Tgdtheory.fiQuantum Model of Memory anen1, February 1, 2006 1 Department of Physical Sciences, High Energy Physics Division, PL 64, FIN-00014, University of Helsinki, Finland.
Recent address: Puutarhurinkatu 10,10960, Hanko, Finland.
Geometric and subjective memories . . . . . . . .
p-Adic physics as physics of intentionality . . . . . .
Spin glass model of memories . . . . . . . . .
Third person aspects of memory Symbolic and cognitive representations of memories Biosupercomputers and memories . . . . . . . .
Different types of memories Geometric and subjective memories . . . . . . . .
'Memories' with respect to geometric time as simulations 12 Mindlike spacetime sheets and simulations . . . .
The difference between intentions and geometric mem-ories . . . . . . . . . . . . . .
What is the temporal extension of mindlike spacetimesheets? . . . . . . . . . . . . . .
Durations of mindlike spacetime sheets representingsubselves What is the subjective duration of 'our' self? . . .
Habits, skills, associations . . . . . . . . . .
Spin glass model of learning and long term memories . . .
Long term memories . . . . . . . . . . . .
Long term memories as geometric memories? . . .
Geometric memories as sensory experiences with theobject of the perceptive field in the geometric past? Long term memories as memories of higher level self? More complicated scenarios . . . . . . . .
Implicit memories Procedural memories . . . . . . . . . . . .
Quantum computation in biological length scales, PenroseHameroff hypothesis, and mirror model of long term mem-ory Is quantum computation possible at all in TGD universe? .
Macrotemporal quantum coherence and molecular sex . .
Do quantum superpositions of tubulin molecule conforma-tions last for a time longer than CP2 time? . . . . .
Naive argument: No . . . . . . . . . .
Could gravitational interaction transform zero modesto quantum fluctuating degrees of freedom? . . .
Could classical gravitation stabilize irreducible boundstate entanglement? . . . . . . . . . .
Long term memory and gravitational MEs . . . . . .
Dark matter hierarchy and hierarchy of long term memories .
Model for long term memories Mirror mechanism . . . . . . . . . .
Classical communications and non-episodal memories Negative energy Z0 MEs as ideal entanglers with thegeometric past . . . . . . . . . . .
Is the right brain hemisphere the quantum entangler? Synesthesia as a key to the mechanism of episodalmemory . . . . . . . . . . . . .
Left-handedness and episodal memory . . . . .
NDEs and long term memories . . . . . . .
Dejavu experiences and memory feats . . . . .
Going to the neuronal level . . . . . . . . . .
Which parts of the brain are the quantum entanglers? Where the classical signals are generated and received? 54 Is memetic code used to code declarative long termmemories? . . . . . . . . . . . . .
What about other synchronous EEG frequencies? . .
Hippocampus and long term memories . . . . . . .
Anatomy of hippocampal system . . . . . .
Memory deficits and hippocampus . . . . . .
Hippocampus and declarative memory . . . . .
Hippocampus provides spatial and temporal context .
Remote emotions and associations? . . . . . .
Memory consolidation and long term potentiation Relationship between cortical and hippocampal EEGs Microtubuli and long term memory . . . . . . . .
Basic findings about the correlation between long termmemory and microtubuli . . . . . . . . .
How microtubuli could relate to declarative long termmemories? . . . . . . . . . . . . .
Relation to the general model of long term memories .
What about effectively 2-D and 3-D memory repre-sentations? . . . . . . . . . . . .
The neural realization of long term memories has remained to a high extent a mystery in the framework of the standard brain science.
The TGD based quantum model for memory have developed graduallyfrom the basic realization that in TGD framework the identification ofquantum states as quantum histories makes it un-necessary to storeinformation about the geometric past to the geometric now. This hasdeep implications.
a) It is possible to separate genuine geometric memory recall from apparent memory recalls such as feature recognition, associations, andimplicit and procedural memories. There are no memory storages inbrain and only memory representations abstracting the essential as-pects of experience are needed.
b) The models of long term memory based on the assumption that information about the geometric past is stored in the recent state ofthe system predict that the new memories should mask the old ones.
It is however known that childhood memories are the stablest ones. InTGD framework this ceases to be a problem.
Mirror mechanism provides a very general mechanism of long term memory. To remember something at a temporal distance T in thegeometric past is to look at a mirror at a distance cT /2. If the mirroris quantum mirror only a timelike entanglement (allowed by the non-determinism of K¨ ahler action) of the mental image of the geometric past with a mental image in brain now is needed. The un-necessityto communicate memories classically implies extreme generality of themechanism: all kinds of memories: sensory, cognitive, verbal,. canbe recalled in this manner. Even the mechanism of memory recall bycue can be generalized since the notion of tele association makes inprinciple sense.
The basic objections against this over-simplified picture is that there is no guarantee that the reflected ME returns to the brain andthat there is no control over the time span of long term memories. Thenotion of magnetic body allows a more realistic formulation. Brain orthe personal magnetic body generates spontaneously negative energyMEs with all fundamental frequencies. These MEs can be also curvedand are parallel to the closed flux tubes defining the personal mag-netic body and connect geometric now with the brain of the geometricpast: multiple reflections are probably required to achieve this. Thelength of the closed magnetic loop defines the time span of the corre-sponding long term memory. The sharing of mental images by timelikeentanglement allows to communicate the desire to remember to the ge-ometric past, and gives rise to the memory recall in the case of episodalmemories. In the case of non-episodal/declarative memories the mem-ory is communicated from the brain of the geometric past by classicalcommunications using positive positive energy MEs which propagate with an effective phase velocity much lower than light velocity alongclosed magnetic flux tubes and generate in the receiving end symbolicrepresentation of the memory.
Macrotemporal quantum coherence is further important piece of the model. The understanding of how macrotemporal quantum coher-ence is made possible by the spin glass degeneracy led to a concreterealization of the mirror model and also provided a connection withthe ideas of Hameroff and Penrose. When a bound state is formed thezero modes of the bound state entangled subsystems become quantumfluctuating degrees of freedom. This means that state function reduc-tion and state preparation cease to occur in these degrees of freedom.
The bound state is in a kind of long-lasting multiverse state, or stateof 'oneness' experientially, and the sequence of quantum jumps de-fined by the duration of the bound state behaves effectively as a singlequantum jump. Macrotemporal quantum coherence making possiblesupercomputer like activities becomes possible.
The spin glass degeneracy associated with the join along bound- aries bonds (the space-time correlates for the bound state formation)lengthens the lifetimes of the bound states dramatically and solves thusthe basic objections against quantum consciousness. The spin glassdegeneracy is due to classical gravitational energy of the system. Thequantum jumps between different classical gravitational configurationsinvolve the emission of gravitational (equivalently Z0) MEs and the in-tention to remember is realized as a transformation of p-adic ME tonegative energy gravitational ME. The fact that classical gravitationalfields couple to classical gauge fields with a coupling which is about108 stronger than the ordinary gravitational coupling, could play animportant role too. Water clusters and macromolecules with sizes inthe range of cell membrane thickness and cell size are good candidatesfor generating gravitonic MEs responsible for all geometric memories.
Also classical Z0 interaction might be involved since gravitonic MEscan be regarded also as Z0 MEs.
A rather detailed neuro level model of long term memory is devel- oped and the model conforms nicely with the basic facts known aboutthe relationship of hippocampus and long term memory.
The ideas related to the quantum model of memory have developed graduallyfrom very general ideas to reasonably concrete models and a connection withbiological quantum computer type systems has emerged. It is good to listthe basic ideas and notions briefly to get and idea about this process whichis still continuing.
Geometric and subjective memories The identification of moment of consciousness as quantum jump betweenhistories implies two kinds of time developments, subjective and geometric,and therefore also two causalities and memories. By the 4-dimensional gen-eral coordinate invariance of quantum TGD, geometric memories containinformation about entire quantum and classical histories. This means thatthere is no absolute need to store memories of the geometric past to thegeometric now. This has dramatic implications for the modelling of brainand allows to get rid of the basic problem of the memory models, namely thefact that the storage of new memories unavoidably tends to destroy the oldmemories whereas it seem that childhood memories are actually the moststable ones.
p-Adic physics as physics of intentionality In purely real contex one ends up with the problem that there is no cleardifference between memories and intentions: intentions are just memoriesabout the geometric future. Why the memories/predictions of geometricfuture and past are so different? The solution of the problem came when Irealized that p-adic physics is physics of cognition, imagination, and inten-tion. p-Adic spacetime regions represent intentions and are about geometricfuture. In quantum jumps transforming intentions to actions p-adic regionsare transformed to real spacetime regions representing geometric memoriesand inducing self-organization patterns giving rise to macroscopic actions.
This amplification process is possible by the quantum criticality of TGDuniverse implying initial value sensitivity. Psychological time correspondsto the front of a p-adic-to-real phase transition proceeding to the directionof geometric future.
Spin glass model of memories One of the relatively early ideas was that the 4-dimensional quantum spinglass property of TGD universe must have fundamental role in the realizationof memories. Spin glass property predicts fractal energy landscape in whichthere are valleys inside valleys inside valleys and memories correspond toself-organization patterns associated with subself having interpretation asprocesses leading to bottoms of various valleys. In TGD framework energyminima are replaced by the maxima of K¨ ahler function defining configuration space geometry as a function of zero modes which are effectively classicalvariables in the sense that in each quantum jump a complete localization occurs in these variables. One can also consider the interpretation of 'energy'as binding energy of bound states as function of zero modes. The higherthe value of the binding energy, the deeper the valley, and the higher theprobability that system ends up to the bound state and the longer the timespent in the bound state.
One can also regard life as a process of carving a 4-dimensional statue gradually quantum jump by quantum jump. The longer the extension ofthe valley in the temporal direction and the larger the number of copies ofthe valley is, the more reliable the memory recall is. The best manner tolearn to remember is to remember. The depth of emotion determines howdeep and long in temporal direction the valleys representing memories are.
MEs provide a mechanism of long term memory which differs from ordinarysensory perception only in that the ME giving rise to a geometric memoryhas much longer duration with respect to the geometric time than the MEgiving rise to ordinary sensory perception. To remember something at tem-poral distance T in the past is to look at a mirror with length L = cT /2.
The mirrors in question must have astrophysical sizes measured in light yearstypically and this of course raises obvious objections against the model. Al-though this mechanism as such is too strong an idealization, it can serveas a starting point. For instance, MEs can be also curvilinear and couldpropagate along closed magnetic flux loops of the personal magnetic bodyserving effectively as wave cavities and suffer few reflections: this wouldmake possible high precision targeting.
At quantum level remembering means sharing of mental images: this corresponds to the quantum entanglement between the subselves of the ge-ometric now and of the geometric past. The classical non-determinism ofK¨ ahler action is essential in making possible entanglement between systems having timelike separation. This would be the mechanism of episodal mem-ory, For non-episodal memories only the the mental image representing thedesire to remember would be shared, and the answer from the geometricpast could be realized as classical communications using MEs. Commmu-nication would be based on some code, perhaps memetic code, and wouldgenerate a conscious experience in the receiving end, typically verbal mem-ory. Positive energy MEs would propagate with ultra low effective phasevelocity inside brain or along magnetic flux tubes of astrophysical size withsub-luminal effective velocity (say alpha wave effective velocity). The mostoften needed non-episodal memories, say short term memories, could be communicated automatically: in this case the memory recall would be ageometro-temporally local operation, much like taking a sample from a datastream representing particular kind of memories with a particular time span.
The option is probably not realized for all non-episodal memories since thiswould require large energy expenditure.
In this framework synaptic strengths code only cognitive representations and learned associations, not genuine information about the events of thegeometric past. Brain can be seen as kind of a collection of standardizedfeatures serving as building blocks of sensory and memory representations.
Long term memory is coded in the classical em/gravitational fields associ-ated with and in coherent light/gravitons generated by MEs in hologram likemanner. Any finite spacetime region receiving the classical em field of co-herent light/gravitons generated by it gets hologram like picture containinginfo about entire geometric time interval spanned by ME. If vacuum currentis localized to some restricted spacetime region (it can be!), the hologrammicinformation is about this region and receiver anywhere along the ME getsmore or less the same information since hologram is in question. Note alsothat the lightlikeness of the boundary of ME implies that ME selves havetemporal extension defined by the length of ME.
Third person aspects of memory Topological quantization implies the notion of field body: field body ac-companies any system be it molecule or human body. Field body serves askind of a manual providing higher level abstract representations about thequantum aspects of the physical body. The model of sensory representationsrealized at personal magnetic body and at Earth's magnetic body explainsboth the first and the third person aspects of our sensory experience. Alsomemories should have third person and transpersonal aspects realized atthe magnetic body of Earth. This prediction is testable: moon travellerconsciousness should have different third person aspect or this aspect couldbe even absent. Third person aspect should be crucial for the generation ofsocial structures and the rapid weakening and reversal of Earth's magneticfield predicted to occur within next 2 millenia might have dramatic effectsfor the future of the civilization.
The sharing of mental images is crucial for the model of the third person aspect of memories. What happens is that subself of brain entangles withwith the subself of the magnetic sensory canvas in the geometric past. Onecould perhaps interpret spontaneous episodal memories as a basic exampleof memories communicated by some subself of magnetic Mother Gaia to us.
Symbolic and cognitive representations of memories Most of our memories are not direct re-experiences. In fact, it would bedifficult to tell whether memory is really in question if this were the case.
Rather, memories are highly conceptual and based on symbolic representa-tions making possible huge filtering and compression of information. Only insome special cases direct re-experiencing occurs. The inherent nondetermin-ism of the p-adic field equations and the classical non-determinism of K¨ action make possible to represent sequences of quantum jumps determiningthe contents of consciousness of self at spacetime level in terms of p-adic orreal spacetime sheets, that is cognitively and symbolically. Symbolic rep-resentations are crucial for memories whereas cognitive representations arecrucial for intentions. Symbolic representations allow to store informationabout geometric past to geometric now: history writing is just this kind ofactivity. Also brain is doing history writing: to remember is also to form anew memory representation.
It is highly plausible that memory representations are highly abstracted and that the signals from the geometric past do not recreate directly theexperience but serve as names for standardized self-organization patterns ofneuronal activity, 'features' giving charicature of the experience. This meansthat it is not easy to distinguish between TGD based model and standardmodel of memories.
Biosupercomputers and memories The most recent but certainly not the last step in the development of ideaswas the realization of a connection between macrotemporal quantum co-herence, quantum spin glass property of the TGD universe, classical andquantum gravitation, and the mirror model of geometric memories.
The interpretation of quantum jump as a creation of a totally entangled holistic state U Ψi which is then analyzed to pieces allows to interpret selfmeasurement cascade as a conscious analysis. The temporal fractality ofconsciousness suggest that the lifecycle of any self can be seen as a genera-tion of multiverse of potentialities followed by analysis (and decay) process.
One can see the situation also differently. The conscious experience of selfis average over moments of consciousness and the eventual thermalizationinduced by the quantum jump sequence destroys all conscious information.
There must be some mechanism hindering this and making macrotemporalquantum coherence possible.
To achieve macrotemporal quantum coherence self must be irreducible self for which self measurements and analysis do not occur. The self mustalso have large number of zero modes transformed to quantum fluctuatingdegrees of freedom and this is achieved if self corresponds at spacetimelevel to a join along boundaries condensate. In this process the zero modesof the condensing spacetime sheets become quantum fluctuating degreesof freedom.
In this 'state of oneness' self is able to carry out quantum computer like information processing which is the diametrical opposite ofanalysis. The decay of this bound state to its components corresponds tothe analysis period at the level of self.
Macrotemporal quantum coherence is possible by the quantum spin glass property of TGD universe making the lifetimes of bound states much longerthan in the universe obeying standard physics. Different almost degeneratevacuum spacetimes differ only because they have different classical gravita-tional energies. The quantum transitions between these almost degeneratestates involve emission of MEs representing gravitons. These topologicalgraviton rays are reflected from the curved almost vacuum spacetime sheetacting as a gravitational mirror and self energy diagrams involving emissionand absorption of the gravitonic ME have interpretation as correlates forthe geometric memory recall. The time scale of human memories is betweenmillisecond and 100 years and this time scale characterizes the gravitationalenergies for systems having sizes between cell size and cell membrane thick-ness (the number theoretical miracle is that all p-adic length scales in thisreange correspond to Gaussian Mersennes). Microtubules are excellent can-didates for realizing long term declarative memories at bit level so that aconnection with Penrose-Hameroff views emerges.
Different types of memories TGD predicts two kinds of memories corresponding to two different time de-velopments. There is deterministic (in generalized sense) time developmentwith respect to the geometric time and the nondeterministic time develop-ment by quantum jumps with respect to the subjective time. The memorieswith respect to subjective time are about previous conscious experiencesand 'real' whereas geometric 'memories' are prophecies giving simulationsof the geometric past and future assuming that quantum jumps do not alterthe macroscopic properties of the spacetime surface. A good visualizationis following: each quantum jump represents particular geometric memorywhereas the heap of these memories gives rise to subjective memory. Thecomparison between expectations and reality is obviously a central part of mentality and the heap structure implies that this comparison is a basicfunction of conscious mind not reducible to anything simpler. It is well-known that our memories involve a lot of construction and are more likestories consistent with what we actually have experienced than actual doc-uments of what happened. Perhaps geometric memories constrainted bysubjective memories give rise to the 'story'.
One can distuinguish between several memory types such as short term memory and long term memory, episodal memory, procedural memory, im-plicit memory and associative memory, and it is interesting to try to findwhether these memories could be understood in the proposed conceptualframework. In the discussion below concrete mechanisms for the realizationof geometric memory are not discussed: the reader interested on this aspectof geometric memory can consult [D1].
