Well, I always say, "better late than never." I prepared these remarks prior to reading the cogent, coherent contributions from Burton, Rick, and John. Having quickly scanned them, I can see that there will be a number of possible topics for discussion. Thanks to Rick in particular for a nice summary of the modular organization of brain and current conceptions of function. His presentation clearly shows the power and productivity of contemporary approaches to describing and understanding brain function. I have a very long way to go to begin to formulate a conception of nervous system organization and function that will be able to compete with that body of knowledge. Nevertheless, I have taken this opportunity to point to what seem to me to be areas of weakness in contemporary schemes.
1. Sexism - Neuralism - Dualism - Emergentism - Holism - Hemispherism - Hierarchism - etc. Whence cometh hierarchism? Could it just be a reflection of the contemporary, critical mind presuming a separation where none exists? Could it just be the result of buying into classical physical and neural theory too much? Would a ring magnet provide a useful heuristic? It has two poles only after a gap has been introduced, when the circle has been broken. Isn't a hierarchical approach a natural concomitant of our expulsion from the Garden of Eden, i.e., of our imagined separation from primordial wholeness due to our persistence in mistaking the map for the territory? (See below)
Michael Faraday inferred, according to Cantor et al (1991, 88), from effects shown above that "poles exist only when (and because of) the reactive medium introduced by the cutting the iron ring. For ordinary magnets this meant that he considered the 'outer medium as essential to the magnet' such evidence argued the necessity of admitting a physical role for space." And so it shall be, I believe, for understanding the brain.
2. I rather like, from a heuristic POV, Al's consideration of a randomly connected network of McCulloch-Pitts neurons, a network that exhibits all-or-none behavior (i.e., it proceeds to a maximum number of neurons firing at a maximum frequency) whenever a sufficient number of neurons achieve a certain probability of firing. Putting this together with the idea of filtering action by axon branch points, which I called attention to last week, suggests that the architecture of a neural network will shape its oscillatory frequency. And, when it or its various "limbs" (i.e., subassemblies) were oscillating at the preferred frequency, one might experience a percept or its subcomponents, as described in the stabilized retinal image experiments, to "jump into and out of perception." I suspect, however, that there are lots of other possibilities for interpreting these results, such a microscopic eye movements or even subconscious intentions to shift the gaze.
3. The issue of what, if anything, a "representation" might be is central to the debate about neural contributions to conscious awareness and experience. One thing seems clear to me, the idea of sensory inputs generating neuronal representations of the external world that exist as more-or-less static symbolic structures and are deciphered, read, or observed in the theater of the mind, is quite preposterous. It just ain't so: such a naive sensationalist doctrine is incorrect. The postmodern situation is nicely summarized by van Gelder & Port (1995, 3):
"The cognitive system is not a computer, it is a dynamical system. It is not the brain, inner and encapsulated; rather it is the whole system comprised of nervous system, body, and environment. The cognitive system is not a discrete sequential manipulator of static representational structures; rather, it is a structure of mutually and simultaneously influencing change The cognitive system does not interact with other aspects of the world by passing messages or commands, rather, it continuously coevolves with them."
Al is, of course, very well aware of the realm of dynamical systems, and his idea that mind emerges as a natural, but unexpected product of brains functioning in societies, is a variety of dynamical approach. However, I would say that Al doesn't quite go far enough, that he needs to become just a little more postmodern and let go of that artificial boundary between that "encapsulated" inside and separate outside. In short, Al needs to really include the world (universe) as part of the dynamic system that is the mind. As we all know, we each struggle against our own particular version of maya - illusion fostered by unconscious allegiance to preconceived concepts - and Al is no exception. I should, BTW, count myself lucky to be so little subject to maya as Al Scott. The figure is from Neumann's classic, The Origins and History of Consciousness (1954):
I have been struggling with the concept of a "neural representation" for a long time without much success. To say that an object or stimulus was "represented" in the brain has always seemed to me to be one of the most obscure and obscuring statements in all of neuroscience. Al's fine presentation on cell assemblies has brought the issue to a head for me and I was stimulated to compose the following. As with so much of my work, it is an exercise in imaginative thought or "play thinking."
3a. A unique approach to "representations" is via a process philosophy route that recognizes perception to be a bimodal process. According to this approach, coming to know an object perceptually begins with a phase of "perception in the mode of causal efficacy," a prehensive unification (concrescence) of the respective entities. This process of becoming an actual entity in union with its universe is mostly unconscious and, thus, unappreciated by the modern mind. Only after that initial phase does the later phase of perception occur, "perception in the mode of presentational immediacy." This phase of perception is sensory and conscious and dominates our awareness to the extent that we mistake it for the whole process (see Whitehead, 1925, 1929a; Griffin et al, 1993; see also Gibson, 1986; c.f. Ullman, 1980). In his seminal work, Science and the Modern World, Whitehead described the processes as follows:
"The entity of which we become aware in sense perception is the terminus of our act of perception...Perception is simply the cognition of prehensive unification; or more shortly, perception is cognition of prehension." (italics added)
For more on prehensions and process philosophy see other webpage entries.
