness.14 Reductionist research on shifting synaptic strengths is showing positive results in confirming much of the selection and map formation theories. Some difficulty arises from the fact that diffusible substances (or “microhormones”) such as nitric oxide and at least three possible others, trigger groups of neurons probabilistically.15 The individual synapse cannot be the computer bit of the brain, rather, local groups of neurons behave like buffered attractors. Major difficulty arises when researchers try to follow the dynamic field receptivity of groups of neurons. Pettit and Schwark report:
...It is difficult to detect reorganization in subcortical maps, which are three-dimensional and can exhibit large somatotopic shifts over relatively small distances. ...It appears that mechanisms underlying receptive field reorganization exist at multiple levels of sensory systems. In normal brain function, such mechanisms might help to adapt to a continuously changing sensory environment.16
A serious theoretical model for why neuronal groups have evolved involves both linear and non-linear synaptic conductance. It is called the normalization model, and shows why group behavior allows for more fit responses to continuously variable stimulus distributions.17
There is an on-going “units of selection” controversy in biology over group vs. individual selection that is similar to the controversy in social sciences over methodological holism vs. individualism. After absorbing Kauffman’s ideas on biological attractors, I can’t help but think that Edelman should call it, Theory of Neuronal Attractors. Research is revealing that neurons behave strongly like attractors.18 The environment is open-ended, each individual unit is subject to a unique set of constraints at each instant. Thus, all selection is localized at the individual unit instant by instant, perhaps around attractors acting on varying spatio-temporal scales. I think it is misleading to speak of group selection in open systems.
The driving forces of an animal’s behavior are evolutionarily selected value patterns that helps their brain and body maintain conditions fulfilling the purposes of survival and reproduction. These systems are homeostats, and include regulation of heartbeat, breathing, sexual responses, feeding responses, endocrine functions and autonomic responses. Homeostasis is the buffered capacity of a system to return after being disturbed. Ralph Waldo Emerson, in his essay Experience, addresses this capacity for constancy: “If I have described life as a flux of moods, I must now add that there is that in us which changes not and which ranks all sensations and states of mind.” These homeostatic values are the base of morality, built-in values adhering around purpose. Miguel de Unamuno, in The Tragic Sense of Life, expressed this secular morality as, “Our desire is to make ourselves eternal, to persist, and whatever conspires to this end we call good, and evil is whatever tends to lessen or annihilate our consciousness.”
The concept of cognitive maps has a long history. The psychologist Edward Tolman viewed organisms as intrinsically goal directed, and as forming “cognitive maps” of their environments, in his 1932 book Purposive Behavior in Animals and Men. The philosopher Gilbert Ryle argued that the individual must “map” various mental concepts and determine their position in relation to other concepts, in his 1949 book The Concept of Mind. Elliott Sober draws attention to the analogy between biological models and maps in general; and argues that even though we don’t understand why probability works (without circularity), it is useful to use probability to make significant generalizations, or maps. I would note that the interpretation of probability (the map), is dependent on what your purpose is, or utility value. This freedom to interpret probability based on utility (an adaptive mechanism), is seen in various interpretations of quantum mechanics, Darwinism (population thinking), classification of species, even political polling. Mark Twain was right, there are “lies, damn lies, and statistics.”
Edelman’s theory covers not just perceptual mapping, but other mapping processes and their inter-relationships along reentry circuits with each other. Global mapping is a dynamic structure containing multiple reentrant motor and sensory maps interacting with nonmapped neuronal regions. The body and brain work together to produce the system property of consciousness, the brain requires the body in order to think. There is also a mapping of types of maps, operating free from immediate sensory input, capable of activating or reconstructing portions of past global and perceptual mappings.
The mechanisms of maps, homeostatic values and selection are central to the system property of memory. Memory is critically related both to perceptual categorization and to learning. The mechanisms of memory transfer, from short-term to long-term, are receiving a great deal of scientific research effort, confirming parts of the TNGS.19 Edelman writes: “To have memory, one must be able to repeat a performance, to assert, to relate matters and categories to one’s own position in time and space. To do this, one must have a self, and a conscious self at that.” David Gelernter emphasizes that “Thinking is primarily, overwhelmingly remembering.”