Geometric and subjective memories There are two times in TGD: subjective and geometric. In accordance withthis there are also two kinds of memories: subjective and geometric1.
a) The temporal binding of the experiences associated with quantum jumps occurred after the last "wake-up" of the self gives rise to subjectivememory defined as memory about earlier conscious experiences and is identi-fiable as an immediate conscious memory, "phychological now", presumablyof duration of fraction of second in case of sensory experiences. There is infi-nite hierarchy of subjective memories and if long term memories are genuinesubjective memories (this need not be the case!), they could correspond toconscious short term memories of higher level selves somehow communicatedto the lower level. An essential element is the possibility of subselves insideself having much shorter lifetime and organized in a subjecto-temporal se-quence: without them the average over the quantum jumps would destroythe information and it would not be possible to remember the digits of aphone number. Various rhytmic actions (such as micro tremor of eyes at80 Hz frequency and muscle tremor) could generate a sequence of subselveswith constant duration and thus a clock measuring subjective time.
b) Geometric memories are like a classical physics based model for the universe. They are memories with respect to geometric rather than subjec-tive time and predict what must have happened in the geometric past and 1The attribute 'subjective', as it is used in TGD context, does not have quite the same meaning as it usually has as something non-objective and unreliable: 'subjective'derives its meaning from 'subjective time' as consciously experienced time as opposed tothe geometric time of physics.
what will happen in the geometric future assuming that world is classical(no quantum jumps). The temporal extension of the mindlike spacetimesheets and the notion of the association sequence (3-surfaces consisting of asequence of spacelike 3-surfaces with timelike separations providing a sim-ulation of classical history) make possible geometric memories. A naturalhypothesis is that the macroscopic spacetime associated with the final stateof the quantum jump represents the geometric memory. Of course, only partof it becomes conscious and temporal binding implies that self experienceskind of temporal average of the geometric memories associated with thequantum jumps. An attractive possibility is that our long term memories,which have narrative character and are unreliable, correspond to geomet-ric memories. This would mean that there is no need for memory storagemechanisms, four-dimensional brain would take automatically care of mem-ory storage.
Intentionality manifests itself in many ways: as expectations of the fu- ture, planning, goals, desires, fears, imagination, intuition etc. It seemsnatural, and this is the only possibility given the fact that it is not possibleto know anything about future quantum jumps, to identify all aspects of in-tentionality with the predictions of the expected geometric future providedby the mindlike spacetime sheet. Geometry as such contains nothing inten-tional. Rather, the intentional aspects of the conscious experience reflectthe attitudes towards the expectations provided by the geometric memory.
'Memories' with respect to geometric time as simulations Geometric memories are predictions/simulations for what would happen ifno further quantum jumps would occur and what would have happened ifno quantum jumps had occurred in the past. Simulations and expectationsrather than real memories are in question. Geometric memories becomereliable in the classical limit, when the effect of quantum jumps becomesnegligible. In the deterministic world of classical physics geometric memorieswould be absolutely reliable. It is indeed possible to predict rather reliablywhat will happen in the solar system during the next decade. Geometricmemories are a prequisite of the intentionality often regarded as a basiccharacteristic of conscious mind: beliefs, expectations, plans, etc. involvegeometric memory in an essential manner. The computational approach tomind assumes only geometric memories.
The memory with respect to geometric time is possible even assuming that single quantum jump determines the contents of conscious experiencecompletely. However, if the contents of conscious experience are determined completely by the initial and final quantum histories of single quantum jump,it is in principle impossible to have genuine memories about previous con-scious experiences. This does not make it impossible to have a model for themost probable subjective life history through simulation. Quantum statisti-cal determinism could make these simulations possible. One must howeveradmit that the hypothesis about subjective memory, naturally identifiableas a short term immediate memory defining the duration of psychologicalmoment, makes things extremely simple and natural. One could also arguethat in a universe without subjective memory it would not be possible todiscover the notion of quantum jump so that internal consistency of thetheory of consciousness requires genuine memory about earlier consciousexperiences.
Mindlike spacetime sheets and simulations It is a fact that we can plan future in the time scale of life time. We can alsoquite reliably extrapolate to the past without direct memory of what hap-pened. The simplest explanation is that the time extension associated withthose mindlike spacetime sheets, which we have access to, is of the orderof lifetime or perhaps even longer. The simplest model for the simulationwould be based on an ensemble of thoughts scattered around entire materialspacetime history defined by, say, my body. Cognitive neutrino pairs wouldrealize thoughts as Boolean algebra of statements and could be present ev-erywhere in condensed matter, in particular in water, which is expected tohave very rich hierarchy of spacetime sheets. Self would experience the sumof the abstracted experiences of ensemble members and experience a simu-lation about what happens in future and what happened in past assumingthat quantum jumps will not occur in future and did not occur in past.
Of course, selves could also do what computers do, namely mimick other selves by building cognitive representations about them at their own space-time sheets. This would make it un-necessary to jump between the levels ofthe self hierarchy. These representation could have quite different temporaland spatial scales and the presence of the time scaled versions about timedevelopment of other selves would realize the fractality aspect related to theidea about Universe as a hologram. DNA could be an example of this kindof simulation of the entire lifespan of individual in molecular length and timescales. Monte Carlo simulation of elementary physics experiment could bealso regarded as a simulation of this kind.
The difference between intentions and geometric memories Intentionality, understood here as time-directedness, manifests itself in manyways: as expectations of future, planning, goals, desires, fears, imagination,etc. The basic element of mentality is the comparison between the expecta-tions of future and what actually occurred. In TGD framework this tensionbetween potential and actual can be understood. The temporal extension ofthe mindlike spacetime sheet makes possible expectations of what happensin the future assuming that no quantum jumps occur or at least that quan-tum jumps do not change the macroscopic spacetime. Single quantum jumpcontains information about this kind of expectations. Subjective memory inturn tells what happened actually. Therefore it seems natural, and this isthe only possibility given the fact that it is not possible to know anythingabout future quantum jumps, to identify the predictions of the expected geo-metric future provided by the mindlike spacetime sheet as a basic prequisiteof intentionality.
Subjective memory makes it possible to compare the expectations with what really occurred since subjective memory is kind of a heap of predictionsof future arranged with respect to the value of the psychological time. Theorigin of at least some emotions, which often involve a comparison of whathappened and what was expected to happen, is perhaps here. It is quitewell possible that all comparisons must be realized as comparisons of thesubjective and geometric time developments. It seems that self can alsocompare its subselves, which correspond to simultaneous mental images.
The possibility of this comparison provide a solution to the paradox raised by the innocent question 'How do I know that the me of today isthe same as the me of the yesterday? How do I even know that I existedyesterday?'. The solution might be simple: mindlike spacetime sheets haveextension which can be much longer than the duration of the subjectivememory. Therefore subjective memories contain information about the ge-ometric me of the yesterday and geometric me of today and since theseme's resemble each other quite a lot, the conclusion is that also the yester-day's me was a conscious self living in this same body. It is however quitepossible that temporal entanglement with higher selves still rememberingmy past wake-up states is also involved and realized as a formation of joinalong boundaries bonds between the mindlike spacetime sheets of my selfand of higher level self. Higher level self could also communicate directlythe subjective memories about my existence to me.
The difference between intentions and memories remained a puzzle for a long time. The answer was finally provided by the view about psycho- logical time as a value of the geometric time characterizing the position ofthe p-adic-to-real phase transition front propagating to the direction of thegeometric future. The MEs representing intentions are p-adic whereas thoserepresenting memories are real.
What is the temporal extension of mindlike spacetime sheets? With respect to subjective time self and its subselves can be characterizedby the typical durations of the wake-up state. With respect to the geo-metric time self (or rather, mindlike spacetime sheet) can be characterizedby its own duration and the durations of the mindlike spacetime sheetswhich it contains. The time span for the predictions and memories pro-vides an estimate for the duration of mindlike spacetime sheets. mindlikespacetime sheets can have timelike separations. mindlike spacetime sheets ofgeometric past could represent memories so that conscious memories couldbe regarded as multitime experiences and the distances between mindlikespacetime sheets could be quite large, of order lifetime.
Durations of mindlike spacetime sheets representing sub-selves Sensory experiences seem to correspond to a well defined geometric nowhaving perhaps duration of order .1 seconds. Thus it seems that mindlikespacetime sheets representing my sensory subselves have rather short timeextension, of order .1 seconds. 'Ontogeny recapitulates phylogeny principle'(ORP) suggests that the extension is of same order as the duration of theimmediate subjective memory, something like .1 seconds. This predictionis certainly consistent with the typical resolution of the sensory experience,say the ability of the visual system to discriminate subsequent pictures asseparate pictures. Quite generally, the p-adic time scale Tp = Lp/c charac-terizing the mindlike spacetime sheets gives the first guess for the durationof the mindlike spacetime sheet and duration of geometric memory providedby it. Note that .1 seconds gives for the p-adic length scale Lp and estimatewhich is about circumference of Earth! The fact is that we have childhood memories, plan future and make reliable predictions. This is not in contradiction with the duration of themindlike spacetime sheets associated with sensory subselves. The mindlikespacetime sheets representing subselves (mental images) can be located ingeometric past or future so that multitime experiences with mindlike sheetsof past and future contributing to the experience are possible.
The duration of .1 seconds is the duration of typical subselves represent- ing our mental images. The geometric duration of the mindlike spacetimesheet representing our 'main self' should be much longger since it containsmindlike spacetime sheets distributed along entire life span.
The subselves which have fallen asleep, wake-up again generating new wave of sensory experience.
For instance, mental images (after images) typically re-appear periodically. We are also mental images of larger selfin the hierarchy and the periodical appearenc e of of our mental imagessuggests that also we appear periodically as mental images of this largerself. This would mean reincarnation in the geometric past so that our lifewould be lived again and again. Entire trains of mindlike spacetime sheetscould wander through time again and and experience what it is to live in aparticular body. Therefore my body could live again and again: by p-adicevolution each life would tend to be slightly better than the previous one.
The civilizations of past could be still well and alive and even more civilized!This picture could perhaps explain why persons in their old age sometimesbegin to live their childhood again.
What is the subjective duration of 'our' self ? Our conscious experience is some kind of an averaged sum over all con-scious experiences associated with the quantum jumps occurred after thelast 'wake-up'. If the averaging is completely democratic, the only possi-bility is that our sensory subselves have duration not much longer than thethe time resolution of the sensory experience of order .1 seconds. Contraryto the original beliefs, this does not in principle pose any limitation to theduration of 'our' self.
There are thus several options concerning the duration of our self.
a) Our self could have duration not much longer than the duration of immediate short term memories of order .1 seconds. The ability to rememberdigits of a phone number requires that the duration is indeed longer. Forthis option it is not at all obvious how the subjective experience of personalcontinuity is possible.
b) The duration could also correspond to the wake-up period. Also now the problem is how we know that this self existed already yesterday. Notethat the gradual thermalization of subselves means that subjective memoriesrepresented by subselves get gradually fuzzy so that the digits of a phonenumber are forgotten even if our self has duration of order wake-up time.
c) Our self has a duration of order lifetime, or even longer and only the mental image representing our physical body has duration of order lifetime.
A possible objection is that the mental image representing our self becomesgradually more and more entropic unless it manages to fight against secondlaw. This might of course correspond to ageing.
Option c) deserves a more detailed consideration.
a) The geometric duration of our 'main' mindlike spacetime sheet should be of the order of life span if geometric memory explains long term memories.
'Ontogeny recapitulates phylogeny' principle would suggest that also thesubjective duration of our 'main' self is of order life time. This option wouldexplain elegantly the fact that we possess subjective identity: this kind ofsubjective identity would be a logical deduction in case that our main selfhas duration shorter than life time.
b) This option would mean that we are not actually unconscious dur- ing sleep but are only unable to remember anything about what happenedduring sleep. This would be rather natural since various sensory and cogni-tive subselves are not conscious during sleep periods so that also multitimeexperiences in which sensory subselves wake-up in night time are rare! Itmight be also possible to remember events occurred during sleep state onlyduring sleep.
c) Note that the claims about near death experiences in which entire life is experienced as a kind of film, could be interpreted as very intensiveexperiences in which mindlike spacetime sheets along the entire life span'wake-up' and give rise to multitime geometric memories. Alternatively, ifbodily self with a duration of order lifetime is a subself of our self (perhapsidentifiable as the self associated with our magnetic body), the bodily selfrepresenting entire life cycle could be experienced as a mental image. Alsoshorter bodily subselves forming a subjectotemporal sequence, 'film', couldbe experienced in the absence of the ordinary sensory input.
Habits, skills, associations The universe of TGD is quantum spin glass [C4]. This provides extremelygeneral conceptual framework for understanding how memories/habits/learnedskills/asso- ciations are formed.
a) Mental images (in particular memories) correspond to subselves un- dergoing self-organizing time development by quantum jumps leading toself-organization patterns selected by dissipation. Thus both memes andgenes, in particular long term memories, can be regarded as winners in thefight for survival in which dissipation is the ultimate Darwinian selector. In-hibitory and excitatory nerve pulses might physically realize "frustrations"which make possible large number of almost degenerate energy valleys.
b) The universe of TGD is quantum spin glass characterized by a frac- tal "energy" landscape having valleys inside . inside valleys (directoriesinside.inside directories). This structure is ideal for a hierarhical repre-sentation of memories. Memories must correspond to valleys of the spinglass "energy" landscape into which dissipation takes the system. Memoryformation is active process and memories are charicatures rather than pho-tos and deep valleys of the energy landscape represent these charicatures.
Hippocampus, known to be involved with the formation of the long termmemories, could control the rate of motion in these control variables. Theplastic regions of the brain are the most spin-glassy ones and are the mostprobable seats of the long term memories.
c) System has some territory in the energy landscape. The motion in the zero modes serving as control variables causes a slow shift of the entire terri-tory. Synaptic strengths corresponds naturally to the slow control variablescharacterizing the position of the territory. In the presence of a metabolicenergy feed and sensory input system moves around this territory.
Spin glass model of learning and long term memories The universe of TGD is quantum spin glass [C4]. This provides extremelygeneral conceptual framework for understanding how memories/habits/learnedskills/associations are formed.
a) Mental images (in particular memories) correspond to subselves un- dergoing self-organizing time development by quantum jumps leading toself-organization patterns selected by dissipation. Thus both memes andgenes, in particular long term memories, can be regarded as winners in thefight for survival in which dissipation is the ultimate Darwinian selector. In-hibitory and excitatory nerve pulses might physically realize "frustrations"which make possible large number of almost degenerate energy valleys.
b) The universe of TGD is quantum spin glass characterized by a frac- tal "energy" landscape having valleys inside . inside valleys (directoriesinside.inside directories). This structure is ideal for a hierarhical repre-sentation of memories. Memories must correspond to valleys of the spinglass "energy" landscape into which dissipation takes the system. Memoryformation is active process and memories are charicatures rather than pho-tos and deep valleys of the energy landscape represent these charicatures.
Hippocampus, known to be involved with the formation of the long termmemories, could control the rate of motion in these control variables. Theplastic regions of the brain are the most spin-glassy ones and are the mostprobable seats of the long term memories.
c) System has some territory in the energy landscape. The motion in the zero modes serving as control variables causes a slow shift of the entire terri-tory. Synaptic strengths corresponds naturally to the slow control variablescharacterizing the position of the territory. In the presence of a metabolicenergy feed and sensory input system moves around this territory.
One can consider two general models of learning and memory recall in this framework, the TGD version of the neural network model and thegenuinely TGD based mechanism on the notion of the geometric memory.
Consider first the TGD based version of the neural network model of mem-ory.
a) The possible memories of the system correspond its territory in the "energy" landscape. Learning means slow change of the shape of the ter-ritory so that memory valleys get gradually deeper and system ends up tothem with larger probability in future.
b) Repeated simulated annealing provides a promising memory recall mechanism. The feed of energy from metabolism kicks the system into amotion and dissipation leads it into some valley. If the valley is quite notcorrect (correct subdirectory but wrong subsubdirectory), a smaller kickleads the system to the bottom of some nearby valley which might be cor-rect. By applying a sequence of increasingly smaller kicks system finallyfinds the correct memory valley. The conscious attempt to remember corre-sponds naturally to an external force forcing the system to move in a correctdirection.
There are several objections to this scenario. The first mystery is how system knows that the experience is a memory: there seems to be nothingwhich would distinguish memory from the experience occurring for the firsttime. Second problem is that the formation of the new memories tends todestroy the old ones: the new territory is simply not the old one. Even if onecould circumvent this paradox, it is difficult to understand why the livelyepisodal memories of childhood are the most stable ones.
If long term memories are geometric memories then memory recall mech- anism corresponds to multitime experiences involving generation of mindlikespacetime sheets in both geometric now and past.
a) Learning by repetition means keeping some subsystem in some deep valley for a long period of geometric time (system is still in that valley inthe geometric past!). This corresponds to reverberating patterns in neuronalcircuits generated automatically or by learning by repetition. In this picturethe modification of synaptic strengths is not learning of memories but justwhat it seems to be: a modification of responses to sensory inputs necessaryfor survival.
b) The attempt to remember creates mindlike spacetime sheets located in the geometric past. The probability that a newly created mindlike spacetimesheet is located in the memory valley of long time duration is high and thusconscious memory recall becomes probable. Also very emotional and 'catchy'experiences generating long lasting memory valleys are easily remembered.
Childhood memories are often very emotional ones and therefore also themost stable ones.
No final vision about what memories are in TGD framework exists yet.
What is certain is that one can distinguish between geometric and subjectivememories. The idea that episodal memories are ordinary sensory experienceswith the object of the perceptive field in the geometric past is very attractiveand speculative hypothesis which might work in TGD Universe, but moreconventional explanation sounds more realistic in the context provided bythe standard neuroscience. What is lacking still is a clear vision about theprecise physical realization of long term memories.
Long term memories An important question is whether our long term memories correspond toeither geometric or subjective memories or whether they involve both aspectssomehow.
Long term memories as geometric memories? The unreliability and narrativeness of the long term memories would sup-port strongly the interpretation of at least episodal long term memories asgeometric memories, that is multitime experiences involving active mindlikespacetime sheets scattered along entire life span. This option is consistentwith the short duration of subjective memories, which can be even of order.1 seconds characterizing the duration of immediate sensory memories.