3b. I shall return to the question of the nature of non-sensory, direct perception shortly. In the meantime and with the notion of perception as a bimodal process firmly in mind, let's consider a more specific example of a "representation" within the sphere of perception:
Let a neuronal assembly be said to represent an object if and only if there is a isotonic isomorphism between the two. This, I suggest - or something close to it - is in fact what is usually meant when the term, representation, is used, i.e., that there is a continuous mapping of some sort between subject and object via the body, i.e., we see something with our eyes, hear something with our ears, touch something with our fingers, lips, or nipples, etc. (The exact nature of the mapping is immaterial for my current purpose; isotonicity is convenient but not mandatory.) Now, there are three possibilities for isotonic isomorphisms between neural assemblies and objects of which we are aware. These are the basis for constitutive, virtual, or empty representations
A constitutive isotonic isomorphism, also known as a sensory perception, is one that is based on an immediately prior (or on-going) prehension, a phase of non-cognitive, non-sensory interaction that establishes a three way unity between the knower (self), the neural assemblies (to be) and the object (to be) of knowledge. It establishes the context of knowing, including its embodied character. This phase of direct perception, though unconscious, is the information-rich phase. This phase is, in fact, informationally awesome and potentially or actually unmanageable if it 'leaks over' into embodied conscious awareness. In contrast, the conscious content of a constitutive isotonic isomorphism is sensory receptor and nervous system dependent, articulated, and informationally sparse. Fine tuning the relationship between the non-sensory, direct component and the information poor, neurally mediated component is an ongoing process of learning & development. Nevertheless, perceptually dominated cognition generally involves abstractions that lead toward reality (less abstractness, see Whitehead, 1925, 170). Thus the success of science. Thus the applicability of mathematics to reality.
A virtual isotonic isomorphism, also known as a conception (or, when unwanted, a hallucination), is a two way prehension between a more-or-less well-trained cell assembly and the self, i.e., between the neural representation of the body and the subject. A conception is a memory of a constitutive isotonic isomorphism. Conceptually dominated cognition generally involves abstractions that lead away from reality (more abstractness, see Whitehead, 1925, 170).
This distinction between constitutive and virtual isomorphisms says a bit of what others have already said about neural assemblies as representations, but - more importantly - brings out of the shadows and places in proper perspective the almost forgotten, yet most important, component for formation of a constitutive representation: the phase of non-sensory prehension between subject and object. In the presence of an ambiguous figure, such as the Necker cube, the directional opposition of perceptual and conceptual domination of the cell assembly gives rise to the alternation of perspective. This scheme also seems compatible with the effects of removal in adulthood of early cataracts. Of course, visual input will be ineffective as a sensory modality for distinguishing objects because no cell assemblies/subassemblies have been trained to recognize objects in the visual fields. The whole developmental sequence will need to be done, but many important critical periods for neural reaction and connections will have been missed and that confounds the issue.
An empty isotonic isomorphism, also know as gibberish or mere verbiage: blah, blah, blah. Is this what dualists or neuralists are talking when they talk about perception without recognition of the absolute need for a phase of non-sensory, direct apprehension between subject and object?
3c. This leads to the question of the nature and mechanism of 'non-sensory' or 'direct' perception. I do not know exactly what it is....It is expressed, however, as a joyful, fully in-formed, albeit non-conscious unification. It is the expression of the "feeling" component of being and becoming, the component that is so starkly absent from the materialistic conceptions of reality and which lies at the heart of the process approach. My guess is that it is identical with or closely related to what David Bohm calls the implicate order, to that level of reality where non-locality is the rule and from which event/particles unfold into the explicate, manifest order. I also intuit that it is closely related to what those who have considered the so-called esoteric and mystical traditions call "the ultimate ground of being" (e.g., Huxley, The Perennial Philosophy). However, not being able to say exactly what it is makes it neither unreal nor unnecessary for understanding perception. In its absence, it seems to me that that "neural representations" are nothing more than vacuous abstractions that have been gifted with intelligence by the speaker. Needless to say, the ego dominated modern mind, like the moon whose fondest desire is for the light of the sun to be extinguished so that it (the moon) will be the brightest object in the sky, denies that aspect of self that participates with the ultimate ground of being. The Tao Te Ching admirably expresses the real, but non-capturable aspect of reality that lies at the heart of our ability to know the universe in verse I:
- "The Tao that can be told is not the eternal Tao.
- The name that can be named is not the eternal name.
- The nameless is the beginning of heaven and earth.
- The named is the mother of ten thousand things.
- Ever desireless, one can see the mystery.
- Ever desiring, one can see the manifestations.
- These two spring from the same source but differ in name;
- this appears as darkness.
- Darkness within darkness.