The primary consciousness of lower animals is phenomenal experience arising “from the correlation by a conceptual memory of a set of ongoing perceptual categorizations.” It is a “remembered present.” Higher-order consciousness in humanity is the achievement of temporal extension through the ability to distinguish conceptual-symbolic models of the processes of primary consciousness from ongoing experience. “The remembered present is placed within a framework of past and future. Once a self is developed through social and linguistic interactions on a base of primary consciousness, a world is developed that requires naming and intending.” Concept formation in the TNGS is linked to subjectivity, intentionality and volition.
Edelman attempts to move the science of consciousness as close to natural science and as far away from social science as possible. He does this through his “qualia assumption.”
There is an on-going “units of selection” controversy in biology over group vs. individual selection that is similar to the controversy in social sciences over methodological holism vs. individualism. After absorbing Kauffman’s ideas on biological attractors, I can’t help but think that Edelman should call it, Theory of Neuronal Attractors. Research is revealing that neurons behave strongly like attractors.18 The environment is open-ended, each individual unit is subject to a unique set of constraints at each instant. Thus, all selection is localized at the individual unit instant by instant, perhaps around attractors acting on varying spatio-temporal scales. I think it is misleading to speak of group selection in open systems.
The driving forces of an animal’s behavior are evolutionarily selected value patterns that helps their brain and body maintain conditions fulfilling the purposes of survival and reproduction. These systems are homeostats, and include regulation of heartbeat, breathing, sexual responses, feeding responses, endocrine functions and autonomic responses. Homeostasis is the buffered capacity of a system to return after being disturbed. Ralph Waldo Emerson, in his essay Experience, addresses this capacity for constancy: “If I have described life as a flux of moods, I must now add that there is that in us which changes not and which ranks all sensations and states of mind.” These homeostatic values are the base of morality, built-in values adhering around purpose. Miguel de Unamuno, in The Tragic Sense of Life, expressed this secular morality as, “Our desire is to make ourselves eternal, to persist, and whatever conspires to this end we call good, and evil is whatever tends to lessen or annihilate our consciousness.”
The concept of cognitive maps has a long history. The psychologist Edward Tolman viewed organisms as intrinsically goal directed, and as forming “cognitive maps” of their environments, in his 1932 book Purposive Behavior in Animals and Men. The philosopher Gilbert Ryle argued that the individual must “map” various mental concepts and determine their position in relation to other concepts, in his 1949 book The Concept of Mind. Elliott Sober draws attention to the analogy between biological models and maps in general; and argues that even though we don’t understand why probability works (without circularity), it is useful to use probability to make significant generalizations, or maps. I would note that the interpretation of probability (the map), is dependent on what your purpose is, or utility value. This freedom to interpret probability based on utility (an adaptive mechanism), is seen in various interpretations of quantum mechanics, Darwinism (population thinking), classification of species, even political polling. Mark Twain was right, there are “lies, damn lies, and statistics.”
Edelman’s theory covers not just perceptual mapping, but other mapping processes and their inter-relationships along reentry circuits with each other. Global mapping is a dynamic structure containing multiple reentrant motor and sensory maps interacting with nonmapped neuronal regions. The body and brain work together to produce the system property of consciousness, the brain requires the body in order to think. There is also a mapping of types of maps, operating free from immediate sensory input, capable of activating or reconstructing portions of past global and perceptual mappings.
The mechanisms of maps, homeostatic values and selection are central to the system property of memory. Memory is critically related both to perceptual categorization and to learning. The mechanisms of memory transfer, from short-term to long-term, are receiving a great deal of scientific research effort, confirming parts of the TNGS.19 Edelman writes: “To have memory, one must be able to repeat a performance, to assert, to relate matters and categories to one’s own position in time and space. To do this, one must have a self, and a conscious self at that.” David Gelernter emphasizes that “Thinking is primarily, overwhelmingly remembering.”