Geometric memories could be realized as multitime experiences involving mindlike spacetime sheets located around several moments of the geometrictime, provide the simplest realization for the long term memories.
a) The model solves the basic difficulties of the neural net models of long term memory. In the neural net models long term memories are representedby synaptic strengths. The problem is that the learning of new memoriesdestroys old memories. In particular, the stability of the childhood memoriesis difficult to understand. It is also hard to understand how brain knows thatthe experience represents memory. One cannot avoid the difficulty by sayingthat novelty detection tells that experience occurs for the first time since the notion of novelty does not make sense if conscious experience contains onlyinformation from single moment of geometric time.
b) TGD model is consistent with neural net models and actually gener- alizes them. Neural net in the spirit of TGD corresponds to brain as systemmoving in spin glass energy landscape. Self-organization by quantum jumpsleads the system to a bottom of an energy valley representing memory. Thismodel is consistent with the fact that there is no upper bound for auto-biographical memory. One can also understand how learning occurs. Therepetition of an experience means that energy valley becomes a canyon intime direction so that mindlike spacetime sheets in the geometric past havea large probability to end up to the region representing memory. In partic-ular, reverberating nerve pulse patterns are ideal for representing long termmemories.
c) Highly emotional experiences generate deep valleys and increase the probability of the system of the geometric past to stay at the bottom ofvalley. This explains why cildhood experiences are so stable. In fact, onecould identify primitive emotions of pleasure and pain as related to themotion in the spin glass energy landscape. Pleasure and pain could evendirectly correlate with the sign of the increment of the K¨ ahler function in the hopping motion in the spin glass energy landscape. Note that primitivepleasure and pain are are very much like sensory experiences and one couldregard them as sensory experiences of brain about its own motion in spinglass energy landscape. This leads to the generalization of the notions ofsensory experience and motor action to include the motion in spin glassenergy landscape and to a considerably new insight about the meaning ofthe brain architecture.
There are also perinatal experiences, memories about previous lives and transpersonal experiences having natural explanation in terms of geometricmemory realized as multitime experiences associated with mindlike space-time sheets located at different values of the geometric time. Transpersonalexperiences suggests that self is dynamical: if prenatal experiences, memo-ries about previous lives and transpersonal experiences are really what theyseem to be, the geometric time extension of self should dramatically increaseduring these experiences.
If 'our' self has duration of order lifetime, also subjective memories can contribute to our long term memories. As already found, this option does notexclude the possibility that our long term memories correspond to subjectivememories.
Geometric memories as sensory experiences with the objectof the perceptive field in the geometric past? The general theory of qualia to be developed in [cbookII] leads to the conclu-sion that geometric memories could be regarded as special kind of sensoryexperiences for which some objects of the perceptive field located in thegeometric past. One also ends up with a concrete models for the mecha-nism making long term memories possible by 'waking up' subselves of thegeometric past in selective manner by EEG frequencies. The unavoidableconclusion is that massless extremals (MEs) with durations of order lifetime,and hence with sizes which are measured in light years, are necessarily in-volved. Needless to say, one must give up the idea that we are nothing butour brains.
The fact that the lightlike boundaries of MEs serve as quantum holo- grams and have gigantic information storage capacities by the almost de-generacy of the states fits nicely with view. Lightlikeness means that 3-dimensional time=constant slice of Minkowski space is replaced with a slicewhich can have arbitrary long temporal duration so that memories becomeindeed possible. The fact that at least vision represents directly informationabout outer surfaces of 3-dimensional objects rather than objects themselvesbut contains information about time development over an interval of order.1 seconds fits nicely with this view.
The realization of long term memories in terms of magnetic quantum phase transitions induced by ME frequencies requires incredibly high fre-quency resolution. The resolution is of order ∆f /f ∼ ∆T /T giving ∆f /f ∼10−9 for time resolution of about ∆T = 1 seconds. An unrealistically highfrequency resolution is required if temporal coding by EEG frequencies isassumed. There is also another problem: if the signal to the geometric pastand back is between parts of brain, one cannot avoid zigzag type MEs effec-tively representing a repeated reflection between two mirrors. In the p-adiccontext these zigzag MEs are allowed by conservation laws (this might relatewith the fact that long term memories are mostly cognitive) but not in thereal context.
These observations suggests that one should allow MEs and magnetic flux tube structures with length scales of order light lifetime and try to in-vent a more elegant mechanism of long term memory. One might start fromthe mirror idea and consider the possibility that memory recall involves aquestion sent to the geometric past as a classical signal reflected back tobrain in a mirror formed by a magnetic flux tube: perhaps passive Z0 MEsare involved at this stage. Thus MEs with lengths of order of light lifetime (L = cT ) would be required. The answer presumably involves a transfor-mation of Z0 MEs to active em MEs and the generation of quantum en-tanglement unless it is present already: the recalled experience is shared bythe experiencer now and experiencer in the geometric past. The mechanisminvolves several purely TGD based features: the lightlike character of theboundaries of MEs making possible lightlike selves; spacetime sheets with anegative time orientation allowing classical signals to propagate backwardsin time; the magnetic flux tube structures associated with brain having sizesof order light years making possible MEs to form mirrors. Precognition isthe temporal mirror image of this mechanism.
If long term memories are in some sense sensory experiences with the object of the perceptive field in the geometric past, the notion of the mag-netic canvas should work also in these astrophysical length and time scales.
Consider first the constraints on this mechanism.
a) The sensory experiences at different levels of the magnetic hierarchy cannot be identical. This means that standard sensory representation usingmagnetic canvas must be applied to realize the episodal memory. This leavesonly two possibilities. Either the experience is coded to a lightlike vacuumcurrent and this information, when sent into future, regenerates the sensoryexperience there. Alternatively, future self could entangle with the self ofthe geometric past and share its experience.
b) Since MEs correspond to 3-surfaces moving with light-velocity, the only possible realization of the communications between geometric past andgeometric now is in terms of 'laser mirrors' connected by MEs representinggeometrically the light reflected in the mirror. The length of ME is givenby L = cT : 2T is the moment of the geometric past which gives rise tothe memory. Interestingly, Peter Gariaev has suggested that laser mirrorsare involved also with DNA . This means that a ME extending fromthe brain of the geometric now to the geometric past and the ME fromthe brain of the geometric past fuse with the same magnetic flux tube toform a representation for light reflected in a cosmic mirror. The MEs andmagnetic flux tube structures associated with the relevant parts of brainmust form pre-existing, tightly correlated structures since the probabilityfor the formation of this kind of mirrors accidentally is extremely small andthere is no guarantee that they connect parts of the same brain. Secondmirror would be obviously defined by the join along boundaries contact ofME with the magnetic flux tube. Hippocampus is a natural candidate forthe brain structure, at which the first mirror is located. The fact that MEsrepresent channelled energy means that distance is not a problem as far asenergetics is considered.
c) Active memory recall must involve a question sent to the geometric past followed by an answer communicated to future in some manner. Theremust be some difference between precognition and memory recall so that thequestion and answer cannot be realized in the same manner. This serves asan important guideline. Various arguments lead to the view that the desireto remember is communicated to the geometric past by sharing and fusionof mental images made possible by entanglement. In the case of episodalmemories also the memory recall would result in this manner. For non-episodal memories the memory would be communicated from the geometricpast using classical communications.
Sharing of mental images if time-like quantum entanglement is generated between the selves of the geometric past and geometric now. This is possiblein TGD framework, thanks to the non-determinism of K¨ ahler action making also MEs quantum holograms in quantum gravitational sense. The fact thatMEs represent lightlike selves, would be essential for this realization. Thebeauty of this realization is that the information need not be transferredclassically. This realization is actually a special case of the realization interms of zigzag ME in much shorter length scale: in this case a huge num-ber of reflections in the mirror pair would be required and it is difficult tounderstand how one could control the temporal position of the self of thegeometric past in this kind of situation.
This picture deserves some further comments.
a) If the higher levels of the magnetic self hierarchy are intelligent as one might expect (and even more intelligent than us), one can also considerthe possibility that the step in which the interaction of ME representinga question sent to the geometric past with the magnetic flux tube at thehigher level of the hierarchy is far from a mechanical interaction. Rather,the magnetic flux tube structure could act as an intelligent conscious systemrather than a mechanical relay station.
b) The process could also have interpretation as an exchange of two virtual MEs between brain and magnetic flux tube structure: kind of a verylow frequency counterpart of self energy Feynmann diagram realized as ageneralized Bohr orbit. The Feynmann diagrams for the emission of parallelphotons are infrared divergent. This encourages the expectation that theprobability for the presence of MEs parallel to the magnetic flux tubes isvery high and increases with the increasing length of ME. The spontaneityof the episodal memories is in accordance with this view. An interestingquestion is how these MEs relate to 1/f noise.
c) The assumption that the lengths scales of MEs and magnetic struc- tures are identical implies that the frequency of EEG ME equal to the mag- netic transition frequency fm fixes the length of the two MEs involved andthus the temporal location of the long term memory in the geometric past: This represents a frequency coding for the temporal location but in a mannerdifferent from the one proposed originally. In particular, this coding doesnot require ME frequencies to be in EEG range and defined with a relativeaccuracy of order E −9. In standard physics the idea about brain generatingMEs with a frequency scale of the order of the inverse of lifetime does notmake sense: in TGD context situation is different since this process occursin subjective time.
If this picture has captured something essential from the nature of the long term memories, the conclusion is that we are not at the top of themagnetic sensory hierarchy. Human body and brain generates extremelyweak magnetic fields and the corresponding magnetic flux tube structurescould serve as a sensory canvas making possible long term memories. Neardeath experiences [C3] could be understood in this framework if the weakmagnetic fields associated with the higher levels of the fractal hierarchy ofmagnetic structures utilize brain and body as kind of sensory and motororgans. Note that there is flux tubes inside flux tubes structure so thatordinary sensory experiences can be associated also with these flux tubes.
Long term memories as memories of higher level self ? The natural identification of the immediate short term memory as subjectivememory predicts that the life time of a human sensory self cannot be muchlonger than .1 seconds, the duration of psychological moment of time. Ourlong term memories correspond to much longer time interval and cannotthus correspond to our subjective memories. Entire hierarchy of subjectivememories is however predicted and a possible model for genuine long termmemories is as resulting from temporary entanglement with selves belongingto the higher level of the hierarchy. Also this identification is consistent withthe fact that there seems to be no upper bound on autobiographical memory.
Summation hypothesis implies that our genuine long term memories wouldbe sums over a large number of wake-up periods of self in the subjective pastof the self. Therefore one could perhaps understand how ageing self gainsgradually wisdom from experience: also the identification of the long termmemories as geometric memories explains this.
Higher level selves could communicate their subjective and geometric memories as well as the emotions generated by their comparison to us. Thefirst idea to come into mind is that communications occur during totallyentangled state, sleep or trance. For this option it is not at all clear howthe experiences of the higher level selves during entangled state could beours! In fact, we should lose our selves during entanglement with self char-acterized by larger p-adic prime. For instance, during sleep without dreamsentanglement with some higher level self should occur and we do not re-member anything about this. Trance is a second example of this: subjectperson does not remember anything about the trance state. Thus it seemsthat this mechanism cannot give rise to conscious long term memories. Thisdoes not however exclude the possibility that cognitive representations areformed during the communication and lower level self experiences them lateras memories. One function of sleep might be the generation of the entan-glement with higher selves making in turn possible the communication ofgenuine memories of subjective past to our mind. This communication couldrealize these memories as thoughts about the experiences of past realized asnerve pulse patterns regenerating these thoughts.
The so called semitrance mechanism [D7] avoids the objections against communications occurring in totally entangled state. During semitranceparts of brain are entangled with some higher level self. These selves cancommunicate their memories to that part of brain which is awake (commu-nication means generation of mental images). Ancient men received thesecommunications as sensory hallucinations ('God's voice'), very much likeschizophrenics, whereas modern man experiences them as thoughts and emo-tions which are often 'hallicinatory' in the sense that they are not automaticreactions to the sensory input. The TGD based vision for the development oflanguage and civilization modifies Jaynes's vision about bicameral man as aschizophrenic of modern society and relies on the notion of semitrance. Semi-trance mechanism is extremely general and could be present in all lengthscales. For instance, semitrance could provide the inhabitants of cell soci-eties (organisms) and protein societies (cells) with a personal self narrative(genetic determination of cell as self narrative!).
Semitrance mechanism survives the most obvious counter arguments.
a) The general objection is that the memories of the higher level selves are rather abstract. The assumption communication mechanism is restrictedto thoughts and emotions is however consistent with the abstract nature ofthe non-episodal long term memories. The most natural identification ofepisodal memories is indeed as personal geometric memories or possibly asartificially generated sensory hallucinations stimulated by higher level self during semitrance.
b) Since semitrance mechanism is only a communication method, ge- ometric and subjective memories remain the fundamental memory mecha-nisms. Therefore the nice features of the geometric memory are not lost. Forinstance, one can understand learning and the role of emotions and repetionin learning.
More complicated scenarios One can consider also more complicated scenarios for realizing long termmemories.
a) Ensemble of mindlike spacetime sheets could generate continuously cognitive representations remaining in ideal case unchanged and memoriesas ability to re-experience would be carried by mindlike spacetime sheetwhen it wanders to the direction of future. This would require that mind-like spacetime sheets replicate just as material spacetime sheets (DNA, cells,members of species) do. If mindlike spacetime sheets responsible for mem-ories of this kind have finite lifetime, say of order one second, short termmemories could be realized in this manner without cognitive populationexplosion. In fact, cell division might realize long term memories in cellpopulations. Perhaps also DNA replication might be regarded as this kindof memory.
b) The realization of long term memory and communication relying on replication is rather primitive and the fact is that neurons do not replicate.
A natural explanation is that neurons have discovered procedural memory,which means that long term memories could be realized dynamically: stan-dardized nerve pulse patterns generate standardized temporal patterns of an-tineutrino Z0 magnetization. This implies ability to regenerate the thoughtstimulated by the primary experience and associative learning would asso-ciate memories to experiences as thoughts. This picture would correspondto that of ordinary associative nets and is subject to the standard counterarguments such as the loss of old memories caused by the learning of thenew ones.
c) Sustainment of the mental images is indeed one of the basic mecha- nisms behind human intelligence and can be also seen as a manner to enhancethe probability that a geometric memory in the past is recalled. Sustainedmental images are analogous to the icons of the computer screen, which infact supports the idea that the evolution of computers mimicks in manyrespects the evolution of the brain. At program level icons correspond toprogram loops. At neural level to periodic neural process generating again and again the same mental image (not necessarily directly conscious to us).
d) Written language and symbols are the next step to the internal sus- tainment and make possible to achieve a given sensory and cognitive ex-perience in a controlled manner. Program files are obviously analogous tothe written language (the electronic control systems preceding the computerera were effectively computer programs but were not written as computercode, externalized). DNA could be seen also as ROM type memory of livingsystems.
Implicit memories A possible definition of implicit memories is as memories which exist butare not created in conscious experience of the subject person. Also im-plicit learning could be defined in this manner. A good example of implicitmemory is provided by a situation in which unaesthetized patient can quiteaccurately remember what has been said during the operation . An ex-ample of implicit learning is the learning of grammatical rules without anyexplicit (conscious) representation for them. The status of the implicit mem-ories and learning is not established. A possible reason for this is that it isnot easy to understand them in computational paradigm of consciousness.
Connectionism explains implicit learning and memories as unconscious for-mation of associations and mathematically modelled by the dynamics of theneural networks.
In TGD framework implicit learning and memories could correspond to learning and memories at the lower levels of the self hierarchy not usuallyconscious to us. In case that the mindlike spacetime sheet correspondingto our subself forms join along boundaries bond with a lower level self sothat lower level self fuses with the subself in question, its memories canbecome our conscious memories. ORP suggests that this process involvesalso the formation of quantum entanglement and this indeed must occur.
Biofeedback could be understand as a special case of this process. In theTGD based model for the quantum correlates of the sensory qualia thisprocess is key role. The memories communicated by semitrance mechanismcan indeed be and probably often are implicit.
One can consider also formation of join along boundaries bonds between our subselves and subselves of other persons. This is quite possible if oursubselves indeed correspond to topological field quanta representing ELFphotons associated with the EEG frequencies having size of even size ofEarth. Formation of join along boundaries contacts between topological fieldquanta of this size would make for us to experience the memories of other persons. This kind of mechanism could explain the memories of anesthetizedpatient about what happened during the operation as memories of subselvesof the persons participating the operation. An open question is whether themechanism could also explain also out-of body experiences, in which patientlooks himself from outside, sometimes involved with this kind of situations.
Implicit learning could also correspond to the development of various cognitive skills realized as self-organized self cascades so that no explicitrepresentation of the skill is needed: when initial value self wakes up, thecascade proceeds with highly predictable manner due to quantum statisticaldeterminism. Even the ontogeny could be regarded as this kind of skillimplicitely coded in DNA! Procedural memories Procedural memories seem to be mostly stabilized sequences of thoughts andmental images and the proposed model for cascade like generations of selvesprovides therefore a model for procedural memory. Procedural memoriescould be simple cognitive acts occurring again and again as a reaction tosome specific stimulus. mindlike spacetime sheet would carry them whiledrifting into the future. For an ensemble of selves with each self initiatingcognitive acts is in question, reliability of memories would result.
Quantum spin glass model of brain explanains for formation of the pro- cedural as resulting from quantum self-organization.
Dissipation caused by quantum jumps would automatically select skills, habits and eigen be-haviours as surviving self-organizing patterns. These patterns would corre-spond to deep valleys in the fractal energy landscape of the spin glass land-scape, which is effectively four-dimensional. Repetition would automaticallylead to the learning of procedural memories since it would extend the val-leys in time direction so that mindlike spacetime sheets would have largerprobability to enter to the valley and give rise to memory. For instance, re-verbrating nerve pulse patters in the memory circuits of brain would realizethis repetition.
Quantum computation in biological length scales,Penrose Hameroff hypothesis, and mirror modelof long term memory Penrose and Hameroff have proposed that microtubules could act as quan-tum computers. The quantum states involved would be quantum superpo- sitions of tubulin conformations and quantum gravitation would somehowmake these quantum superpositions stable. Long enduring quantum su-perpositions of the conformations of (say tubulin) molecules would allowto perform a multiverse simulation for the conformational behaviour of themolecules and this would certainly have evolutionary value.
