Holographic model of human memory and tridimensionality of the space of perception
Simon Berkovich
Department of Computer Science
The George Washington University
berkov@gwu.edu
1. A hypothesis that human memory operates on the principles of holography had been put forward by K. Pribram [1]. This hypothesis has been investigated, elaborated, and discussed in numerous publications [2,3,4,5]. As said in [5], without this idea “there had been a total absence of even the prospects of general theory of biological memory”.
The reason for the high regard of holography is that, apparently, it offers the only known method for distributed representation of information corresponding to the functionality of the brain. The two basic paradoxical facts about human memory are well interpreted by means of the properties of the holograms (see Fig. 1):
- Like a hologram, human memory, as observed in classical works of K. Lashley [6], does not keep information in a particular location; and in both cases, the entire information contents can be recovered from a piece of the whole structure.
- With holography it is possible to recognize partially matching patterns invariant to certain distortions; this provides the most important feature of human memory - associative access.
Fig. 1
The principle of holography – partial match recall
Light pulses from a laser are split into two beams: a reference beam going
straightly and a signal beam going through an object. The two beams meet at
the recording medium producing an interference pattern – a hologram. For
another object the same can be done with a reference beam at a different angle.
As the recording medium is irradiated by a signal beam from a part of an object
Simon Berkovich
Department of Computer Science
The George Washington University
berkov@gwu.edu
1. A hypothesis that human memory operates on the principles of holography had been put forward by K. Pribram [1]. This hypothesis has been investigated, elaborated, and discussed in numerous publications [2,3,4,5]. As said in [5], without this idea “there had been a total absence of even the prospects of general theory of biological memory”.
The reason for the high regard of holography is that, apparently, it offers the only known method for distributed representation of information corresponding to the functionality of the brain. The two basic paradoxical facts about human memory are well interpreted by means of the properties of the holograms (see Fig. 1):
- Like a hologram, human memory, as observed in classical works of K. Lashley [6], does not keep information in a particular location; and in both cases, the entire information contents can be recovered from a piece of the whole structure.
- With holography it is possible to recognize partially matching patterns invariant to certain distortions; this provides the most important feature of human memory - associative access.
Fig. 1
The principle of holography – partial match recall
Light pulses from a laser are split into two beams: a reference beam going
straightly and a signal beam going through an object. The two beams meet at
the recording medium producing an interference pattern – a hologram. For
another object the same can be done with a reference beam at a different angle.
As the recording medium is irradiated by a signal beam from a part of an object
the corresponding reference beam, feasibly with lesser intensity, is recreated
2. In holography, the dimensionality of informational structures is determined by the restrictions on the wave propagation patterns. There are two basic types of wave propagation (Fig. 2): (a) waves can keep the whole medium behind agitated, for example, this happens on water surfaces; or, (b) waves can move with a sharp front leaving the media behind quiet, for example, this is the case for electromagnetic waves. The latter situation occurs in accordance with the so-called Huygens’ principle when the propagating front is created by secondary waves.
It has been noticed in [7] that holography can be effectively implemented only with the second type of wave propagation because for the first type of wave propagation the region of waves interference is extensive. It turns out that in a strict sense Huygens' principle is efficacious only in a 3D space [8]. As a result, a holographic model of human memory can directly manipulate only with 2D structures. Perception of 3D objects is done through temporal sequences of the 2D cross-sections. As to the objects of higher dimensions, their effectual representation in the holographic memory is not possible [7]. This implies that for the developed model the interface between the memory and the brain has to operate with 2D segments.
Fig. 2
Fig. 3
3. To begin with, a holographic mechanism needs a source of coherent waves to implement the scheme in
Fig. 1. Apparently, in conventional science a source of waves for holography is not at hand neither for the brain, nor for the cosmos. Importantly, in the words of Karl Pribram, “holography does not like boundaries”.