The primary consciousness of lower animals is phenomenal experience arising “from the correlation by a conceptual memory of a set of ongoing perceptual categorizations.” It is a “remembered present.” Higher-order consciousness in humanity is the achievement of temporal extension through the ability to distinguish conceptual-symbolic models of the processes of primary consciousness from ongoing experience. “The remembered present is placed within a framework of past and future. Once a self is developed through social and linguistic interactions on a base of primary consciousness, a world is developed that requires naming and intending.” Concept formation in the TNGS is linked to subjectivity, intentionality and volition.
Edelman attempts to move the science of consciousness as close to natural science and as far away from social science as possible. He does this through his “qualia assumption.”
Qualia, individual to each of us, are recategorizations by higher-order consciousness of value-laden perceptual relations in each sensory modality or their conceptual combinations with each other. Given the fact that qualia are experienced directly only by single individuals, our methodological difficulty becomes obvious. As a basis for a theory of consciousness, it is sensible to assume that, just as in ourselves, qualia exist in other conscious human beings, whether they are considered as scientific observers or as subjects. It is our ability to report and correlate while individually experiencing qualia that opens up the possibility of a scientific investigation of consciousness.
Thus, there is to be a marriage: between the natural science portion of the underlying physiology, that is to be studied through Francis Crick’s program of reductionism; and the social science portion of reports of mental states, that is to be studied through statistics.20 The difficulty we have of accurately reporting the complexity of our states and feelings means this research program is going to take considerable time because of the need for large sample sizes.
A crucial concept to understanding human nature is what Gelernter calls the “spectrum of focus” of consciousness. This is the idea that we live our lives along a sliding scale of thoughtful attention, from high-focus analytical reasoning, through medium-focus emotional states down to low-focus associative creativity. Edelman notes that the upper end of the range produces a thinker who “is so immersed in a specific attentive state related to the project of thought that he or she is truly ‘abstracted’- unaware of time, space, self, and perceptual experience.” At the low end of the range are dreaming and mystical experiences. This spectrum of focus has evolved largely because, as Edelman says: “The brain and the nervous systems cannot be considered in isolation from states of the world and social interactions. But such states, both environmental and social, are indeterminate and open-ended.”
Gelernter refers to his plunge-squish method from Mirror Worlds (see review in Extropy #11), to characterize reasoning; “High-focus thought is capable of penetrating a whole stack of memories at once.” This could be termed a mapping of maps, abstract rationality typifies this state. From Kauffman, “...Similar states typically flow to the same attractor and hence are classified as the same.” An excellent source for understanding high-focus categorization covers memory, induction, pattern completion, and causal reasoning.21 Estes gives formal accounts how instances within a category vary in their typicality and categorization accuracy; how categories that are defined by different rules are learned with different degrees of efficiency; how concepts are combined; and how categorization is influenced by variables such as category size, presentation order, frequency, and featural similarity.
Gelernter proposes affect-linking as a mechanism for continuity in medium-focus thought, “The role of emotion in thought, then, is exactly to glue low-focus thought-streams together.” Kauffman describes a mechanism of fluctuations in the distributed neuronic architecture that could stand for shifting emotional states, “...Minor alterations in network structure and logic can cause nearby states which formerly flowed to the same attractor to flow to two different attractors.” The emotional affect-linking has been characterized as “binding” or “chunking” allowing cognitive maps in the to-be-remembered environment to be processed simultaneously.22
Arthur Reber has discovered “experience, thought, and action can be influenced by past events that we cannot consciously remember (implicit memory) and current events that we cannot consciously perceive (implicit perception).”23 Implicit learning occurs because interpretations of cognitive maps that are not necessarily content-specific, may generalize the map to pick up patterns in the environment that the subject was unconscious of. These are low-focus processes. Gelernter calls emotion: “a ‘content-transcending’ abstraction. The vocabulary of the abstraction is completely separate from the vocabulary of the thing being abstracted.” A function of dreaming is to consolidate the affect-linking of emotions through single episodic memories. This linking could be what Kauffman describes as “...States along trajectories flowing to the same attractor converge on one another.” Rather than the mapping of maps of high-focus thought, this is more like cruising over maps picking up bits and pieces of memory. This low-focus area seems to be a likely candidate for what John Searle calls the Background. “A crucial step in understanding the Background is to see that one can be committed to the truth of a proposition without having any intentional state whatever with that proposition as content.”24
I will close by letting Edelman express his philosophical thoughts arising from his theory of consciousness.