Penrose-Hameroff hypothesis is highly interesting from TGD point of view since TGD Universe is quantum spin glass in the sense that there is aninfinite number of different configurations of spacetime sheets whose ener-gies differ only by the gravitational interaction energy. Also the generationof coherent gravitatons by MEs might have a role to play in the quantumphysics of living matter. Especially so because genuine quantum gravita-tional states are state functionals in the space of 3-surfaces, that is worldof worlds: therefore they should correspond to higher abstraction level ofconsciousness than ordinary elementary particles. Furthermore, the gravita-tional constant associated with the energy of the induced gauge fields is by afactor 108 larger than the gravitational constant associated with elementaryparticles. The task is to put these pieces together.
In the following I will discuss Penrose-Hameroff hypothesis in more detail from the point of view of TGD.
a) TGD Universe indeed allows quantum computing under natural as- sumptions and the huge quantum spin glass degeneracy broken only byclassical gravitation is crucial for the preservation of quantum coherence.
b) Quantum computation occurs optimally for irreducible selves which are in the 'state of oneness' and have no subselves (mental images) so thatthere are no dissipating subsystems. The paradoxal statement of mysticismthat completely empty mind is source of infinite wisdom has therefore aprecise content. Second prequisite is that all but center of mass zero modesof 3-surface representing say join along boundaries condensate of tubulinmolecules (microtubule) transmute to quantum fluctuating degrees of free-dom when 3-surfaces topologically condenses to a larger spacetime sheet.
Otherwise a complete localization in zero modes meaning state functionreduction would occur in each quantum jump and quantum computationwould not be possible in time scales longer than CP2 time about 10−39seconds defining the average duration of a single quantum jump (quantumjump corresponds to 'elementary particle of consciousness' having durationof CP2 time).
c) The problem is that standard physics predicts too short life times for the bound states so that quantum computations would still last for tooshort time. The huge spin glass degeneracy associated with the join alongboundaries bonds however implies that there is an immense number of bound states with almost degenerate energy. This means that the branching ratiofor the decay to unbound states is reduced dramatically and bound statelifetime increases.
d) An unexpected connection with the mirror model of long term memo- ries emerges: it is the topological correlates of gravitons which are mirroredfrom curved almost vacuum spacetime sheet. The reason why for gravitons isthat they have so weak interaction with background. What is especially fas-cinating is that classical gravitational binding energies in the range spannedby the cell membrane thickness and cell length scale (all p-adic length scalesin this range correspond to Gaussian Mersennes) correspond to time range1 millisecond- 100 years, the span of human memories. In particular, micro-tubule conformations could code for declarative long term memories. Also aconnection with the idea that so called 1/f noise (now gravitonic) is crucialfor consciousness and long term memory emerges.
Is quantum computation possible at all in TGD uni-verse? In TGD framework each quantum jump can be interpreted as quantumcomputation performed by entire universe. Unitary process Ψi → U Ψi,where Ψi is a prepared maximally unentangled state, corresponds to thequantum computation.
Then follows state function reduction and state preparation involving a sequence of self measurements and given rise to anew maximally unentangled state Ψf .
The problem is that simplest estimate for the increment of the psycho- logical time in single quantum jump is about 10−39 seconds from the ideathat single quantum jump is a kind of elementary particle of consciousnessand thus corresponds to CP2 time. This would mean that 1039 quantumcomputations occur during single second and conscious experience is averageover these quantum computations and the result of the computation wouldbe averaged out completely. This would look like another manner to saythat quantum computation is not possible.
That Nature would not allow quantum computation in time scales much above CP2 time scale looks strange and there sould be ways to get out ofthe problem.
a) The estimate, which is just a dimensional guess, could be simply incor- rect and the average increment of psychological time could be dynamicallydetermined and be much longer. The question is about how long averagetime interval the p-adic-to-real phase transition front shifts towards geo-metric future in single quantum jump. There are good reasons to believe that this time interval is common for living organisms: otherwise one endsup with strange paradoxes. It seems however difficult to believe that thisinterval is as long as say .1 seconds: we would not experience a continuousstream of consciousness if this were the case and 10 Hz would be naturaltime scale for the rate of all quantum transitions.
b) Situation changes if the quantum entanglement associated with the quantum computer is bound state entanglement stable against self measure-ments and if the quantum computer self is in a state of 'irreducible selfness'and therefore stable against self measurement. Paradoxally, in mystics thiscorresponds to the state of oneness without any mental images: the totalemptiness of mind would be crucial for quantum computation which is themost effective manner to perform information processing! In this case quan-tum jumping could preserve bound state for quite a long time. The haltingwould be caused by an external perturbation destroying the bound state.
The properties of the bound state plus interaction with environment wouldallow to estimate the typical duration of the quantum computation. Thistime would take the role of coherence time. This would suggest connectionwith the standard approach to quantum computation.
Irreducible selfness is not enough for quantum computation. Macrotem- poral quantum coherence in the sense that zero modes for the three-surfacecease to be zero modes under some conditions, is also necessary for quantumcomputation. The reason is that localization in the zero modes correspondsto state function reduction spoiling the quantum coherence.
One can imagine two alternative mechanisms transmuting zero modes to quantum fluctuating degrees of freedom.
a) Topological condensation transforms zero modes to quantum fluctu- ating degrees of freedom quite generally. In this case one would effectivelyhave no zero modes at all. This looks utterly unphysical conclusion.
b) The formation of join along boundaries bonds between 3-D space sheets implies that only the 'center of mass' zero modes remain whereasrelative zero modes become quantum fluctuating degrees of freedom. Thisoption looks realistic and is very natural in the case of tubulins and watermolecules, which indeed form join along boundaries condensates. The for-mation of join along boundaries bonds is indeed the basic mechanism for theformation of macroscopic quantum states and the correlate for bound statequantum entanglement. This would explain why water is so important forlife.
Whether one can one approximate quantum jump sequence as unitary Hamiltonian time evolution in case of a bound state is an open question.
Fractality of consciousness would suggest that one can in case of quantum coherence effectively treat long quantum jump sequence as a single quantumjump (just like one can treat molecules as pointlike particles in a reasonableapproximation) so that Hamiltonian description might make sense. Hamil-tonian time evolution would more or less correspond to a unitary operatorresulting as a product of the actions of the unitary operators U associatedwith the quantum jumps of the sequence. Discretized time developmentwould emerge automatically in this framework. Schr¨ odinger equation at in- finitesimal level would not make sense but this is of course not a practicalproblem.
The fact that oxidative metabolism is anomalously low during a neu- ronal syncrony supports the view that neuronal synchrony might give riseto bound-state entangled multineuron states in 'state of oneness'(the liber-ated binding energy would be usable energy). The quantum computationsperformed by the neuronal groups might last the typical duration of 'fea-ture', which is about .1 seconds, typical time scale of alpha rhytm.
Macrotemporal quantum coherence and molecular sex The formation of bound states is a generic mechanism for generating newquantum fluctuating degrees of freedom and could make possible quantumcomputation like processes and multiverse states of consciousness contain-ing large amounts of conscious information. At macrolevel sexual organismcould be basic example of multiverse state of oneness generated by the for-mation of quantum bound state between partners. Neuroscientists use totalk about rewards and punishments and one might argue that life involveskind of sexual pleasure as a reward for the formation of bound states at alllevels of hierarchy. Spiritual experiences would represent a more abstractexperiences of this kind involving the formation of bound states of the fieldbodies by MEs serving as field bridges.
Some examples are in order.
a) The binding of molecules by lock and key mechanism is a fundamental process in living matter and could generate large number of quantum fluc-tuating degrees of freedom and generate conscious intelligence. This couldexplain why long linear macromolecules are so important for life. From theviewpoint of classical chemistry it is not obvious why DNA is arranged intolong chromosomes rather than separate short threads. In TGD universe thereason why would be that for chromosomes the number of quantum fluc-tuating degrees of freedom and thus the amount of conscious intelligence ismaximized.
b) The Ca++ ions binding to microtubules and molecules like calmodulin could act as switch like bridges between water clusters and microtubulesand thus able to dramatically increase the number of quantum fluctuatingdegrees of freedom and initiate quantum computation like process. Thede-attachment of Ca++ ions would halt the process.
c) The binding of the information molecules to receptors is a universal control mechanism in living matter. In TGD universe information moleculewould initiate genuine quantum information processing lasting for the life-time of the information molecule-receptor complex. In particular, neuro-transmitters could induce molecular states of oneness in receptor-neurotransmittercomplex or perhaps even in larger-sized structures. If neurotransmittershave join along boundaries bonds to other neurons mediated by magneticflux tube structures, they could act as conscious quantum links in quantumweb and induce quantum computation like processes involving distant neu-rons just as link links in the web induce classical computations involvingdistance computers.
d) One could see information molecules and receptors as representatives of opposite sexes: information molecules being active quantum binders freeto move from flower to flower whereas receptors would be the passive partyattached to some structure. The binding of the information molecule to thereceptor would be the analog of sexual intercourse. Usually the receptors arebound to larger structures such as cell membrane and also the zero modesfor some parts of these larger structures could become quantum fluctuatingin the process.
Do quantum superpositions of tubulin molecule confor-mations last for a time longer than CP2 time? In TGD quantum superpositions of molecular(say tubulin) conformationscorrespond quantum superpositions of 3-surfaces representing protein con-formations and the question is whether they could last more than CP2 time.
Naive argument: No The first guess is that the conformational degrees of freedom of protein corre-spond in TGD framework to effectively non-quantum fluctuating zero modedegrees of freedom (zero modes do not contribute to the line element defin-ing the metric of the configuration space of 3-surfaces). In each quantumjump a complete localization in the zero modes must occur by mathemat-ical consistency. Standard quantum measurement theory results if there isentanglement between zero modes and quantum numbers associated with quantum fluctuating degrees of freedom: just like between spin of electronand its classical orbit in magnetic field in Stern-Gerlach experiment.
In each quantum jump the unitary process U generates a multiverse in which different protein conformations are in quantum superposition andstate function reduction selects a quantum superposition of the spacetimesurfaces such that zero modes have same values for all these surfaces andstate looks completely classical. If the average increment of the psychologicaltime in quantum jump is CP2 time about 10−39 seconds, this means that thequantum superposition lasts only for 104 Planck times and is what standardquantum gravity would also suggest apart from 104 factor. Thus TGD wouldnot support Hameroff-Penrose view if the simplest assumptions are correct.
Could gravitational interaction transform zero modes toquantum fluctuating degrees of freedom? The situation changes if the interaction of the tubulin molecule with theexternal world, say other tubulin molecules, somehow transforms zero modedegrees of freedom characterizing the shape of free tubulin to quantum fluc-tuating degrees of freedom, or rather the relative shape degrees of freedom.
The mechanism would be simply the formation of join along boundariesbonds between tubulins of microtubules which would transmute all but theoverall 'center of mass zero modes' to quantum fluctuating degrees of free-dom. I do not know whether molecular chemistry allows examples in whichmolecule can be said to be in quantum superposition of different confor-mations or whether molecules behave effectively classically as assumed inBorn-Oppenheimer approximation for calculating electronic states. In trans-lational degrees of freedom situation is certainly different.
Quantum fluctuations in the shape of the spacetime sheet belonging to a join along boundaries condensate characterized by prime p would be of the order p-adic length scale Lp ∝ p. The ratio of this length scale to the length scale of the larger spacetime sheet at which molecules are condensedtopologically, would be typically a small power of 2: pp/p1 ' 2(k−k1)/2, kand k1 primes or powers of prime. Typically the factor would be 1/2, 1/4, .
Interestingly enough, in biological length scales twin pairs k, k1 = k + 2 ofprimes are very abundant: in this case the value of the factor would be1/2 so that quantum fluctuations in shape would be maximal. Perhaps thismaximation of quantum fluctuations is something deep.
Could classical gravitation stabilize irreducible bound stateentanglement? Bound state entanglement gives rise to a 'state of oneness' in which quantumcomputing system is totally bound-state entangled and does not decay intosubselves in self measurement process and can thus behave effectively as anon-dissipating system and quantum compute. The estimates for the dura-tion of this kind of bound states tend to be much shorter than required .
The question is whether classical gravitational interaction could somehowstabilize these bound states.
The extremely low value of the gravitational binding energy is an objec- tion against the view that gravitational interaction could help to stabilizethe bound states. The huge degeneracy of the bound states could howeverchange the situation.
a) Suppose that spin glass degeneracy gives rise to a huge number of almost degenerate bound states for which only the classical gravitationalenergy is different and that for non-bound states this degeneracy is muchsmaller. The dominant part of the binding energy is of course somethingelse than gravitational. If this is the case, the number of the bound states isso large as compared to the number of unbound states that the branchingratio for the decay to unbound state is very small and bound state entangle-ment can last for much longer time as usually. Although the lifetime of anindividual bound state need not increase, the time spent in bound states anddefining decoherence time become much longer than predicted by standardphysics.
b) If the join along boundaries bonds are sufficiently near to vacuum extremals, they indeed allow immense spin glass degeneracy with slightlydifferent gravitational interaction energies and the desired situation can beachieved.
c) This argument can be refined by using unitarity.
for the transitions to bound states is enchanged by the degeneraty of thebound states, probability conservation implies that the probability for theoccurrence of decohering decays is reduced correspondingly.
A rough order of magnitude estimate for the gravitational binding energy for a cubic blob of water (that is living matter) having size given by p-adiclength scale L(k) is Egr(cubic, k) ∼ ' 2−12725/2(k−137) Gravitational binding energy is larger than the p-adic energy 2π/L(k) forL(k = 179) ' .169 mm. In the range L(163) = 640 nm and L(167) = 2.56µm gravitational binding frequency varies between 1 Hz and 1 kHz, that isover EEG range up to the maximal frequency of nerve pulses. If the bindingenergy gives estimate for the lifetime of the gravitationally bound states,this might fit nicely with EEG energies in typical cell length scales! For k = 157 and k = 151 (the range from cell 10 nm-80 nm, microtubules are at the lower end of this range) the gravitational binding frequency corre-sponds to a time scale of 8.5 hours and 32 years respectively so that the timescales relevant for life are spanned by the Gaussian Mersennes. What soundsparadoxal is that short length scales would correspond to long time scalesbut this indeed follows from the inverse square law for the gravitationalforce.
One can perform a similar estimate for linear structures. Parametrizing the microtubular transversal area to be d = x2L2(151), L(151) = 10 nm,one has Egr(lin, k) = x5 × Egr(cubic, 151) This gives for L(k) ∼ 1 meter, the frequency of .1 × x5 Hz. The time scalevaries between 10/x5 seconds and 32/x5 years and certainly covers the timescale for human long term memories. Of course, this rough estimate involvesnumerical factor which can increase the upper bound.
Note that the increments of the gravitational energy between transitions between almost degenerate bound states are some fraction of the gravita-tional binding energy. Also the gravitational interaction energy associatedwith the classical em fields could contribute significantly to the density ofthe gravitational energy in TGD framework and tend to increase the over-all energy scale. The reason is that the gravitational constant associatedwith classical fields is roughly 108 times larger than the ordinary gravita-tional constant [B1]. Thus, if the energy of classical fields is more than10−8mp ∼ 10 eV per proton the classical field energy of, say, join alongboundaries bonds becomes significant factor. Since hydrogen ground statebinding energy is about 13 eV, this kind of energy density per atomic volumelooks quite reasonable in case of water.
TGD universe is quantum critical system in the sense that spacetime sheets representing magnetic and electric fields with arbitrary large sizesare present and correspond to two phases in equilibrium (compare with iceand water at melting point). Electric-magnetic duality is second funda-mental symmetry of quantum TGD. Magnetic flux tubes carrying constant magnetic field (in lowest order approximation) have as their duals spacetimeregions carrying electric fields (constant in lowest order approximation). Inbiosystems various electrets and magnetic flux tube structures are the con-crete realization of these two phases. Classical gravitational effects generatevacuum 4-currents near the boundaries of these structures serving as sourcesof magnetic resp. electric fields. The boundaries of these structures are sin-gularities of the classical gravitational fields and these gravitational fieldsare good candidates for generating gravitional MEs responsible for long termmemories.
Long term memory and gravitational MEs Interestingly, MEs (topological light rays) with fundamental frequencies withtime scale measured using year as a unit are needed in the mirror model oflong term memories (to remember event at a distance of T in past is tolook in mirror at a distance L = cT /2). The gravitational transitions be-tween huge number of almost degenerate spin glass states could be codedto the fundamental frequencies of MEs. In particular, structures with sizesslightly above cell membrane thickness, such as microtubules, could generatethese MEs as the topological correlates of graviton emission with frequencyequal to the increment of the gravitational binding energy in quantum jumpinvolved. Thus there would be a direct correlation with long term memo-ries and microtubules: microtubule conformations could code for long termmemories.
The mirror mechanism of long term memory has beautiful interpretation in terms of topological correlates for virtual graviton exchange with vacuum.
a) The light reflected in mirror corresponds to topological light rays assignable to gravitons and is reflected from the curved vacuum. Topologicalcounterpart of virtual graviton is emitted by (say) tubulin, absorbed byvacuum and and emitted again by vacuum, and finally absorbed by tubulin.
Curved vacuum acts as a mirror for gravitons and you see yourself in thismirror.
b) Why gravitons are the only possibility in time scale of years is simply that they interact so weakly that they can propagate light years before ab-sorbed by curved vacuum. Time scales come out correctly and microtubulesare known to be crucial for long term memories (Altzheimer's disease in-volves changes at microtubular level).
c) There are also genuine vacuum extremals interpretable as topological graviton rays. These graviton rays could reduce to vacuum MEs except inthe turning point. This would mean 'self-reflection' without scattering from background and interpretable as an absorption and emission of a virtualgraviton. In case of nonvacuum extremals, classical momentum conservationhowever requires that the topological graviton exchanges momentum withthe background spacetime surface and thus is mirrorred from it.
d) One could interpret the low energy topological graviton rays respon- sible for long term memory as a particular kind of 1/f noise accompanyingall critical systems, in particular TGD Universe, which can be regarded as aquantum critical quantum spin glass. Gravitonic 1/f noise would be emit-ted in the transitions between almost degenerate spin glass states and wouldbe kind of analog for gravitational brehmstrahlung.
e) Gravitonic MEs carry also classical em and/or Z0 fields. The require- ment that memory MEs interact weakly with the environment suggests thatZ0 MEs are in question.