- The gate to all mystery."
4. Finally, whether neurons process information about an "external" world which is then processed for "internal" conscious awareness or whether, via their electromagnetic oscillations, they resonate us and our perceptual bodies into the world as embodied consciousnesses, there is the issue of the "binding problem." Modern neuroscience clearly shows that spatially separate groups of neurons in the brainstem, cerebral cortex, cerebellum, basal ganglia, etc. are electrically active during what, for any one of us, is a unitary perceptual event of self & world. How is it that the activity of those separate groups of neurons is perceived as a unity? Wolf Singer, a highly respected neuroscientist, outlines the importance of synchronous neural activity in the gamma range (40 - 60 Hz) from an apparently neuralist position as follows (note what seems to be a empty isomorphism, i.e., vacuous abstraction, "meta-representation in the brain," gifted with awareness):
"The hypothesis examined here is based on the following premises: phenomenal awareness necessitates and emerges from the formation of meta-representations in the brain; meta-representations have become possible through the evolution of higher order cortical areas that process the output of lower order areas in the same way as these process their sensory inputs; to account for the combinatorial flexibility required for perception, these meta-representations consist of dynamically associated assemblies of neurons rather than individual specialized cells; neurons are grouped into assemblies and their responses labeled as related by a binding mechanism based on transient and precise synchronization of their discharges; and the formation of such dynamically associated, synchronized assemblies requires activated brain states and is facilitated by attentional mechanisms."
Synchronous activity is also emphasized by Llinas & Pare (1996) who suggest that conscious awareness of self and world as a unitary event is due to "binding" produced by synchronous action potentials at gamma range frequencies (40 - 60 impulses/sec) in groups of CNS neurons. Gamma band activities may arise due to interaction of specific and non-specific thalamocortical pathways on the cell bodies and dendrites on the relevant neurons.
5. Following on from last week's comments, I suspect that the brain's 'computer chip' for an epoch of embodied conscious awareness with a particular content, i.e., a perception or a conception, is a set of synchronously oscillating neurons - including (especially) subsets of anatomically interconnected neurons in cerebral cortex oscillating at 40 - 60 Hz - that maintain that oscillation for some minimum amount of time. Perhaps study of the fMRI difference between similar perceptual and conceptual cognitive events could reveal something about the locus for perception in the mode of causal efficacy?
6. Sorry to profess and run, but I am completely overwhelmed and out of time...I expect that many of you wonder with what right I "operate so carelessly and primitively with ideas in such a problematic realm without making even the least effort to prove anything? My defense: all our thinking is of this nature of a free play with concepts; the justification for this play lies in the measure of survey over the experience of the senses which we are able to achieve with its aid." (Holton) Please take what you like and leave the rest....
7. Parting Whitehead (1929b) quotation:
"The art of reasoning consists in getting hold of the subject at the right end, of seizing on the general ideas which illuminate the whole, and of persistently marshalling all subsidiary facts round them. Nobody can be a good reasoner unless by constant practice he has realized the importance of getting hold of the big ideas and hanging on to them like grim death."
Geoffrey Cantor, David Gooding, and Frank A.J.L. James, Michael Faraday, New Jersey: Humanities Press, 1991.
James J. Gibson, The Ecological Approach to Visual Perception, Hillsdale, NJ: Lawrence Erlbaum Associates, 1986.
David Ray Griffin, John B. Cobb, Jr., Marcus P. Ford, Pete A.Y. Gunter, and Peter Ochs, Founders of Constructive Postmodern Philosophy: Peirce, James, Bergson, Whitehead, and Hartshorne, Albany, NY: State University of New York Press, 1993.
G. Holton, Constructing a theory: Einstein's model, Amer. Scholar (?)
Aldous Huxley, The Perennial Philosophy, New York: Harper & Row, 1944.
Lao Tsu, Tao Te Ching, Gia-Fu Feng & Jane English, Translators, New York: Vintage Books, 1972.
Erich Neumann, The Origins and History of Consciousness, Princeton, NJ: Princeton University Press, 1954.
Wolf Singer, Consciousness from a neurobiological perspective, In: Brain and Mind: Evolutionary Perspectives, M.S. Gazzaniga and J.S. Altman (Eds.), HFSP, Strasbourg, pgs. 72-88, 1998)
S. Ullman, Against direct perception, Behav. Brain Sci. 3:373, 1980.
Timothy van Gelder and Robert F. Port, It's about time: An overview of the dynamical approach to cognition, In: Mind as Motion: Explorations in the Dynamics of Cognition, Port & van Gelder, eds., Cambridge, MA: The MIT Press, pgs. 1-43, 1995.
Alfred North Whitehead, Science and the Modern World, New York: The Free Press, 1925.
Alfred North Whitehead, Process and Reality, New York: The Free Press, 1929a.
Alfred North Whitehead, The Aims of Education and Other Essays, New York: The Free Press, 1929b.