A crucial concept to understanding human nature is what Gelernter calls the “spectrum of focus” of consciousness. This is the idea that we live our lives along a sliding scale of thoughtful attention, from high-focus analytical reasoning, through medium-focus emotional states down to low-focus associative creativity. Edelman notes that the upper end of the range produces a thinker who “is so immersed in a specific attentive state related to the project of thought that he or she is truly ‘abstracted’- unaware of time, space, self, and perceptual experience.” At the low end of the range are dreaming and mystical experiences. This spectrum of focus has evolved largely because, as Edelman says: “The brain and the nervous systems cannot be considered in isolation from states of the world and social interactions. But such states, both environmental and social, are indeterminate and open-ended.”
Gelernter refers to his plunge-squish method from Mirror Worlds (see review in Extropy #11), to characterize reasoning; “High-focus thought is capable of penetrating a whole stack of memories at once.” This could be termed a mapping of maps, abstract rationality typifies this state. From Kauffman, “...Similar states typically flow to the same attractor and hence are classified as the same.” An excellent source for understanding high-focus categorization covers memory, induction, pattern completion, and causal reasoning.21 Estes gives formal accounts how instances within a category vary in their typicality and categorization accuracy; how categories that are defined by different rules are learned with different degrees of efficiency; how concepts are combined; and how categorization is influenced by variables such as category size, presentation order, frequency, and featural similarity.
Gelernter proposes affect-linking as a mechanism for continuity in medium-focus thought, “The role of emotion in thought, then, is exactly to glue low-focus thought-streams together.” Kauffman describes a mechanism of fluctuations in the distributed neuronic architecture that could stand for shifting emotional states, “...Minor alterations in network structure and logic can cause nearby states which formerly flowed to the same attractor to flow to two different attractors.” The emotional affect-linking has been characterized as “binding” or “chunking” allowing cognitive maps in the to-be-remembered environment to be processed simultaneously.22
Arthur Reber has discovered “experience, thought, and action can be influenced by past events that we cannot consciously remember (implicit memory) and current events that we cannot consciously perceive (implicit perception).”23 Implicit learning occurs because interpretations of cognitive maps that are not necessarily content-specific, may generalize the map to pick up patterns in the environment that the subject was unconscious of. These are low-focus processes. Gelernter calls emotion: “a ‘content-transcending’ abstraction. The vocabulary of the abstraction is completely separate from the vocabulary of the thing being abstracted.” A function of dreaming is to consolidate the affect-linking of emotions through single episodic memories. This linking could be what Kauffman describes as “...States along trajectories flowing to the same attractor converge on one another.” Rather than the mapping of maps of high-focus thought, this is more like cruising over maps picking up bits and pieces of memory. This low-focus area seems to be a likely candidate for what John Searle calls the Background. “A crucial step in understanding the Background is to see that one can be committed to the truth of a proposition without having any intentional state whatever with that proposition as content.”24
I will close by letting Edelman express his philosophical thoughts arising from his theory of consciousness.
By taking the position of biologically based epistemology, we are in some sense realists and also sophisticated materialists. Given how meaning is defined in this book, we must accept a position of qualified realism. Our description of the world is qualified by the way in which our concepts arise. According to biologically based epistemology and qualified realism, knowledge must remain fragmentary and corrigible. We have suggested a favored set [of philosophical positions]: qualified realism, sophisticated materialism, selectionism, and Darwinism. Selfhood is of critical philosophical importance. Please remember, however, that no scientific theory of an individual self can be given (our qualia assumption). |
|
Notes:
1. Gelernter, D. The Muse in the Machine: Computerizing the Poetry of Human Thought. New York: The Free Press, 1994.
2. Subbotsky, E. Foundations of the Mind: Children’s Understanding of Reality. Cambridge, MA: Harvard Univ. Press, 1993.
3. A variety of references are listed for information on these systems:
Kauffman, S. The Origins of Order: Self-Organization and Selection in Evolution. New York: Oxford Univ. Press, 1993.