If this view is correct, the time scales of long term memory at DNA level would correspond to very long time scales characterizing consciousnessat the level of species. As a matter fact, the gravitational binding energyassociated with L(139) ∼ .1 nm (atomic physics) corresponds to the age ofthe universe: perhaps this explains why Schr¨ oedinger equation applies to the description of atom. 1/R dependence of the gravitational interaction energywould explain why very short length scales code biological information aboutvery long time scales rather than vice versa.
Dark matter hierarchy and hierarchy of long term mem-ories The emergence of the vision about dark matter hierarchy has meant a rev-olution in TGD inspired theory of consciousness. Dark matter hierarchymeans also a hierarchy of long term memories with the span of the mem-ory identifiable as a typical geometric duration of moment of consciousnessat the highest level of dark matter hierarchy associated with given self sothat even human life cycle represents at this highest level single momentof consciousness. This section gives only a rough overall view about theimplications for understanding long term memory.
Dark matter hierarchy leads to detailed quantitative view about quan- tum biology with several testable predictions [D4]. The applications to livingmatter suggests that the basic hierarchy corresponds to a hierarchy of Planckconstants coming as ¯ h0, λ ' 211 for p = 2127−1, k = 0, 1, 2, .
[D4]. Also integer valued sub-harmonics and integer valued sub-harmonicsof λ might be possible [A1]. Each p-adic length scale corresponds to thiskind of hierarchy and number theoretical arguments suggest a general for- mula for the allowed values of Planck constant λ depending logarithmicallyon p-adic prime [A1]. Also the value of ¯ h0 has spectrum characterized by Beraha numbers Bn = 4cos2(π/n), n ≥ 3, varying by a factor in the rangen > 3 [A1].
The general prediction is that Universe is a kind of inverted Mandelbrot fractal for which each bird's eye of view reveals new structures in long lengthand time scales representing scaled down copies of standard physics andtheir dark variants. These structures would correspond to higher levels inself hierarchy. This prediction is consistent with the belief that 75 per centof matter in the universe is dark.
1. Living matter and dark matter Living matter as ordinary matter quantum controlled by the dark matter hierarchy has turned out to be a particularly successful idea. The hypoth-esis has led to models for EEG predicting correctly the band structure andeven individual resonance bands and also generalizing the notion of EEG[D4]. Also a generalization of the notion of genetic code emerges resolvingthe paradoxes related to the standard dogma [D6, D4]. A particularly fas-cinating implication is the possibility to identify great leaps in evolution asphase transitions in which new higher level of dark matter emerges [D4].
It seems safe to conclude that the dark matter hierarchy with levels labelled by the values of Planck constants explains the macroscopic andmacro-temporal quantum coherence naturally. That this explanation is con-sistent with the explanation based on spin glass degeneracy is suggested byfollowing observations. First, the argument supporting spin glass degener-acy as an explanation of the macro-temporal quantum coherence does notinvolve the value of ¯ h at all. Secondly, the failure of the perturbation theory assumed to lead to the increase of Planck constant and formation of macro-scopic quantum phases could be precisely due to the emergence of a largenumber of new degrees of freedom due to spin glass degeneracy. Thirdly,the phase transition increasing Planck constant has concrete topological in-terpretation in terms of many-sheeted space-time consistent with the spinglass degeneracy.
2. Dark matter hierarchy and the notion of self The vision about dark matter hierarchy leads to a more refined view about self hierarchy and hierarchy of moments of consciousness [D2, D4].
The larger the value of Planck constant, the longer the subjectively ex-perienced duration and the average geometric duration T (k) ∝ λk of thequantum jump.
Dark matter hierarchy suggests also a slight modification of the notion of self. Each self involves a hierarchy of dark matter levels, and one is led to askwhether the highest level in this hierarchy corresponds to single quantumjump rather than a sequence of quantum jumps. The averaging of consciousexperience over quantum jumps would occur only for sub-selves at lowerlevels of dark matter hierarchy and these mental images would be ordered,and single moment of consciousness would be experienced as a history ofevents. One can ask whether even entire life cycle could be regarded as asingle quantum jump at the highest level so that consciousness would not becompletely lost even during deep sleep. This would allow to understand whywe seem to know directly that this biological body of mine existed yesterday.
The fact that we can remember phone numbers with 5 to 9 digits sup- ports the view that self corresponds at the highest dark matter level tosingle moment of consciousness.
Self would experience the average over the sequence of moments of consciousness associated with each sub-self butthere would be no averaging over the separate mental images of this kind, betheir parallel or serial. These mental images correspond to sub-selves havingshorter wake-up periods than self and would be experienced as being timeordered. Hence the digits in the phone number are experienced as separatemental images and ordered with respect to experienced time.
3. The time span of long term memories as signature for the level of dark matter hierarchy Higher levels of dark matter hierarchy provide neat quantitative view about self hierarchy and its evolution. For instance, EEG time scales corre-sponds to k = 4 level of hierarchy and a time scale of .1 seconds [D2], andEEG frequencies correspond at this level dark photon energies above thethermal threshold so that thermal noise is not a problem anymore. Variouslevels of dark matter hierarchy would naturally correspond to higher levelsin hierarchy of consciousness and the typical duration of life cycle wouldgive an idea about the level in question.
The level would determine also the time span of long term memories as discussed in [D4]. k = 7 would correspond to a duration of moment ofconscious of order human lifetime which suggests that k = 7 corresponds tothe highest dark matter level relevant to our consciousness whereas higherlevels would in general correspond to transpersonal consciousness. k = 5would correspond to time scale of short term memories measured in minutesand k = 6 to a time scale of memories measured in days.
The emergence of these levels must have meant evolutionary leap since long term memory is also accompanied by ability to anticipate future in the same time scale. This picture would suggest that the basic differencebetween us and our cousins is not at the level of genome as it is usuallyunderstood but at the level of the hierarchy of magnetic bodies [D6, D4]. Infact, higher levels of dark matter hierarchy motivate the introduction of thenotions of super-genome and hyper-genome. The genomes of entire organcan join to form super-genome expressing genes coherently. Hyper-genomeswould result from the fusion of genomes of different organisms and collectivelevels of consciousness would express themselves via hyper-genome and makepossible social rules and moral.
Model for long term memories In the following an attempt is made to understand how long term memoriescould be realized at neuronal level. I hope that my fragmentary knowledgeabout the details of brain science would not mask from the reader the beautyand simplicity of the general mechanism. The model is constructed first atgeneral level and then basic facts about long term memory are discussed inthe framework of the model.
In TGD framework one can make a precise distinction between genuinememories and apparent memories such as procedural and implicit memo-ries, associations, feature recognition, and standardized neuronal 'features'serving as building blocks of memories. The basic question is whether therepresentations of the long term memories are realized in the brain geo-metrically now or in the brain of the geometric past. In TGD the latteroption is allowed by timelike quantum entanglement made possible by thenon-determinism of K¨ ahler action. The very fact that the memory storage of past memories to the geometric now is not needed, means that there is noneed to carve long term memories to associative structures so that geomet-ric now would contain representations about moments of the geometric past.
Only the representation of the event at time when it occurred is needed. Forexample, this implies that long term potentiation (LTP) is just learning andadaptation to a new situation and can only be related to the modification ofmemory representations and possibly the construction of new standardizedfeatures.
Mirror mechanism is the simplest quantum mechanism of episodal memoriesand involves only a sharing of mental images by entanglement. The brainhemisphere sends a negative energy ME to the geometric past reflected at alarge distance and returning back to the hemisphere and induces a sharingof mental images. The desire to remember something and the memory of thepast fuse to a single mental image shared by the brains of the geometric pastand now. The desire to remember would be communicated to the geometricpast also in case of non-episodal memories whereas memory itself would becommunicated classically by positive energy MEs.
In a more realistic situation multiple reflections for a curvilinear negative energy ME along a closed magnetic flux loop would occur and guaranteeprecisely targed communications to the geometric past. The sizes of theseloops would be measured in light years.
Z0 Mes and Z0 magnetic flux loops associated with the personal Z0 magnetic body are the most realisticcandidates since in this case the interaction with matter is minimized.
The notion of memory field supports this idea.
Retrograde amnesia leads to a selective loss of memories in some time interval, and the notion ofmemory field provides a possible explanation. This means that brain struc-tures with a given memory field entangle with those events of the geometricpast which are located in some time interval ∆T at temporal distance T inthe past. A closed Z0 magnetic flux tube with a given length L(T ) wouldobviously be a correlate for a memory field with a given time span T .
The sharing of mental images mechanism requires only that gravitational/Z0 MEs take care of only quantum entanglement and because it allows arbi-trary kinds of episodal long term memories. The electric stimulation ofneurons can induce complex episodal memories. This can be understood ifthe episodal memory recall involves only the entanglement by the negativeenergy ME and the field pattern associated with ME does not matter atall. The unique experimental signature of the quantum entanglement mech-anism is that no direct correlates for the memories themselves are necessaryin the brain geometrically now. One can wonder what distinguishes the re-sulting experience from precognition by the self of the geometric past: couldit be that to precognize now is to remember in the geometric future? The direct sharing of sensory experience is non-economical in the sense that the amount of the irrelevant information is very high. The concep-tualization involved with the symbolic representation allows to representonly the absolutely essential aspects. In case of classical communicationssymbolic representations is of course the only practical possibility. Since the brain of the geometric past serves as a passive entangler and does nothave the possibility to process the communicated information, the sharingof the mental images is not flexible enough and does not allow an activeprecisely targeted memory recall. It is also very difficult to tell whethersensory experience represents memory or a genuine experience.
Classical communications and non-episodal memories For non-episodal memories classical communication mechanism suggests it-self as a more appropriate mechanism. Classical signalling requires the cod-ing of the data to the shape of the field pattern propagating along positiveenergy ME, which could be curvilinear and analogous to a radiation propa-gating in a wave cavity defined by a magnetic loop of the magnetic body.
MEs are indeed optimal for the coding of the classical signal since the vacuum current for given moment of geometric time is non-deterministic.
Classical communications would allow and also require the minimizationof the data communicated. These memories would not be sensory unlessback-projection to the sensory organs is involved at the receiving end. Theformation of the symbolic representation is subject to errors: for instance,temporal order of events can change. It is known that declarative memoriescan often involve changes of the temporal order. It must be emphasizedthat declarative need not be synonymous with non-episodal. Declarativememories could be also episodal and correspond to sharing of a symbolicmental images of the geometric past.
A dramatic reduction of the phase velocity is required, and is also as- sumed to occur for Z0 MEs in the model of nerve pulse and it could occursfor EEG MEs because of the interaction with brain. The phase velocities ofEEG waves in brain provide a good estimate for the effective phase velocitiesof MEs. The mechanism reducing the velocity is quantal: positive energyME drifts to the direction of geometric future quantum jump by quantumjump and this reduces the effective phase velocity. Thus it would seem thatclassical communications might rely on positive energy EEG and ZEG MEsin EEG frequency range. Z0 cyclotron frequencies are quite generally inalpha band so that Z0 MEs could be responsible for the communication ofthe long term memories using memetic code whereas EEG MEs might be re-sponsible for various sensory representations at the personal magnetic bodyand even magnetosphere. Note that the "features" of Freeman  havingduring of about .1 seconds are good candidates for the representation of theclassical signals. If EEG MEs are involved, the modulation of hippocampaltheta frequency is a candidate for the representations of classical signal.
There are are two basic options for how the classical communication could occur.
a) Positive energy ME would not leave brain at all and would therefore have ultra slow effective phase velocity along the brain structure in question,say axon, so that it would not leave brain during its travel to the geometricfuture.
b) Positive energy ME would be curvilinear and parallel with magnetic flux loop of the personal magnetic body serving effectively as a wave guide.
In this case the reduction of the phase velocity to EEG wave phase veloc-ity would be enough. For instance, for the phase velocity of alpha wavespropagating along loops with the size of the order of the Earth's circumfer-ence, the time span of the memory would be of the order of one year. Inthis picture one of the functions of the part of EEG and ZEG representingevoked responses could be classical communications making possible non-episodal memories. Only part of these memories would be conscious to us.
The length of the magnetic loops is expected to directly correlate with theperiod of EEG frequency involved with the classical communication via therelatioship L = vT would provide a second correlate for the notion of thememory field. There are indeed reasons to expect that the structures com-municating signals to the geometric future are specialized to communicatesignals to a certain distance.
The most plausible neurophysiological excitations associated with the received signal are Ca++ waves known to have extremely wide velocity spec-trum. For the option a) the required velocity would be of order neuronalsizes per year, and this is perhaps unrealistically low velocity. It is alsodifficult to see how the neuronal noise would not spoil the signal. For theoption b) the positive energy ME entering brain at the moment of memoryreceival would induce Ca++ waves in turn inducing neural activity.
For classical signalling the transformation of the classical signal to a con- scious experience is needed. Positive energy Z0 MEs could directly generatemembrane oscillations and nerve pulse patterns via the general mechanismof nerve pulse and EEG discussed in [D5]. EEG MEs could in turn inducecyclotron transitions at the magnetic flux tubes of the Earth's magneticfield and induce modulations of Z0 MEs in turn affecting nerve pulse gener-ation. Also a transformation of the signal to Ca++ waves could be possible.
The conscious experience does not involve sensory component unless thereis back-projection to the level of sensory organs involved.
Interesting questions relate to the interpretation of the ultraslow effective phase velocity of MEs acting as bridges connecting two space-time sheets.
a) The classical fields from a larger spacetime sheet A can be transferred to a smaller spacetime sheet B topologically condensed on A by inducing themotion of the wormhole contacts, which in turn generate classical fields atthe smaller space-time sheet. The fields can also penetrate along join alongboundaries bonds connecting the boundaries of two space-time sheets.
b) Quite generally, the "topological" half of Maxwell's field equations implies that tangential component of E and normal component of B are con-tinuous at the junctions connecting the boundaries of two space-time sheets.
One could assume that quantum effects can be modelled phenomenologicallyby introducing the phenomenological D and H fields introduced also in theMaxwell's theory. In the Maxwell's theory the discontinuity of the normalcomponent of the D field equals to the density of the free surface chargesand the discontinuity of the tangential component of the H field equals tothe free surface current. These conditions can be assumed also now, at leastas the first approximation.
c) One could model the propagation of MEs topologically condensed at a spacetime sheet labelled by a p-adic prime p ' 2k, k prime or power ofprime, by introducing the di-electric constant (k) and the relative perme-ability µ(k) satisfying the condition (k)µ(k) = 1/v2 > 1/c2 = 1, wherev is the effective phase velocity of ME depending in general on its funda-mental frequency. The fields D and H would be defined as D = (k)E,H = B/µ(k): this condition generalizes to that for the Fourier componentsof the fields. The reduction of the effective velocity for the propagation ofthe topologically condensed MEs to say alpha wave phase velocity does notseem plausible.
d) The propagation of MEs which serve as bridges between boundaries of two spacetime sheets (say cell membrane spacetime sheet and cell exteriorspacetime sheet) must be modelled differently. One could introduce a gener-alized di-electric constant (k1, k2) and permeability µ(k1, k2) characterizingthe pair of spacetime sheets such that the effective phase velocity v(k1, k2)of MEs acting as bridges satisfies (k1, k2)µ(k1, k2) = 1/v2(k1, k2), and alsonow depend on the fundamental frequency of ME. A very large value of(k1, k2) implying the needed very small value of the effective phase velocitywould mean that the orthogonal component of the electric field does not ap-preciably penetrate inside ME from either spacetime sheet. Since MEs arethe fundamental topological field quanta, this looks a natural assumption.
The extremely low effective phase velocity should be due to the replacementof the wormhole contact coupling with the join along boundaries couplingcausing the "stucking" of MEs. Note that the join along boundaries cou-pling is topological sum coupling for boundaries whereas wormhole contactsrepresent topological sum coupling for interior.
Furthermore, join along boundaries contacts can have a macroscopic size whereas wormhole con-tacts are CP2-sized: this could explain the huge reduction of the effectivephase velocity for the boundary MEs.
Negative energy Z0 MEs as ideal entanglers with the geo-metric past Gravitational, or equivalently Z0-, MEs with negative energies are especiallyfavoured for quantum communications. The reasons are many-fold. Theinteraction with the matter is very weak in long length scales but strong incellular length scales, negative energy implies that ME is identifiable as avirtual particle and analogous to a part of a Feynmann diagram so that nodissipation is involved and quantum communication is possible. The reversalof the arrow of geometric time means also that there is not macroscopicdissipative dynamics which would spoil the quantum coherence.
The requirement that the entanglement process is highly selective sug- gests a resonance mechanism.
This requires that receiving and sending structures are similar and generate ULF MEs with fundamental frequen-cies measured typically in cycles per year. If negative energy energy ME isin question, as suggested by the idea that a classical communication to thegeometric past is involved, it cannot be emitted unless there exists a receiverabsorbing the negative energy and in this manner providing energy for thesender by buy now-let others pay mechanism. For negative energy MEsresonance mechanism plus a simple classical signal serving as a a passwordcould also guarantee that correct part of the brain receives the signal.
Negative energy MEs represent time reversed level of the p-adic length scale hierarchy so that the dissipative effects associated with the space-timesheets with the normal arrow of time should not interfere with the quantumcommunication. This at least, when the energy of the negative energy MEhas a magnitude larger than the thermal energy associated with the space-time sheets with which it interacts: there is simply no system which couldmake a transition to a lower energy state by the absorption of a negativeenergy ME. Furthermore, since the systems with reversed arrow of geometrictime are expected to have extremely low density, the dissipative effects inthe reversed direction of time are expected to be small.