The fascination with holography has been also extended towards the Universe where the properties of holography are applied to rationalize the weirdness of quantum mechanics. To explicate the key point of non-locality David Bohm introduced a concept, which he called “implicate” or “enfolded” order; the holography is attracted just as analogy showing that the whole world can be enfolded in each of its parts [4].
Processing in neuronal system coordinates activities in different brain areas. Vision system produces 2D patterns; other senses mainly produce 1D sequences signals. Access to holographic memory is done by 2D patterns, so to enable access to the memory by 1D inputs it is necessary to have serial-to-parallel conversion providing 2D patterns. As a result, inputs from various senses can be accommodated and cross searches between them become possible.
The best that neuroscientists can say for the moment is that human memory is a stored pattern of connections between neurons in the brain. Each neuron makes about 5,000 to 10,000 synaptic connections with other neurons. Such a structure is not an effective computer engineering apparatus, but now Yasser Roudi and Peter Latham at University College London [10] have found that even with 10,000 connections per neuron, a network could only store about 100 memories – regardless of how many neurons were in the network. Current explanation of the brain attracts the Long-Term Potentiation to strengthen synaptic connections for the task of memory. However, switching times in neuron processing are about 10-2 sec and covalent modifications of proteins for LTP last minutes [11]. These time scales are absolutely incompatible with the speed of brain operations. As to other performance characteristics, notably, reliability and fault-tolerance, the pattern of connections between neurons cannot be even thought of to match human memory. Hence, workings of human brain tend to be attributed to perplexing quantum world or problematic non-Turing computations (see, e.g. [12]).
Information processing functions associated with human brain constitute a problem of acute discomfort to thinkers for millennia. Said St. Augustine, V century: “The way in which minds are attached to bodies is beyond man’s understanding”, Blaise Pascal, XVII century: “Man is to himself the most wonderful object in nature; for he cannot conceive what the body is, still less what the mind is, and least of all how a body should be united to a mind”, Steven Weinberg, XX century: “How memories are stored in the brain is not likely to be affected by the discovery of the final theory”.
The fascination with holography has been also extended towards the Universe where the properties of holography are applied to rationalize the weirdness of quantum mechanics. To explicate the key point of non-locality David Bohm introduced a concept, which he called “implicate” or “enfolded” order; the holography is attracted just as analogy showing that the whole world can be enfolded in each of its parts [4].
Processing in neuronal system coordinates activities in different brain areas. Vision system produces 2D patterns; other senses mainly produce 1D sequences signals. Access to holographic memory is done by 2D patterns, so to enable access to the memory by 1D inputs it is necessary to have serial-to-parallel conversion providing 2D patterns. As a result, inputs from various senses can be accommodated and cross searches between them become possible.
The best that neuroscientists can say for the moment is that human memory is a stored pattern of connections between neurons in the brain. Each neuron makes about 5,000 to 10,000 synaptic connections with other neurons. Such a structure is not an effective computer engineering apparatus, but now Yasser Roudi and Peter Latham at University College London [10] have found that even with 10,000 connections per neuron, a network could only store about 100 memories – regardless of how many neurons were in the network. Current explanation of the brain attracts the Long-Term Potentiation to strengthen synaptic connections for the task of memory. However, switching times in neuron processing are about 10-2 sec and covalent modifications of proteins for LTP last minutes [11]. These time scales are absolutely incompatible with the speed of brain operations. As to other performance characteristics, notably, reliability and fault-tolerance, the pattern of connections between neurons cannot be even thought of to match human memory. Hence, workings of human brain tend to be attributed to perplexing quantum world or problematic non-Turing computations (see, e.g. [12]).
Information processing functions associated with human brain constitute a problem of acute discomfort to thinkers for millennia. Said St. Augustine, V century: “The way in which minds are attached to bodies is beyond man’s understanding”, Blaise Pascal, XVII century: “Man is to himself the most wonderful object in nature; for he cannot conceive what the body is, still less what the mind is, and least of all how a body should be united to a mind”, Steven Weinberg, XX century: “How memories are stored in the brain is not likely to be affected by the discovery of the final theory”.