Harrison, L. Kinetic Theory of Living Pattern. New York: Cambridge Univ. Press, 1993.
Vallacher, R. & Nowak, A. eds. Dynamical Systems in Social Psychology. San Diego: Academic Press, 1994.
Gazzaniga, M. Nature's Mind: The Biological Roots of Thinking, Emotions, Sexuality, Language, and Intelligence. BasicBooks, 1992.
Charnov, E. Life History Invariant. New York: Oxford Univ. Press, 1993.
Dalva, M. & Katz, L. “Rearrangements of Synaptic Connections in Visual Cortex Revealed by Laser Photostimulation.” Science 8 July 1994: 255-9.
4. Research support is accumulating for distributed organization as opposed to a dedicated model.
Wu, J., Cohen, L. & Falk, C.X. “Neuronal Activity During Different Behaviors in Aplysia: A Distributed Organization?” Science 11 February 1994: 820-3.
Schwartz, A. “Direct Cortical Representation of Drawing.” Science 22 July 1994: 540-2.
Wilson, M. & McNaughton, B. “Reactivation of Hippocampal Ensemble Memories During Sleep.” Science 29 July 1994: 676-9.
5. Fischman, J. “New Clues Surface About the Making of the Mind.” Science 3 December 1993: 1517.
Barkow, J., Cosmides, L. & Tooby, J. eds. The Adapted Mind. Oxford Univ. Press, 1992.
6. Sober, E. Philosophy of Biology. Boulder, CO: Westview Press, Inc. 1993
7. Thaler, D. “The Evolution of Genetic Intelligence.” Science 8 April 1994: 224-5.
8. Nishi, R. “Neurotrophic Factors: Two Are Better Than One.” Science 19 August 1994: 1052-3.
9. Nozick, R. The Nature of Rationality. Princeton, NJ: Princeton Univ. Press, 1993.
10. See essays in Holland, J., Holyoak, K., Nisbett, R. & Thagard, P. Induction: Processes of Inference, Learning, and Discovery. MIT Press, 1986.
11. Van der Heijden, A.H.C. Selective Attention in Vision. New York: Routledge, 1992.
12. Norwich, K. Information, Sensation, and Perception. San Diego: Academic Press, 1993.
13. Salzman, C.D. & Newsome, W. “Neural Mechanisms for Forming a Perceptual Decision.” Science 8 April 1994: 231-7.
14. Zeki, S. A Vision of the Brain. Cambridge, MA: Blackwell Scientific, 1993.
Restak, R. The Modular Brain: How New Discoveries in Neuroscience Are Answering Age-Old Questions About Memory, Free Will, Consciousness, and Personal Identity. Scribners, 1994.
Kim, S.-G., Ugurbil, K. & Strick, P. “Activation of a Cerebellar Output Nucleus During Cognitive Processing.” Science 12 August 1994: 949-51.
15. See Barinaga, M. “Learning by Diffusion: Nitric Oxide May Spread Memories” Science 28 January 1994: 466, for mention of Edelman’s successful prediction of NO mechanism in 1990.
Also, see: Pascual-Leone, A., Grafman, J. & Hallett, M. “Modulation of Cortical Motor Output Maps During Development of Implicit and Explicit Knowledge.” Science 4 March 1994: 1287-9.
Jagadeesh, B., Wheat, H. & Ferster, D. “Linearity of Summation of Synaptic Potentials Underlying Direction Selectivity in Simple Cells of the Cat Visual Cortex.” Science 17 December 1993: 1901-4.
Murphy, T., Baraban, J., Wier, W.G. & Blatter, L. “Visualization of Quantal Synaptic Transmission by Dendritic Calcium Imaging.” Science 28 January 1994: 529-32.
Bolshakov, V. & Siegelbaum, S. “Postsynaptic Induction and Presynaptic Expression of Hippocampal Long-Term Depression.” Science 20 May 1994: 1148-52.