Since the generation of negative energy MEs does not require energy feed, the memory recall to the geometric past occurs more or less spontaneously,and the scanning of the geometric past becomes possible. The intentionalityof the memory recall would be realized as generation of a p-adic Z0 trans-forming to a negative energy Z0 ME, when the real system jumps to a higher energy state. This process makes possible precisely targeted intention alsoin the case of memory recall since the transitions in question cannot occurspontaneously. In the case of precognition precognizer must intentionallyreceive negative energy MEs from the geometric future so that energy feedis needed. This perhaps explains why precognition is so rare. Note that p-adic variant of pre-cognition having interpretation as intentionality occurseasily since p-adic energy is conserved only in a piecewise manner.
The most often needed non-episodal memories, say short term memories, could be communicated automatically: in this case the memory recall wouldbe a geometro-temporally local operation, much like taking a sample from adata stream representing particular kind of memories with a particular timespan. The option is probably not realized for all non-episodal memoriessince this would require large energy expenditure.
Is the right brain hemisphere the quantum entangler? There are some reasons to suspect that the quantum communications withthe geometric past occur more dominantly in the right brain hemispherewhereas classical communications would occur in the left hemisphere. Thiswould explain among other things the holistic aspects of right brain con-sciousness. Left brain hemisphere is specialized more to symbolic processingof information and would indeed be more suitable to classical communicationof this information.
Clearly, right brain would be passive receiver whereas left brain would be active expresser. DNA strands would be an example of this dichotomy atmolecular level. This dichotomy would be realized also at the level of geneexpression using em and Z0 MEs as the model of biophotons involving inessential manner negative and positive energy MEs suggets. Of course, thisstatement must be take only in the spirit of fractality and wuold hold trueonly in certain range of p-adic time scales.
The following arguments lend some support for the proposed division of labour between right and left brain hemispheres.
Synesthesia as a key to the mechanism of episodal memory What forces brain region to send negative energy Z0 MEs and thus to re-member? "Hunger!" is the possible answer! During synesthesia the metabolismin the left cortex is reduced by by 18 per cent due to the abnormally highmetabolism in memory circuit (for the model of synesthesia see [C1]). Per-haps the generation of the negative energy Z0 MEs is forced by the starvation of the neurons of the left cortex induced by the over-activity of the neuronsof the memory coordination circuit. The starving cortical neurons of the lefthemisphere would send massive amounts of negative energy Z0 MEs to thedirection of the geometric past inducing entanglement bridges by the mirrormechanism with the brain of the geometric past in turn inducing episodallong term memories by the sharing of the mental images. Thus the mirac-ulous ability of synesthetes to remember episodally could be understood toresult as a by-product of a neuronal emergency reaction.
There are good reasons to expect that same mechanism might be at work also in the normal situation but involve a less dramatic artificial starvationof the neurons of the right brain hemisphere. Clearly, the role of hippocam-pus is dramatically different from what is usually believed and also forcesto question the naive belief that neuronal activity is a measure of the con-tribution of brain area to the conscious experience. While building longterm memory representations as classical signals hippocampus and memorycircuit would steal energy from certain areas of cortex, and the resultingmetabolic starvation would force them to send negative energy MEs to gainenergy in this manner. This in turn would lead to the generation of longterm episodal or non-episodal memories as a side product. Quite generallyit is known that limbic brain and cortex tend to work in complementarymodes: when the cortex is in a high state of arousal, limbic brain is in astate of low arousal and vice versa. Perhaps the passive brain region isinvolved with memory recall and the active one with the construction ofsensory or memory representations.
Left-handedness and episodal memory It is known that persons with many left-handed family members have betterability for episodal memory recall and that this probably relates closely tothe communication between left and right hemispheres. We begin to haveverbal memories only after the age of four: at this time also the connectionbetween right and left hemispheres has matured. The proposed mechanismof non-episodal memories requires that the right brain hemisphere sharesthe mental image representing the desire to remember and the left brainhemisphere communicates the memory classically. Als the communicationbetween right and left hemisphere is necessary for this process to occur.
Children before the age of four could live in a kind of a dream time expe-riencing mostly sensory episodal memories and presumably not being abledistinguish memories from genuine experiences. This would also explainwhy we do not have declarative memories dating to the time before the age How could one understand the tendency of persons with many left- handed family members to have better episodal memory recall? The abilityto have sensory memories can appear also when a damage occurs to theregions of the left hemisphere. It could be that classical communicationsbetween the hemispheres are worser than usually when episodal memoryrecall is favoured, and are replaced by quantum communications. The men-tal images in the left brain hemisphere would entangle with those in theright hemisphere entangling in turn with the geometric future and give riseto episodal memories. Thus the quantum communications between hemi-spheres might be better than usually. This kind of persons would be more"holistic" than ordinary persons.
NDEs and long term memories That negative energy Z0 MEs could be responsible for episodal long termmemories is supported by near death experiences. Persons having near deathexperiences are clinically dead: in particular, EEG is absent. If these personsindeed have conscious experiences and if they are able to remember themas it seems, and since EEG signals are out of question, only the Z0 MEsgenerated during NDE remains as a viable alternative in TGD framework.
Brain or possibly body should be involved with the receival of geometricmemories if spin glass degeneracy is essential for the entanglement by Z0(gravitational) MEs.
Life review is one important aspect of the NDE experiences: entire 4- dimensonal body is experienced simultaneously. The starvation of neuronsforcing them to generate negative energy Z0 MEs could explain the episodalmemory feats of synesthetes and the eidetic memory, and would naturallybe at work also during NDE experience. This is not the only possibility.
This experience might also be partially due to the absence of the dominat-ing p-adic-to-real phase transition changing intentions to actions. This lifereview memory could be interpreted as geometric memories not masked bythe normal contributions to the contents of consciousness. An interestingpossibility is that this contribution is generated by theta and delta bandsof EEG during lifetime and is present also normally but, being stronglymasked, is not recognized.
Dejavu experiences and memory feats Dejavu experiences provide a challenge for any realistic model of memory.
In Dejavu the sensory experience is accompanied by the feeling 'I have ex-perienced this already earlier'.
A natural working hypothesis is that purely sensory memories, sensory re-experiences, do not contain information about the value of the geomet-ric time associated with the sensation. This means that sensory memoriescannot be distinguished from real experiences. On the other hand, cogni-tive and symbolic memories differ so radically from the sensory experiencesthat there is no difficulty of distinguishing them from genuine experiences.
Therefore one knows that the experience represented by this kind of memoryoccurred in geometric past or represents an expectation of future. Symbolic(real) and cognitive (p-adic) representations are very probably continuallytransformed to each other. If this view is correct, then the simultaneous oc-currence of the sensory and cognitive memories implies dejavu experience.
The event giving rise to the sensory and cognitive memories might haveoccurred only few seconds earlier.
This view has some nontrivial implications concerning the character of conscious experience of children. Cognitive abilities are thought to appearonly after the age of four or five years. If also symbolic memories are ab-sent, small children might live in a kind of dream time, as also members ofprimitive cultures, such as aboriginals, are believed to live in. Also dreamconsciousness could involve in an essential manner sensory memories as sug-gested by temporal acontinuity of dream consciousness. One could also seedreams as transformations of cognitive representations to sensory ones andsuch reverse to what occurs in wake-up consciousness so that surreal dreamlogic could basically result from p-adic non-determinism. The back projec-tion to the sensory organs would be an essential element of the mechanism.
The absence of a temporally organized consciousness would explain why we do not possess memories from the age before four. Perhaps also thebicameral consciousness, which according to Jaynes preceided modern con-sciousness, was kind of dream time consciousness in which memories weredirect sensory experiences, like voices experienced as voices of gods and vi-sual hallucinations. According to Jaynes, also schizophrenics are modernbicamerals.
Some time ago I saw a TV document about some autistic persons, who have very serious cognitive defects like inability count the number of objectsif it exceeds two, are capable of miraculous memory feats. One of thesefascinating individuals was an artist who could draw in full detail a picture about an area of London containing thousands of buildings after havin seenthe area once from a helicopter. Another autistic artist, virtuoso pianist,could reproduce every piece he had heard with highly personal style. Perhapsalso great musical wunderkinds like Mozart have had similar direct sensorymemory for music. Also a brain damage spoiling cognitive abilities can leadto the blossoming of exceptional artistic gifts. If the neuronal metabolicstarvation forces the generation of negative energy Z0 MEs in turn givingrise to long term episodal memories then one could indeed understand howbrain damage could have this kind of positive consequences.
The explanation suggesting itself is that the loss of cognitive memory is compensated by sensory memory in this kind of situations. A plausiblereason for why average human being has dominantly cognitive memoriesis simple. Sensory memory contains huge amounts of un-necessary data:symbolic and cognitive memories have much higher survival value since onlythe relevant data are stored.
Sensory genii have very hard time in the modern society unless they work as artists! In light of foregoing, the poor cognitive abilities of animals suggest that also animals remember predominantly sensorily and live in dream time (notehowever that rats have hippocampal theta). For instance, dogs might havesensory memory dominated by odours. The challenge is to invent tests forthis hypothesis. One could also try to device a non-destructive methodleading to a temporary loss of cognitive consciousness and making possibleto spend a day as a dog.
Going to the neuronal level The following attempt to develop the model of long term memory at theneuronal level is made involves many uncertainties and must be taken as anexercise in order to get accustomed with the ideas involved.
Which parts of the brain are the quantum entanglers? It is known that the electrical stimulation of amygdala, hippocampus, andtemporal lobes can generate lively sensory memories. The simplest explana-tion is that quantum entanglement with the sensory representations of thegeometric past is in question. The role of the electric stimulation would beonly the generation of time like entanglement, not providing any informationcharacterizing the memory. This would mean that large portions of braincan participate to the generation of episodal memories.
The fact that the part of body must be able to generate negative energy Z0 MEs with a proper ULF time scale, poses constraints on the systeminvolved. Cellular sub-systems and microtubules are good candidates in thisrespect since the transition frequencies for the transitions involving change ofclassical gravitational are in the required range. Since resonance mechanismis probably involved, there are good reasons to believe that similar system isis involved with both the receival and sending of the message. Microtubularstructures are good candidates adn accompany both neurons and glial cells.
Energetics poses also constraints. The receivers of negative energy MEs should have an easy access to the metabolic energy resources compensatingthe negative energy. In fact, the receiver must be in an excited state, whichdecays when negative energy Z0 ME is received (dropping ions to a largerspace-time sheet could be also involved). Glial cells serve as metabolic re-sources of the brain and interact with neurons via Ca++ waves and are thefirst guess for the system entangling with negative energy MEs. Other partsof brain and body, even sensory organs, can get metabolic energy by entan-gling with astrocytes via negative energy MEs so that the desired sharingof mental images would indeed result.
The notion of memory field  was derived from the study of short term memory and applies to the neurons of the frontal lobes at least. Thespan T of the memory field is essentially the time span of the long termmemory. T correlates strongly with the fudamental frequency associatedwith the negative energy M E if quantum entanglement is involved, and thelength of magnetic loop and curvilinear negative energy Z0 ME satisfiesL ∼ cT = c/f , where f is a frequency related to a transition in whichgravitational energy of the system is question changes.
When f is expressed in terms of the size of the water blob generating gravitational negative energy Z0 ME in spin glass transition this gives T ∝L−5, where L is the size of the water blob serving as a gravitational quantumantenna. Z0 MEs with T varying in the range 8.5 hours- 32 years in thelength scale range 80 nm-10 nm are generated. One day (24 hours) wouldcorrespond to a length scale 33 nanometers: 3.3 times the thickness of thecell membrane. In case of neurons only the intracellular structures havingmuch larger sizes and much higher gravitational binding energies might serveas entanglers (larger spacetime sheets would be in question) and give riseto short term memory. The time scale of 1 minute corresponds to about .3micrometers, millisecond corresponds to L(167) ' 2.3 micrometers, whereasL(163) corresponds to a time scale of 1 second. This would suggests thatsub-neuronal water blocks larger than the size of cell nucleus could generateshort term memories which need not be conscious-to-us. Perhaps the fluxloops of the magnetic body of the cell nucleus are involved.
For linear structures like microtubules one has T ∝ 1/L. Even in this case a rather strong dependence on the time span of the long term memoryon the system generating negative energy Z0 MEs results. The fact thatmicrotubules are ideal for representing conscious information symbolically,suggests that neuronal/astrocytic microtubules serve as the entanglers atsending/receiving end of the quantum communication line responsible forlong term memories. This picture also suggests that the Z0 magnetic fluxloop of a given astrophycal length scale is associated with a microtubule ofa given length.
Where the classical signals are generated and received? There are several bits of information helping to guess how long term mem-ories might be realized.
a) The damage of the hippocampus leads only to a loss of the ability to generate new declarative memories but does not lead to a loss of longterm memories from the period when hippocampus was intact. Thus itseems that hippocampus plays essential role in the communication of ournon-episodal declarative memories to the geometric past and that at leasta dominant part of the receivers are somewhere else than in hippocampus.
Since the stimulation of both amygdala, hippocampus and temporal lobesinduces long term episodal memories, it would seem that all these structurescan serve as quantum entanglers.
b) New neurons and glial cells are regenerated in hippocampus and the regeneration is especially intense during ischemia which can destroy a lot ofneurons [Fordahl]. This would suggest that both glial cells and neurons areessential for the realization of long term memories.
These pieces of data give some guide lines in the attempt to build a more detailed model of long term memories.
a) The generation of classical signals requires metabolic energy and this suggests that the generation occurs as near as possible to energy resources.
Glials cells are known to be the providers of the metabolic energy. Syn-chronously firing neuron groups are accompanied by astrocytes forming gapjunction connected structures. For a long time it was believed that astro-cytes play only the role of passive energy storages but it has become clearthat there is signalling between astrocytes and neuronal groups based onCa++ waves. Astrocytes couple also strongly to sounds: for instance, itis known that very mild blow in head inducing sound waves can lead to aloss of consciousness. Perhaps the astrocyte structures associated with hip-pocampal neurons generate positive energy MEs responsible for the classical communications making our non-episodal memories possible.
b) The receival of the classical signal does not require metabolic energy.
If astrocytes are involved with the sending of the classical signal, then neu-rons would be naturally the receivers of the signal and the energy receivedwith the signal would partially explain why synchronous firing of neuronalgroups seems to require less metabolic energy than expected. Of course,quantum entanglement by negative energy MEs wither energy sources couldalso explain this.
Is memetic code used to code declarative long term mem-ories? Memetic code is a good candidate for the coding of declarative long termmemories. The duration of single memetic codeword would be about .1seconds and the duration of a single bit would be about 1 millisecond. Thishypothesis fits nicely with the facts that the Z0 cyclotron frequencies arearound 10 Hz; that the frequency of neuronal synchronal firing is aboutkHz; and that there seems to be no direct electromagnetic counterpart forthe synchronous firing.
Quite recently it became clear that TGD predicts counterpart of Tesla's scalar waves [C5, C6]. These waves represent a pulse of electric field propa-gating with a velocity of light and an electric field in the direction of propa-gation. These waves corresponds in TGD to spacetime sheet of finite lengthand duration (L = cT ) carrying constant electric field and propagating withvelocity of light to the direction of the field. This solution type is extremelygeneral and dual to the magnetic flux tubes. Electrets (and also zelectrets)are one manifestation of these structures in living matter (membrane poten-tial is one example of this kind of structure).
One could consider the hierarchy of Z0 MEs representing geometrically a hierarchical structure of commands and that memetic code corresponds tothe lowest level with bit represented by an zelectric pulses whose polaritydetermines whether '1' or '0' is in question: very much like in case of com-puters. Zelectret sequences would ultimately give atomic nuclei kicks in adirection depending on the value of the bit.
What about other synchronous EEG frequencies? Genuine theta (hippocampal theta which spans both theta and alpha bands)and delta bands could correspond to more abstract levels of consciousnessnot directly experienced by us usually. During slow wave sleep theta and delta bands dominate and the interpretation in terms of the binding of themental images to memory representations is highly suggestive. Hence thesebands would contribute to our consciousness in the geometric future ratherthan in the geometric now.
a) Theta band might relate to long term memory consolidation by a con- struction of temporal replicas of ordinary long term memory representationsgenerated already during the wake-up period. Sleep state is certainly idealin this respect.
b) Naive extrapolation suggests that delta band memories correspond to a rather long temporal distance T (that is very low frequency f = 1/Tfor gravitonic MEs). Delta band memories would be therefore generatedby structures with sizes below the thickness of cell membrane. One couldunderstand why delta band is strongest in childhood and weakens towardsold age. If delta band memories correspond to very long temporal distancesT , it is useless to generate these memories at the old age since there wouldbe no brain receiving these memories. The long time span of the deltaband memories would explain why childhood memories are stable and whysome persons 'return' to their childhood at the old age. The return to thesensory world of childhood at old age suggests that delta band memoriesmust be sensory memories. Delta band representations might even give riseto transpersonal memories experienced during the later lives. The absenceof ordinary sensory input masking delta band memories would explain whyearlier life cycles can be recalled in meditative states.
c) The contribution of theta and delta band memories to our conscious- ness could also relate to the third person aspect of consciousness. Thetaand delta waves could be associated with the magnetospheric sensory rep-resentations giving rise to multibrained selves. The entanglement betweensleeping brains inducing a loss of personal consciousness would induce a kindof collective stereo consciousness in which a large number of invididual viewsabout world fuse together would be in question. The search for correlationsbetween the EEGs of sleepers having a close personal relationship mightbe rewarding. For istance, DNA could quantum entangle and give rise toconscious memories in very long time scale at the level of species.
Note that the presence of synchronous or asynchronous EEG correlate of memory generation is present also during memory recall does not seemto be necessary since the memory is indeed in the geometric past.
One important question is whether positive energy EEG MEs are involvedwith long term memories or only with sensory representations (assumingthat sensory representations are realized at the magnetic body). The ideathat Z0 MEs take care of memories and EEG MEs of sensory representationsis attractive idea at least.
Fascinating questions relate to cognitive representations since these in- volve p-adic physics. Frontal lobes are known to be the seat of planning,volition and cognition. Therefore p-adic cognitive representations, p-adicentanglement and the p-adic selves characterized by positive entanglementnegentropy should be realized in the neural circuits involving frontal lobes.