The contemporaneous philosophical thought tacitly admits that the considered problem cannot be incorporated into the existing knowledge; see, e.g., an article in the encyclopedia [13]: “It may well be that the relation between mind and body is an ultimate, unique, and unanalyzable one. If so, philosophical wisdom would consist in giving up the attempt to understand the relation in terms of other more familiar ones and accepting it as the anomaly it is”.
We have suggested a paradigm analogous to the concept of “Cloud Computing” that gives
an elegant constructive solution to the problem of the organization of mind where individual brains are not stand-alone computers but collective users getting shared access to portions of the holographic memory of the Universe [14, 15, 16]. A critical consideration of such an idea can be found in [17]. With the existing worldview attaining the extraordinary performance of the brain is absolutely inconceivable.
We have suggested a paradigm analogous to the concept of “Cloud Computing” that gives
an elegant constructive solution to the problem of the organization of mind where individual brains are not stand-alone computers but collective users getting shared access to portions of the holographic memory of the Universe [14, 15, 16]. A critical consideration of such an idea can be found in [17]. With the existing worldview attaining the extraordinary performance of the brain is absolutely inconceivable.
References
1. K.H. Pribram, “Languages of the Brain”, Brandon House, London, 1981.
2. P.R. Westlake, “The Possibilities of Neural Holographic Processes within the
Brain”, Kybernetik, 7, No. 4, pp. 129-153, 1970
3. Ken Wilber (editor), “The Holographic Paradigm and Other Paradoxes”,
Shambhala Publications, Boulder & London, 1982
4. M. Talbot, “The Holographic Universe”, Harper Perennial, New York, 1991
5. P. Pietsch, “Shufflebrain. The Quest for the Hologramic Mind”, Houghton Mifflin
Company, Boston, 1981
6. K. S. Lashley, “In search of the engram”(1950), in the book: Neurocomputing
Foundations of Research, editors J. Anderson and E. Rosenfeld, pp. 57-63,
MIT Press, 1988
7. S. Ya.. Berkovich, “The Dimensionality of the Informational Structures in the
Space of Perception (Posing of Problem)”, Biophysics, 21, pp. 1136-1140, 1976
8. R. Courant, D. Hilbert, “Methods of Mathematical Physics, Vol. I”,
Interscience Publishers, New York, 1953
9. S. Berkovich, “Shaping the computational model with content-addressable memory”,
E-Newsletters for Science and Technology, Issue 1, Number 1, 2007, pp 1-3,
European Academy of Sciences,
http://myweb.polyu.edu.hk/~mmktlau/E_Publication%20(EAS)/EASEN_Issue_1.htm
10. “Brain connections cause rethink over human memory”, New Scientist,
issue 2621, page 23, 18 September 2007
11. Amit Etkin, Gleb Shumyatsky, Christofer J. Pittinger, and Eric R. Kandel,
“Genetic Analyses of Functional Connectivity in the Nervous System”, pp. 221- 240,
in “Databasing the Brain”, John Wiley & Sons, Inc, 2005
12. Roger Penrose, "Shadows of the Mind", Oxford University Press, New York, Oxford, 1994
13. Jerome Shaffer, “Mind-Body Problem”, The Encyclopedia of Philosophy, Vol. 5,
p. 345, Macmillan Publishing Co., New York, 1972
14. S. Y. Berkovich, “On the “barcode” functionality of the DNA, or The phenomenon
of Life in the physical Universe”, Dorrance Publishing Co, Pittsburgh, PA, 2003
(almost a full version of this book is at http://arxiv.org/abs/physics/0111093)
15. Simon Y. Berkovich and Hanan M. Al Shargi, “Constructive Approach to
Fundamental Science: Selected Writings”, University Readers, San Diego, 2010
16. Berkovich S. “On the information processing capabilities of the brain: shifting the
paradigm”, Nanobiology, 2, 1993, pp. 99-107
http://www.seas.gwu.edu/~berkov/Theory.htm
17. Donald R. Forsdyke, “Samuel Butler and human long term memory
Is the cupboard bare?”, Journal of Theoretical Biology, 258, 2009, pp. 156-164
http://post.queensu.ca/~forsdyke/mind01.htm
Welcome Simon! It is very nice indeed to have a distinguished computer scientist in our ranks, as well as Soviet expatriate, since Russian literature potentially relevant to our topic is wholly unknown to us.