16. Pettit, M. & Schwark, H. “Receptive Field Reorganization in Dorsal Column Nuclei [DCN] During Temporary Denervation.” Science 24 December 1993: 2054-6.
17. Carandini, M. & Heeger, D. “Summation and Division by Neurons in Primate Visual Cortex.” Science 27 May 1994: 1333-6.
18. Turrigiano, G., Abbott, L. & Marder, E. “Activity-Dependent Changes in the Intrinsic Properties of Cultured Neurons.” Science 13 May 1994: 974-7.
19. Knowlton, B. & Squire, L. “The Learning of Categories: Parallel Brain Systems for Item Memory and Category Knowledge.” Science 10 December 1993: 1747-9.
O’Dell, T., Huang, P., Dawson, T., Dinerman, J., Snyder, S., Kandel, E. & Fishman, M. “Endothelial NOS and the Blockade of LTP by NOS Inhibitors in Mice Lacking Neuronal NOS.” Science 22 July 1994: 542-6.
Karni, A., Tanne, D., Rubenstein, B., Askenasy, J. & Sagi, D. “Dependence on REM Sleep of Overnight Improvement of a Perceptual Skill.” Science 29 July 1994: 679-82.
Nguyen, P., Abel, T. & Kandel, E. “Requirement of a Critical Period of Transcription for Induction of a Late Phase of LTP.” Science 19 August 1994: 1104-7.
20. Crick, F. The Astonishing Hypothesis: The Scientific Search for the Soul. New York: Scribner, 1994.
21. Estes, W. Classification and Cognition. New York: Oxford Univ. Press, 1994.
22. Cohen, N. & Eichenbaum, H. Memory, Amnesia, and the Hippocampal System. Cambridge, MA: MIT Press, 1994.
23. Reber, A. Implicit Learning and Tacit Knowledge. New York: Oxford Univ. Press, 1993.
24. Searle, J. The Rediscovery of the Mind. Cambridge, MA: MIT Press, 1992.
2. Subbotsky, E. Foundations of the Mind: Children’s Understanding of Reality. Cambridge, MA: Harvard Univ. Press, 1993.
3. A variety of references are listed for information on these systems:
Kauffman, S. The Origins of Order: Self-Organization and Selection in Evolution. New York: Oxford Univ. Press, 1993.
Harrison, L. Kinetic Theory of Living Pattern. New York: Cambridge Univ. Press, 1993.
Vallacher, R. & Nowak, A. eds. Dynamical Systems in Social Psychology. San Diego: Academic Press, 1994.
Gazzaniga, M. Nature's Mind: The Biological Roots of Thinking, Emotions, Sexuality, Language, and Intelligence. BasicBooks, 1992.
Charnov, E. Life History Invariant. New York: Oxford Univ. Press, 1993.
Dalva, M. & Katz, L. “Rearrangements of Synaptic Connections in Visual Cortex Revealed by Laser Photostimulation.” Science 8 July 1994: 255-9.
4. Research support is accumulating for distributed organization as opposed to a dedicated model.
Wu, J., Cohen, L. & Falk, C.X. “Neuronal Activity During Different Behaviors in Aplysia: A Distributed Organization?” Science 11 February 1994: 820-3.
Schwartz, A. “Direct Cortical Representation of Drawing.” Science 22 July 1994: 540-2.
Wilson, M. & McNaughton, B. “Reactivation of Hippocampal Ensemble Memories During Sleep.” Science 29 July 1994: 676-9.
5. Fischman, J. “New Clues Surface About the Making of the Mind.” Science 3 December 1993: 1517.
Barkow, J., Cosmides, L. & Tooby, J. eds. The Adapted Mind. Oxford Univ. Press, 1992.
6. Sober, E. Philosophy of Biology. Boulder, CO: Westview Press, Inc. 1993
7. Thaler, D. “The Evolution of Genetic Intelligence.” Science 8 April 1994: 224-5.
8. Nishi, R. “Neurotrophic Factors: Two Are Better Than One.” Science 19 August 1994: 1052-3.
9. Nozick, R. The Nature of Rationality. Princeton, NJ: Princeton Univ. Press, 1993.
10. See essays in Holland, J., Holyoak, K., Nisbett, R. & Thagard, P. Induction: Processes of Inference, Learning, and Discovery. MIT Press, 1986.