These circuits have been even proposed to be 'conscious circuits' but thisprobably reflects the erranous identification of consciousness as cognitiveconsciousness only. Cognitive representations could be realized at magneticcognitive canvas using beta frequencies as resonant frequencies and betaMEs would entangle with the points of the cognitive magnetic canvas p-adic mental images representing intentions and plans. The transformationof these p-adic mental images to real ones would somehow generate general-ized motor actions, in particular ordinary motor actions. That frontal lobescontain motor areas conforms with this view.
Hippocampus and long term memories The findings about hippocampal system provide a good test for the generalideas about long term memory. For a review about the role of hippocampusin long term memory see .
Anatomy of hippocampal system The anatomy of hippocampus is discussed in : here only very rough sum-mary is given: possible inaccuracies are due to my amateurish knowledge ofbrain science.
Hippocampus is located with the inferior medial wall of the temporal lobe posterior to the amygdala. Hippocampus decomposes into anterior andposterior regions. Hippocampus consists of a number of subcomponents,and adjoining structures, such as the parahippocampal gyrys, entorhinal andperirhinal cortex and uncus. The main body of the hippocampus consistsof the detante gyrus (here brain cells are regenerated), the subiculum andthe sectors referred to as CA1,CA2,CA3 and CA4. The uncus is a bulbar allocortical protrusion located in the anterior-inferior medial part of thetemporal lobe.
There are three major neural pathways leading to and from the hip- These include the fornix-fimbrial fiber system, and a supra- callosal pathway which passes through the cingulate, and via the entorhi-nal area: this is the mesocortical gateway to the hippocampus. Throughthe fornix-fimbrial pathways hippocampus makes major interconnectionswith the thalamus, septal nuclei, medial hypothalamus, and exerts eitherinhibitory or excitatory influences on these nuclei.
The entorhinal cortex acts to relay information to and from the hip- pocampus. The hippocampus maintains via the entorhinal cortex intercon-nections with the neocortical multi-modal associations areas of the tem-poral, frontal, and parietal lobes, including surrounding structures, e.g.,the parahippocampal gyrus, and allocortical tissues, the perirhinal cortex,septal nuclei and amygdala. The parahippocampal gyrus, entorhinal andperirhinal cortex, being directly interconnected with the hippocampus andthe neocortex, act to relay input from the neocortical association areas tothis structure.
The entorhinal cortex consists of 7 to 8 layers rather than only 6 layers.
The entorhinal cortex maintains massive interconnections with all multi-modal neocortical association areas (as well as with the amygdala, hip-pocampus, septal nuclei, olfactory bulb, etc.) but none of the primary sensory areas which presumably relates to the fact that hippocampus isresponsible for declarative rather than sensory memories.
Memory deficits and hippocampus Memory deficits provide important information about the role of hippocam-pus with respect to the memory. In anterograde amnesia the ability to gener-ate new long term declarative memories is lost and it is known that a damageto the hippocampus can cause this defect. Thus it seems that hippocam-pus is crucially involved with the construction of long term memories. Alsothe damage to the medial temporal lobes and subcortical structures such asmedial thalamus and mammillary bodies can destroy the ability to generatelong term memories. This supports the view that hippocampus is kind ofa central entangler binding together mental images from various parts ofbrain: most naturally entanglement occurs along the three neuronal path-ways going through hippocampus and presumably associated with toruslikemagnetic flux tubes.
In retrograde amnesia memories about some period of time in past are lost. It seems that this deficit does not correlate with the damage of hip-pocampus. Thus the cautious conclusion is that long term memory recalloccurs also elsewhere in brain. The selectivity of the retrograde amnesia sug-gests that the notion of the memory field applying in the case of short termmemory  generalizes. The brain structures responsible for the receival oflong term memories are specialized in the sense that they entangle with themental images of the geometric past located only in an interval around cer-tain temporal distance T . If the memories involve only few reflections alonga closed magnetic flux loop, the corresponding Z0 (gravitonic) MEs havefundamental frequency f = 1/T and correspond to spin glass transition formicrotubules or for 3-dimensional sub-neuronal structures at a length scalebetween cell size and cell membrane thickness if the simplest estimate makessense. This kind of resonant selectivity might be possible to achieve if thereceiving system is driven to the bottom of the spin glass landscape witha depth which corresponds to the gravitonic energy E = 2πf If memoriesinvolves large number of reflections, it is difficult to imagine, how this kindof selectivity could be achieved.
Hippocampus and declarative memory It is known that there are several memory types and hippocampus is re-sponsible for the construction of only declarative memories, which are ver-bal and highly symbolic representations of the geometrical aspects externalworld. Hippocampus is not essential for the recognition of familiar objectsnor for procedural/motor memories which are implicit memories. The nat-ural identification of declarative memories is as memories communicatedclassically using some coding but one cannot exclude sharing of mental im-ages. Memetic code or its scaled up/scaled down is a good candidate in thisrespect. The modulation of hippocampal theta might provide the coding.
Sensory memories can be induced by the electric stimulation of both hippocampus, amygdala and temporal lobes. This suggests that lower lev-els of self hierarchy which we do not experience directly can have sensorymemories. The entanglement by negative energy Z0 ME with the geometricpast giving rise to an episodal memory is the most natural interpretation forthe effect. Neural loops are the geometric correlates for entanglement at thelevel of CNS, and timelike quantum entanglement of parts of the electricallystimulated structures with primary sensory areas with the mediary of theseloops should be involved. If the stimulation is too strong, hallucinationsresult. In this case the sensory representations in thhe brain geometricallynow are presumably activated and back projection to the sensory organs would occur. An interesting possibility is that the strength of stimulationcorrelates with the temporal distance of the sensory representation in thegeometric past activated in the stimulation.
Hippocampus provides spatial and temporal context The right hippocampus of the taxi drivers in London is enlargened. Thissupports the view that hippocampus provides kind of a symbolic map ofthe spatial layout of the environment. Studies in animals suggest that hip-pocampus adds a spatial context to the mental images from cortex entangledwith mental images in subhippocampal structures entangled with the men-tal images in hippocampus. The spatial map is based on various spatial cuesserving as landmarks. Left hippocampus is in turn involved with the verbalmemories and this suggests that it is responsible for providing a temporalcontext and time ordering of events. This suggests that hippocampus is re-sponsible for the temporal and spatial organization of conscious experiencebesides generating memory representations. Perhaps a high level sensoryrepresentations at the magnetic body is in question.
Hippocampus is known to contain place cells providing cognitive repre- sentations for the objects of perceptive field. These place cells are pyramidalcells containing magnetic crystals which suggests that they act as projectorsto the magnetic memory canvas. All kinds of features could be associatedwith these landmarks, and more generally, with the symbolic objects of thememory field.
Long term potentiation (LTP) does not occur in hippocampus but hip- pocampus is highly dynamical with synaptic contacts being generated allthe time and even the size of hippocampus continually changing. It wouldseem that hippocampus provides by its own dynamical structure a contextfor various data coming from cortex, kind of a geometro-symbolic model forthe external world. The mental image associated with this model of exter-nal world quantum entangles with the mental images in cortex, amygdala,hypothalamus, etc.
Not only spatial but also temporal context is important and hippocam- pus should provide also this. Purely sensory memories do not carry anyinformation about whether memory is in question or not. For symbolic rep-resentations the situation is different. Symbolic representations would berealized as association sequences, perhaps in the time scale of hippocampaltheta such that each 3-surface of association sequence contains lower levelassociation sequences contains. Memetic code words of duration .1 secondswould be at the lowest level and perhaps correspond to mesoscopic features of Freeman .
The intronic portion of DNA could provide the fundamental hardwave representation of memes in terms of sequences of 21 DNA triplets: spokenlanguage would be only a tip of an iceberg if this picture is correct [D3].
Positive energy em and Z0 MEs could realize these memes in the shape ofvacuum current, which at given moment of time is non-deterministic andtherefore optimal in this respect. Memetic code realized in terms of Z0magnetization direction for cognitive antineutrinos is a further candidatefor realizing the symbolic representations. This highest level representa-tion adding context to the other data located in the geometric past wouldentangle via Z0 MEs with the brain of the geometric now in case of episo-dal memories. The fact that hippocampus is thought to be involved withthe trassfer of items in short term memory to long term memory in cortexconforms with the mirror mechanism.
Entorhinal cortex serves as somekind of a relay station between hip- pocampus and neocortex. Entorhinal cortex has very special structure be-ing 7-to-8 layered rather than 6-layered. Entorhinal cortex maintains richconnections to various multimodal regions in temporal, parietal and frontalcortices but not to the primary sensory areas. This is consistent with theidea about three-leveled hierarchy multimodal areas→ entorhinal → cortex-hippocampus, with the fact that the mental images associated with hip-pocampal memory representations are symbolic rather than sensory, andwith the assumption that multimodal areas, entorhinal cortex, and hip-pocampus entangle.
Hippocampal theta corresponds to EEG frequency range varying from about 4 Hz to 12-14 Hz and thus spans both theta and alpha bands. Hip-pocampal theta can be seen as a correlate for the binding of various corticaland subcortical mental images to a single mental image representing boththat aspect of consciousness which makes possible organized view aboutspace and time and declarative memory. MEs at hippocampal theta fre-quencies could project to the magnetic memory canvas providing an ab-stract representation about world analogous to sensory representation butwithout sensory qualia. It must be emphasized that the memory representa-tion should provide an essential part of our everyday consciousness makingpossible space and time categories of everyday conscious experience. Noveland painful stimuli indeed induce hippocampal theta as well as orientingreactions, learning, selection and discrimination.
Remote emotions and associations? Amygdala seems to be responsible for the formation of emotional aspectsof the memories in accordance with entanglement paradigm.
is known to be sensitive to emotional contextual cues which can triggerperceptive experiences similar to previous ones. Associative memories seemto be in question.
Whether the associative memory is in the geometric now or past is not obvious and timelike quantum entanglement might perhaps allow to induceremote associations in the geometric past. If the cue is entangled with thecue in the geometric past, the activation of this cue by quantum entangle-ment could activate neural process generating the memory in the geometricpast. This kind of mechanism would provide a general mechanism of activememory retrieval. The active scanning of memory neurons with memoryfields characterized by different values of T would be a second mechanism ofthis kind. In fact, there need not be any sharp difference between ordinaryassociations and associations in past.
Memory consolidation and long term potentiation The notions of memory consolidation and long term potentiation relate tothe more standard views about long term memory and it is interesting totry to interpretthem in TGD framework. Memory consolidation means thestrengthening of memories by 'replaying' them. Certainly a repetion of men-tal image provides a manner to learn and establishing a long term memoryalso in TGD. The mere generation of gravitational MEs associated with agiven mental image means consolidation: no modification of the existingneural connectivity is needed. Of course, standardized mental images areprobably generated but this is not construction of memories in the strictsense of the word.
Memory consolidation involves hippocampal theta. In TGD framework hippocampal theta is a correlate for that part of consciousness which givesrise to an organized view about space and time: not necessarily in the ge-ometric now however. Mirror mechanism implies that this process definesautomatically memory representations about the state of brain so that mem-ory consolidation is an automatic side effect.
It has been proposed that during REM sleep hippocampus is 'replaying' the memories unconsciously . The fact that there is no sensory input atnight time would suggest that sleeping brain is like an empty magnetic tapefreely usable for the memory construction. Theta and delta bands could re- late to the memory representations replayd during sleep period but could bealso responsible for the construction of higher level sensory representationsimportant for non-episodal memories.
There are however objection against the idea that REM sleep is special- ized with the replaying. First, hippocampal theta, believed to be crucial forthe formation of long term declarative memories, is not synchronous duringREM sleep. Secondly, during dreams only the posterior portion of the hip-pocampus is active whereas during learning the active part is the anteriorportion of the hippocampus.
TGD based vision suggests a first principle explanation for the activity of hippocampus during sleep and dreams. Both classical communicationsto the geometric future and the receival of negative energy MEs from thegeometric future require metabolic energy feed. Since the metabolism re-lated to the motor activity and sensory preception is absent during sleep,the optimal realization of the long term memories is based on the entangle-ment with the sleeping brain of the geometric past. This would also explainwhy we do not have conscious experiences about memory recalls from thegeometric future. Sleeping brain can also help the situation by performingmemory recalls itself. REM sleep would not be in any special role exceptthat it could make possible episodal sensory memories.
The memories about dream experience fade out rapidly after wake-up.
This suggests that the lengths of the magnetic flux tubes along which clas-sical communications occur during dreams, are short and therefore also thetime span of the resulting declarative memories is brief. This as it shouldbe since otherwise dreams would make possible pseudo memories. We couldbe conscious during dreams but would not remember it since long termmemories would not be generated during this period. Alternatively, dreammemory representations could be generated by the larger self to which we arefused during sleep. The above mentioned findings about the hippocampalactivity during dreams could mean that magnetic flux loops of declarativememory get longer in posterior-anterior direction: this would mean a con-crete identification for the neurophysiological correlates of the declarativememory fields. Also the dominating frequency of EEG/ZEG would becomelower in this direction.
The basic question relates to the intepretation of the hippocampal theta.
There are two options.
a) Hippocampal theta is associated with the EEG MEs responsible for theclassical communications to the geometric future making possible long termmemories.
b) Z0 MEs take care of the classical communications to the geometric fu- ture (memetic code) whereas hippocampal theta contributes to the consciousexperience of the geometric now by generating high level sensory represen-tations at the personal magnetic body.
For the latter option hippocampal theta could be also involved with the generation of entanglement between various parts of brain crucial forthe construction of long term memories making possible an organized viewabout space and time. This assumption conforms with the idea that EEGrhytms are responsible for the synchrony and entanglement. This would nothappen during REM sleep since hippocampal theta is asynchronous duringdreaming and during cortical synchrony (not much sensory input). Visualdream consciousness is indeed sensory consciusness without an organizedview about space and time categories. This applies also to the non-REMverbal dreams. Furthermore, the desynchronization of both hippocampaland cortical EEGs implies a confused state of mind. This would suggestthat hippocampus indeed contributes also to our consciousness in the geo-metric now, and makes possible the organized view about space and timeby constructing higher level sensory representations.
Long term potentiation (LTP) has been suggested as a mechanism by which hippocampus generates long term memories by strengthening thesynaptic communications between neurons. In TGD framework this inter-pretation does not make sense: rather LTP can be seen as a special case ofassociative learning which is just gradual modification of the brain structureas a response to the conscious experience. Of course, LTP modifies graduallymemory representations but these memory representations do not containinformation about past.
As noticed, LTP does not occur in hippocampus itself. Instead, hip- pocampus grows rapidly in neuron number and synaptic connections duringlong term memory generation. This conforms with the view that hippocam-pus is more or less a real time dynamical representation for what mightbe called changing context. In particular, new neurons generated in hip-pocampus could be essential in representing the context and could generategravitonic MEs crucial for the entanglement.
Relationship between cortical and hippocampal EEGs Cortical desynchronization accompanies hippocampal synchronization andvice versa. The simultaneous desynchronization of cortical and hippocam-pal EEGs involves distractability and hyper-responsiveness so that personbecomes owerwhelmed, confused, and may orient to and approach severalstimuli.
These findings can be understood in TGD framework.
a) During cortical asynchrony there are good reasons to build long term memories so that hippocampus should be in synchronized state and bindvarious mental images to long term memories.
b) During cortical synchrony there is nothing to represent as long term memories and hippocampus can do something else. Perhaps participate inimagination and day dreaming as suggested by the fact that also duringREM sleep hippocampal theta is asynchronous.
c) When both cortical and hippocampal theta are desynchorized, not only the long term memory representations fail to be generated but also theconstruction of spatial and temporal context and this leads to confusion anddifficulties with orientation to various stimuli.
Microtubuli and long term memory When I began consciousness theorizing whole-daily around about 1994, Ibecame deeply fascinated about microtubuli (as probably most others in thefield of quantum consciousness). I launched off by developing a rudimentarymodel about how microtubuli could act as quantum antennae in the TGDuniverse: massless extremals were the key element of the model. Needlessto say, too much of the general theory of consciousness and of biosystemsas macroscopic quantum systems needed for a deeper understanding wasunconscious-to-me at that time.
After the rapid self-organization of the theory during this year and still continuing (I am living last days of August 2002 while writing this), itoccurred to me that it might be a good idea to take a fresh look on therole of the microtubuli. While re-reading the wonderfully inspiring articleof Nanopoulos dating back to 1995 , I realized that the TGD based viewabout macrotemporal quantum coherence, the mirror mechanism of longterm memory, and the quite recent discovery of cognitive codes and theirphysical realization, provide the tools for developing a view about the roleof microtubuli in long term memory.
What made me somewhat skeptic about the importance of the micro- tubuli for our consciousness was the naive view that the size L of the systemsystem generating the memory increases when the geometrotemporal dis-tance T of the long term memory increases. Microtubuli would be consciousbut from our point of view this would represent something analogous to bitlevel in computers.
The understanding of how the macrotemporal quantum coherence is gen- erated however challenged this view. TGD Universe is quantum spin glass and spin glass degeneracy is broken only by the classical gravitational bind-ing energy. Quantum transitions between almost degenerate quantum spinglass states correspond to frequencies defined by the differences of the classi-cal gravitational binding energy and generate gravitational MEs responsiblefor the quantum mirror mechanism. Gravitational binding energy increaseswith the system's size and this means an effective inversion of the lengthscale hierarchy, so that systems like microtubuli can contribute to our con-scious experience much more signicantly than some subsub.subself level atthe bottom of the self hierarchy might be expected to do.
Basic findings about the correlation between long termmemory and microtubuli A basic difference between ordinary cell and neuron is that the microtubuliassociated with the T shaped centriole in case of the ordinary cell, are in neu-ron replaced by long microtubule bundles starting in a region near nucleusand connecting it to dendrites and axonal ends. The natural guess is thatat least these microtubuli are closely involved with the brain consciousness.