ReplyDeleteIn your 1976 paper "The Dimensionality of the Informational Structures in the Space of Perception (Posing of Problem)" that appeared in the journal "Biofizika," the abstract refers to the "known hypothesis of Poincaré on the physiological determinacy of the property of three-dimensionality of the space of perception." Could you briefly tell us what Poincare's hypothesis is? (I, for one, am not familiar with it.)
Just rereading the relevant part of your 1976 paper I see Poincaré's criteria for defining dimensionality (as well as a higher dimensional embedding space), but not how he thought this was physiologically determined in the case of perceptual space.
ReplyDeleteIn his latest work - “Pourquoi l’espace a trois dimension?”, Dernieres Pensee, Flamarion, Paris, 1913 – H. Poincare put forward the idea that the tridimensionality of the space of perception is determined by the physiological specifics of the human brain.
ReplyDeleteI have considered the holographic model of human memory, and in elaboration of the Poincare approach, showed how it leads to the tridimensionality of the space of perception. This comes as consequence of Huygen’s principle – holographic mechanism can rely only on narrow wavefronts, and according to Huygen’s principle this can happen only in 3D space.
H. Poincare presents a recursive definition of dimensionality. A point has dimension 0.
A line can be divided by a point in separated parts, so a line has dimensionality 1.
A surface area can be divided by in separated parts by a line, so surface has dimensionality 2. A volume can be divided by a surface in separated parts, so volume has dimensionality 3. And so on.
At any given moment of time, a holographic mechanism can handle only a 2D slice.
In a certain time interval, this mechanism can retain a sequence of 2D slices, i.e. crossections of a 3D object. So, holographic mechanism can effectively process only objects of a dimension not greater than 3.
I also wish to welcome Simon to the group. One issue that I have with Poincare is that in Dernieres Pensees he also says that it takes "a little good will" to give three dimensions to our visual experiences. I find it to be noteworthy that even with the case of viewing a hologram we still don't perceive interiors, and that perceived regions can be bounded by edges, which are one dimensional (making the resulting experience two dimensional by Poincare's own criterion for dimensionality). I wonder then whether he isn't conflating the issues of not being flat (in metric concept of not possessing a Euclidean metric) with the issue of dimensionality (a topologic concept). I don't think that the issue of "good will" should confuse these issues.
ReplyDeleteI find Simon's proposal very intriguing, and always have since I first read it back in 1980. The only basic question I have, though, is that the theory seems to start as a theory of memory and memory retrieval but then talks of perception. So I would ask, Simon, how does the dimensionality of perceptual space depend upon the dimensionality of memory information?
ReplyDeleteThere are really several issues, but the overall question of whether perceptual space (especially visual space) must be of N dimensions in order to interface with a holographic-type memory system is, I feel, the main one.
ReplyDeleteCan Simon or Bob find the passage in Poincare about the question of physiological determinancy of perceptual space? It might be nice to have that quoted here (or part of it). If it is long, we should perhaps make it a separate posting.
Bill, unfortunately while I read this material long ago when I was initially researching these issues, I don't have a copy of Poincare's Dernieres Pensees. If Simon has a copy maybe he can post something from it though.
ReplyDeleteA belated welcome to Simon! I am very much in favour of new and stimulating hypotheses such as yours. I think our two approaches are compatible. I support the idea (as does Bernard) that phenomenal consciousness itself lies outside the brain i.e. in higher-dimensional space a la Brane theory: and it (or part of it) could do so in a holographic form. My theory at present focusses only on the 'inner core' of consciousness (i.e. the experienced phenomenal 'world' ) but there is no reason why hologram-based memory banks could not surround this (unexperienced—i.e. outside Plato's Cave—remember Jason Brown's cogent remark "No one asks what lies beyond the space of a dream.").