11. Van der Heijden, A.H.C. Selective Attention in Vision. New York: Routledge, 1992.
12. Norwich, K. Information, Sensation, and Perception. San Diego: Academic Press, 1993.
13. Salzman, C.D. & Newsome, W. “Neural Mechanisms for Forming a Perceptual Decision.” Science 8 April 1994: 231-7.
14. Zeki, S. A Vision of the Brain. Cambridge, MA: Blackwell Scientific, 1993.
Restak, R. The Modular Brain: How New Discoveries in Neuroscience Are Answering Age-Old Questions About Memory, Free Will, Consciousness, and Personal Identity. Scribners, 1994.
Kim, S.-G., Ugurbil, K. & Strick, P. “Activation of a Cerebellar Output Nucleus During Cognitive Processing.” Science 12 August 1994: 949-51.
15. See Barinaga, M. “Learning by Diffusion: Nitric Oxide May Spread Memories” Science 28 January 1994: 466, for mention of Edelman’s successful prediction of NO mechanism in 1990.
Also, see: Pascual-Leone, A., Grafman, J. & Hallett, M. “Modulation of Cortical Motor Output Maps During Development of Implicit and Explicit Knowledge.” Science 4 March 1994: 1287-9.
Jagadeesh, B., Wheat, H. & Ferster, D. “Linearity of Summation of Synaptic Potentials Underlying Direction Selectivity in Simple Cells of the Cat Visual Cortex.” Science 17 December 1993: 1901-4.
Murphy, T., Baraban, J., Wier, W.G. & Blatter, L. “Visualization of Quantal Synaptic Transmission by Dendritic Calcium Imaging.” Science 28 January 1994: 529-32.
Bolshakov, V. & Siegelbaum, S. “Postsynaptic Induction and Presynaptic Expression of Hippocampal Long-Term Depression.” Science 20 May 1994: 1148-52.
16. Pettit, M. & Schwark, H. “Receptive Field Reorganization in Dorsal Column Nuclei [DCN] During Temporary Denervation.” Science 24 December 1993: 2054-6.
17. Carandini, M. & Heeger, D. “Summation and Division by Neurons in Primate Visual Cortex.” Science 27 May 1994: 1333-6.
18. Turrigiano, G., Abbott, L. & Marder, E. “Activity-Dependent Changes in the Intrinsic Properties of Cultured Neurons.” Science 13 May 1994: 974-7.
19. Knowlton, B. & Squire, L. “The Learning of Categories: Parallel Brain Systems for Item Memory and Category Knowledge.” Science 10 December 1993: 1747-9.
O’Dell, T., Huang, P., Dawson, T., Dinerman, J., Snyder, S., Kandel, E. & Fishman, M. “Endothelial NOS and the Blockade of LTP by NOS Inhibitors in Mice Lacking Neuronal NOS.” Science 22 July 1994: 542-6.
Karni, A., Tanne, D., Rubenstein, B., Askenasy, J. & Sagi, D. “Dependence on REM Sleep of Overnight Improvement of a Perceptual Skill.” Science 29 July 1994: 679-82.
Nguyen, P., Abel, T. & Kandel, E. “Requirement of a Critical Period of Transcription for Induction of a Late Phase of LTP.” Science 19 August 1994: 1104-7.
20. Crick, F. The Astonishing Hypothesis: The Scientific Search for the Soul. New York: Scribner, 1994.
21. Estes, W. Classification and Cognition. New York: Oxford Univ. Press, 1994.
22. Cohen, N. & Eichenbaum, H. Memory, Amnesia, and the Hippocampal System. Cambridge, MA: MIT Press, 1994.
23. Reber, A. Implicit Learning and Tacit Knowledge. New York: Oxford Univ. Press, 1993.
24. Searle, J. The Rediscovery of the Mind. Cambridge, MA: MIT Press, 1992.