What happens in microtubuli is indeed very intimately related to what happens in synapses. The minimal modification of the standard neurosciencebelief system is that microtubuli control how synapses, still assumed to beresponsible for the memory representations, are modified during learningidentified as generation of long term memories. In  a lot of basic factsabout microtubuli plus the evidence for the correlation between microtubuliand long term memory is discussed and references can be found in thisarticle. Here I just summarize the basic points of the discussion of .
a) The production of tubulin and MT activities correlate with peak learn- ing, memory and experience in baby chick brains. Experiments with babyrats show that when they first open they eyes, neurons in their visual cortexbegin producing vast quantities of tubulin.
b) The experiments with trained goldfishes show that the drug colchicine produces retrograde amnesia. The interference with MTs responsible for thestructural modification of certain synapses is believed to affect memory fix-ation. In TGD framework one must carefully distinguish between learningand memory: microtubuli could provide both the long term memory repre-sentations and also control learning by controlling synaptic strengths.
c) The selective dysfunction of animal brain MTs by the drug colchicine causes defects in learning and memory which mimick the symptons of Alzheimer'sdisease (AD). It has been reported that in rats a continuous MT disruptioninduced by a chronic colchicine administration results in a dose-dependent learning deficit, and memory retention is also impaired. It has also beenstressed that these colchicine-induced cognitive defects resemble those ofAD, e.g., amnesia of the recent learning and loss of formerly establishedmemories. These findings encourage to think that that microtubuli are in-volved both with the generation of the memory representations and longterm memory recall by mirror mechanism in accordance with the idea thatmicrotubuli act as both receiving and sending quantum antennae in thesense that they generate MEs making possible timelikequantum entangle-ment. MEs generate coherent photons or gravitons according to the originaldefinition of quantum antenna [C7]. Certainly, the antenna which sends isalso optimal for receiving.
d) It has been suggested and also supported by detailed experimental studies that the impairment of MTs, leading to tangled and dysfunctionalneural cytoskeleton, may be one explanation for the pathogenesis of AD.
e) In specific hippocampal regions of the brain of schizophrenic patients, distorted neuronal architecture has been found due to a lack of 2 MAPs.
This suggests that the splitting of consciousness characterizing schizophrenyhas a geometric correlate already at the microtubular level: macroscopicbound state entanglement responsible for the binding to longlived holisticmicrotubular mental images and the generation of memory representationswould not occur as they should.
How microtubuli could relate to declarative long term mem-ories? For several reasons microtubuli are taylor-made for the realization of longterm declarative memories in TGD Universe (the structure of microtubuliis discussed in some detail in [C2], where the realization of cognitive codesis discussed). Microtubuli are however not the only candidates: also 2-Dmembrane like structures and genuinely 3-D structures could be involvedand correspond to different types of long term memories.
a) Microtubuli can entangle with each other and with the surrounding world in conformational degrees of freedom to yield macrotemporal quan-tum coherence. Also cognitive neutrinos could be present. Microtubuleassociated proteins (MAPs) can mediate naturally bound state entangle-ment between conformational patterns of different microtubuli. This makespossible macrotemporal quantum coherence and processes resembling quan-tum computation when bound states are formed. MAPs can act as switchesinitiating quantum computation and halting it. The simplest possibility isthat MAP protein becomes just disconnected at some levels of the hierarchy of spacetime sheets.
b) Tubulin dimers allow two different conformations and the patterns of tubulin conformations are ideal for binary representations of data naturalfor the representation of long term declarative memories. In [C2] a cogni-tive code explaining the numbers associated with microtubular geometry isdiscussed and a model for how the conformational patterns are coded intoconscious experience in the phase transition in which spontaneus electricpolarization occurs and forces all tubulin dimers to the ground state confor-mation. That microtubuli allow the realization of the symbolic counterpartsof cognitive representations realized using cognitive neutrinos and possiblyalso by p-adic MEs, conforms with the fact that colchicine which affectsMTs, induces cognitive defects characteristic of Alzheimer's disease. Thelinearity of microtubuli would be obviously essential and at least parts ofthe sensory pathways could be responsible for the representations of thesememories.
c) In the standard view about long term memories one cannot iden- tify microtubuli as seats of long term memory representations. The reasonis simply that microtubule conformations are quite too short-lived for thispurpose. This leaves only the identification of the synaptic strengths as arepresentation of long term memories. In TGD the situation is just the re-verse and flexibility requires fast enough dynamics. The time scale definingsensory resolution is obviously a bottle neck time scale. The time scale forthe phase transition leading to ground state of tubulin dimer in an externalelectric field and the time scale related to the control of the external elec-tric field at the microtubular spacetime sheet are the most obvious guesses.
The first time scale should be of order of the time scale of conformationaldynamics, about nanosecond. The latter time scale would be basically theduration of nerve pulse if nerve pulses are responsible for the phase transi-tion in question. In TGD framework the modification of synaptic strengthscan be more naturally seen as representing generation of new 'habit rou-tines' rather than memory representations which are much more involvedand information rich.
d) Microtubuli are ideal for quantum mirror mechanism of long term memories. As already found, in case of spherical structures the dependenceof gravitational binding energy on size of the structure is Egr ∝ L5, whereasthe gravitational binding energy depends on the length L of a linear struc-ture as Egr ∝ L. For membrane like structures Egr ∝ L3. Since microtubulelengths vary in the range 10 nm- 1 mm, this means that the temporal dis-tance T ∝ 1/L of long term memory varies between 32 years 2.8 hours (veryroughly; increase of the overall time scale due to the fact that increment of the gravitational binding energy in the transition is smaller than the gravi-tational binding energy itself). Inside axons microtubuli can bind to longerstructures by MAPs and even meter sized structures associated with sensorypathways are possible. This lowers the lower bound for the time span to 10seconds. The longest microtubuli are responsible for the representation ofthe shortest term memories realizable in this manner. Of course, memorycircuits should regenerate again and again microtubular memory represen-tation and in this sense synaptic strengths become an essential part of thememory representation.
e) Colchicine affects both memory recall and memory generation. This inspires the working hypothesis that microtubuli of a given length L ∝1/T in the geometric past entangle with a microtuble of same length inthe geometric now during memory recall. For instance, the receiver in thegeometric now could correspond to a postsynaptic microtubule whereas thesender in the geometric past corresponds to a presynaptic tubule. This isnot the only alternative, receiving cells could be even glial cells.
f) That the memories of childhood are the most stable ones could be interpreted as reflecting the fact the microtubuli act both as receiving andsending quantum antennae, and that the long microtubuli responsible forgenerating the short term memory representations and for receiving them de-teriorate towards the old age with much higher probability than the shorterones. It could be possible to induce selective amnesiae restricted to memo-ries with a temporal distance ∼ T by a treatment which affects microtubuliof given length ∼ L ∝ 1/T .
g) Microtubuli could be also ideal for the communication of non-episodal memories involving classical communication by ultra slow Z0 MEs perhapsaccompanied by Ca++ waves known to have an extremely wide velocityspectrum. Ca++ ions are associated with the outer surface of the micro-tubuli and dynamically comparable to a crop field in a wind. Ultra-sloworientational waves for these Ca++ ions representing sensory inputs andpropagating along axons could make possible a classical communication ofdata from the geometric past as declarative memories. For sensory pathwaysthe sequences of microtubuli could have a total length of order one meter.
For the average length L0 = 10 µm of the microtubule in brain, the timespan T0 = 10 seconds would give v0 ∼ 1µm/s, a typical velocity of in cellularlevel. In this case 10 nm length of microtubule would correspond to 10−2seconds of time. This would mean that roughly 13 parallel sequences of 13bits of information about 10 millisecond period. The bit rate of one bit permillisecond corresponds to the information storage capacity of the memeticcode. For longer time intervals T and microtubule lengths L the bit rate would scale like (L/L0) × (T0/T ) = v/v0. For T = 1 year and L = L0 onewould have roughly one bit per hour. It seems that this mechanism can beat work only for short term memories whereas long term memories wouldinvolve closed magnetic loops.
Relation to the general model of long term memories It is interesting to relate the proposed model with the general model of longterm memories.
a) Long term memory is lost when tubulins return to ground state un- less there is some mechanism regenerating the conformational state. In brainthe function of neuronal loops generating the nerve pulse patterns repeat-edly would take care of regenerating the memory representation. If this viewis correct, then also memories of childhood involve this kind of continual re-generation. Sensory pathways do give rise to long term memories unless thefeedback from brain to primary sensory organs (otoacoustic sounds and themovement of eyes during REM sleep) regenerates these memory represen-tations. During dream long term memories correspond to small value of T :does this allow to conclude that the feedback to the primary sensory organsduring dreams results in long term memories with T about few minutes?The maximation of the lengths of the sensory pathways (left side of thebody is connected to right brain hemisphere and vice versa) would relateto the maximization of the representational capacity if this mechanism is atwork. Notice that the continual regeneration of memories with say tempo-ral distance of T = 15 minutes does not seem sensical since these memorieswould not be received by that part of the 4-D brain which corresponds tothe p-adic-to-real phase transition front. The most natural assumption isthat sensory representations are regenerated for time interval of order T sothat the maximal values of T and stablest memories correspond to relativelyshort microtubuli in the interior of neuron.
b) Hippocampus is believed to be crucial for the generation of long term declarative memories and responsible for spatio-temporal organization ofperceptive field. Hippocampus could act as a kind of entanglement centerentangling with 'features' at various brain areas and project them to thesensory magnetic canvas (the episodal component representing spatial rela-tionships might accompany also non-episodal memories!). Feature subselveswould have microtubular selves as subselves: this would mean entanglementbetween hippocampal and other microtubular memory representations. Themicrotubuli acting as central entanglers in hippocampus should be relativelyshort, with lengths not much longer than the length determined by the lower bound for temporal distance T for long term memories. The maximal lengthL of hippocampal axons should correspond to this T and L ∼ 10−2 metersfrom the size of the hippocampus might be a reasonable guess giving a timescale of about 15 minutes (these estimates are just orders of magnitude).
c) The recall of long term memories could basically correspond to a tran- sition of a neuronal microtubule to a higher energy state by an emission ofnegative energy Z0 ME. The process would be preceded by the emission ofa p-adic Z0 ME representing the intention to remember and transformed toa real negative energy MEs in the jump to a higher energy state. The neu-ronal/astrocytic microtubules of the right brain hemisphere could be special-ized to send/receive negative energy MEs, whereas the astrocytic/neuronalmicrotubules of the left hemisphere would be specialized to send/receivepositive energy MEs. Of course, this is just a naive guess inspired by theright/left–holistic/reductionic dichotomy. What is however clear that micro-tubuli with abnormally small metabolic energy feed would be responsible forgenerating long term memory recalls and those with abnormally large energyfeed responsible for generating long term memories.
d) Tubulin dimers correspond to the Mersenne prime p = Mk = 2k − 1, k = 13, and the n-ary 13-adic time scale nearest to p-adic prime nearestto .1 second time scale of the memetic code word is T (20, 13) ' .8 secondswhereas single bit lasts for T (20, 13)/13 ' 61 milliseconds. 8 seconds israther natural time scale from the point of view of human consciousness.
Corresponding frequencies are 1.25 Hz in delta band, and 16.25 Hz in thelower end of the beta band which conforms with the fact that cognitioncorrelates with the beta band activity of EEG. That delta frequency alonedoes not give rise to conscious experience would be due to the fact that nophase transition giving rise to a conscious experience occurs if all tubulinspossess same ground state conformation. The facts that delta band weakensduring ageing and also memory generation mechanisms deteriorate towardsthe old age, conform with the idea that this band is responsible for thegeneration of memory codewords. If this view is correct, hippocampal thetashould be responsible for the binding of mental images rather than codingof our long term memories. Of course, also a lower level representations intime scale of hippocampal theta could be in question.
e) At this stage it is not possible to answer the question whether micro- tubuli correspond to subselves or subsub.selves. If the entangled micro-tubuli correspond to our subselves, the microtubuli belonging to differentneurons should be able to entangle with each other. This requires the pres-ence of join along boundaries bond contacts between pre- and postsynapticmicrotubuli. MEs with lengths of neuron length scale could serve as this kind of contacts and generate time like entanglement between the microtubuli ofneurons along the neural pathway.
What about effectively 2-D and 3-D memory representa-tions? Microtubuli need not be solely responsible for our long term memory repre-sentations. The fact that microtubuli seem to correlate with cognition anddeclarative memories which involve typically representations linear with re-spect to time suggests that the effective dimension D of the structure in-volved determines the character of the long term memory and also that ofsensory experience. Moreover, it is quite possible that a large number ofentangled neurons results in a kind of 'stereo consciousness' fusing a largenumber of slightly different views about the same sensory input. This wouldmean large number of entangling Grandmother neurons.
a) Cell membranes consist of a large number of parallel rather than serially ordered units. Hence cell membranes could be responsible for thestorage of sensory memories, which are 2-dimensional at the basic level, sayvisual images. The neuron size of 10−4 meters corresponds to the lowerbound of about millisecond for T ∝ L3.
b) Three-dimensional blobs of biomatter in length scale range 1 micron- 10 nanometers span the range 1 millisecond-32 years for temporal distanceT . This allows to consider the possibility that 3-D structures could be alsoresponsible for long term memory representations. If one takes seriously thedimensional rule, 3-D structures should give to genuinely three-dimensionalsensory memories and make 3-D spatial imagination and sensory experiencepossible. It is not obvious whether neurons contain any 3-D lattice likestructures besides liquid crystal blobs of ordered water.
structures could also result as composites of 2-D structures.
anen (1990), Topological Geometrodynamics Internal Report [padTGD] M. Pitk¨ anen (1995), Topological Geometrodynamics and p-Adic [cbookI] M. Pitk¨ anen (2001), TGD inspired theory of consciousness with [cbookII] M. Pitk¨ anen (2001) Genes, Memes, Qualia, and Semitrance, [Fordahl] M. Fordahl (2000), Scientists trigger neuron regeneration in mouse brain EDT http://www.nandotimes.com .
 W. J. Freeman (2001), Making sense of brain waves: the most baffling frontier in neuroscience, http://sulcus.berkeley.edu .
 P. Gariaev et al (2000), The DNA-wave biocomputer, International Journal of Computing Anticipatory Systems. Ed. Daniel Dubois, Pub-lished by CHAOS, Vol. 10, 2001.
 P. S. Goldman-Rakic (1999), The "psychic" neuron of the cere- Ann. N. Y. Acad. Sci. 1999 Apr. 30;  R. Joseph, Hippocampus, http://www.brain-mind.com .
 P. M. Mericle and M. Daneman (1996) Memory for unconsciously per- ceived events: evidence for anesthetized patients, Consciousness andCognition,5, p. 525.
 D.V. Nanopoulos (1995) Theory of Brain function, Quantum Mechan- ics, and Superstrings, CERN-TH/95-128, hep-ph/9505374.
 L. R. Squire(1992) Memory and the Hippocampus: A Synthesis From Findings With Rats, Monkeys, and Humans, Psychological Review, vol.
99,no 2, 195-231.
 M. Tegmark (1999), The importance of quantum decoherence in brain processes, arXiv: quant-ph/9907009.
 R. P. Vertes, K. E. Eastman (2000), The case against memory consoli- dation in REM sleep, Behavioral and Brain Sciences 23 (6).
[A1] The chapter Was von Neumann Right After All? of [TGD].
[B1] The chapter Anomalies Related to the Classical Z0 Force and Gravita- tion of [padTGD].
[C1] The chapter Self and Binding of [cbookI].
[C2] The chapter p-Adic Physics as Physics of Cognition and Intention of [C3] The chapter Biological Realization of Self Hierarchy of [cbookI].
[C4] The chapter Quantum Theory of Self-Organization of [cbookI].
[C5] The chapter Quantum Control and Coordination in Bio-systems: Part I of [cbookI].
[C6] The chapter Quantum Control and Coordination in Bio-Systems: Part II of [cbookI].
[C7] The chapter Quantum antenna hypothesis of [cbookI].
[D1] The chapter General Theory of Qualia of [cbookII].
[D2] The chapter Quantum Coherent Dark Matter and Bio-Systems as Macroscopic Quantum Systems of [cbookII].
[D3] The chapter Genes and Memes of [cbookII].
[D4] The chapter Dark Matter Hierarchy and Hierarchy of EEGs of [D5] The chapter Quantum Model for Nerve Pulse of [cbookII].
[D6] The chapter Many-Sheeted DNA of [cbookII].
[D7] The chapter Semi-trance, Mental Illness, and Altered States of Con- sciousness of [cbookII].
Modern Organocopper Chemistry. Edited by Norbert Krause Copyright > 2002 Wiley-VCH Verlag GmbH ISBNs: 3-527-29773-1 (Hardcover); 3-527-60008-6 (Electronic) 2Transmetalation Reactions Producing Organocopper Reagents Paul Knochel and Bodo Betzemeier Organocopper reagents constitute a key class of organometallic reagents, with nu-merous applications in organic synthesis . Their high reactivities and chemo-selectivities have made them unique intermediates. Most reports use organocopperreagents of type 1 or 2, which are prepared from organolithiums. This trans-metalation procedure confers optimal reactivity, but in many cases it permitsonly the preparation of relatively unfunctionalized organocopper reagents. Morerecently, substantial developments have been taking place in transmetalations toorganocopper reagents starting from organometallic species that tolerate the pres-ence of functional groups , while synthetic methods permitting the preparationof functionalized organolithiums and organomagnesium compounds have alsobeen developed. All organometallics in which the metal M is less electronegativethan copper, and all organometallic species of similar electronegativity but withweaker carbon-metal bonds, are potential candidates for transmetalation reac-tions . Thus, reaction conditions allowing the transmetalation of organo-boron,-aluminium, -zinc, -tin, -lead, -tellurium, -titanium, -manganese, -zirconium and-samarium compounds have all been found, resulting in a variety of new organo-copper reagents of type 3. Their reactivity is dependent on the nature of the origi-nal metal M, which in many cases is still intimately associated with the resultingorganocopper reagent (Scheme 2.1) [3–5].
Herpes Zoster Ophthalmicus: More than Meets the EyeGuillermo Rocha, MD, FRCSC; Mercedes Muzychuk, OD The varicella virus is common; in recent decades the In this paper, a review of the basics of the herpes infection was found to affect 50% of Canadians by age 5 zoster virus will be covered. Aspects of ophthalmic and 90% by age 12.6 Although the rate of infection with