ReplyDeleteI was also intrigued with Berkovich's statement about "traveling wave solutions propagating in a helicoidal form". This reminded me of the hitherto unexplained helicoid Ehrenhaft phenomenon (the complex and beautiful helical patterns spun out by minute dust particles in a strong beam of light). These are illustrated in my book "The Walls of Plato's Cave."
I think we should think of VS--and perceptual space--as being structurally, if not informationally--isolated. Why that is so seems the big puzzle. But I would still like to know what Poincare says about a physiological basis to perceptual dimensionality. Can you supply the passage, Simon?
ReplyDelete'What lies beyond the space of a dream,' as Jason Brown asks, is a cogent question, because from an informational standpoint--content--it seems to share content of perception as well as (apparently) precognitive content *to be perceived* in the future, and thus eventually in VS. In the case of hypogogic/hypnopompic hallucinations there are dramatic "morphings" of dream into "reality" and the reverse, some of which I have experienced myself. I can recall experiences in which the dream visual field literally transformed into the normal visual field with my eyes open, but it was not like changing the channel on a TV at all. It really was very much like morphing for me because the transformation seemed to occur at the microlevel, as if the dream image found a way of changing into the "real" "normal" things I was then seeing--not at all like a "dissolve" in a movie either, in which there is an overlap between two images. Exactly how that process might occur may be something that Lothar Kleine-Horst may explain in relationship to what the Ganzheit psychologists called "microgenesis" of visual perception (German: Akutalgenese), the leading Austrian-American researcher of being Heinz Werner (Clark University).
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteThanks for finding our blog which has been inactive now for several years. Not sure how your ideas connect to questions concerning visual space, though, and what you describe sounds like science fiction. Most of what you write concerns a third person (objective) view of an individual, not the first person (subjective) point of view each of us uniquely knows. Identical twins are typically very close psychologically but they are still two different people--i.e., two different subjective points of view--so merely making a "print" (or clone) as you describe does not solve the problem of the first person point of view, something each of us cannot share with another, though the late James Culbertson did think it was possible to connect to someone else's consciousness in his book "The Minds of Robots" and subsequent books that refined his ideas. The head of our lab at UCSD, V.S. Ramachandran, has also written about what is called the "Twin Paradox" in a paper published last year (2017) with Baland Jalal entitled "The Twin Vantage Point Paradox: A Thought Experiment." Rama has been thinking about this paradox for 40 years.
ReplyDeletehttps://www.researchgate.net/profile/Baland_Jalal/publication/322445618_The_Twin_Vantage_Point_Paradox_A_Thought_Experiment/links/5a6713bcaca2720266b436fe/The-Twin-Vantage-Point-Paradox-A-Thought-Experiment.pdf
Hy!
ReplyDeleteYour block is an informal block. I am very impressed by continuing with your information and continuing to write.
Thanks for sharing.
here
Relative to Coxeter's "Regular Polytopes" I recently explained in my paper "The Dimensionality of Visual Space" that the structure of visual space is to be distinguished from structures *in* visual space, much as that distinction is made in cosmology (physical geometry). That distinction is also necessary to understand/explain certain visual illusions such as the "reversible" Necker cube.
DeleteReally a great and valuable blog, I appreciate your work which you have done about hologram. please suggest me the best Security holograms manufacturer in India and Keep sharing with us.
ReplyDeleteExcellent Blog. I think very useful for all people. keep up the great posts! kindly suggest me the affordable and best holographic wads manufacturer in India.
ReplyDeleteHello,
ReplyDeleteThe content of your blog is awesome.Thanks for sharing this blog.Anyone please suggest me holographic wads manufactures in India at affordable price.
This comment has been removed by the author.
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteThis comment has been removed by the author.
ReplyDelete