From hameroff@U.ARIZONA.EDU Tue Feb 9 21:20:54 1999 Date: Mon, 8 Feb 1999 20:15:41 -0700 From: Stuart Hameroff Subject: [q-mind] Reply to Klein and Nunn - Henry Stapp >From Henry Stapp This is a reply to certain comments on my vN/W theory by Stan Klein [q-mind 4 Feb to 5 Feb, 1999]. I answer also some questions from Chris Nunn about predictions of my theory pertaining to Bierman's experiments. First, however, I give the February 7 version of the section entitled "Idealism and Materialism": it goes significantly beyond the version I sent earlier, and the changes are pertinent here. Idealism and Materialism I have stressed just now the idea-like character of the physical state of the universe, within vN/W quantum theory. This suggest that the theory may conform to the tenets of idealism. This is partially true. The quantum state undergoes, when a fact become fixed in a local region, a sudden jump that extends over vast reaches of space. This gives the physical state the character of a representation of knowledge rather than a representation of substantive matter. When not jumping the state represents potentialities or probabilities for actual events to occur. Potentialities and probabilities are normally conceived to be idea-like qualities, not material realities. So as regards the intuitive conception of the intrinsic nature of what is represented within the theory by the physical state it certainly is correct to say that it is idea-like. On the other hand, the physical state has a mathematical structure, and a behaviour that is governed by the mathematical properies. It evolves much of the time in accordance with local deterministic laws that are direct quantum counterparts of the local deterministic laws of classical mechanics. Thus as regards various structural and causal properties the physical state certainly has aspects that we normally associate with matter. So this vN/W quantum conception of nature ends up having both idea-like and matter-like qualities. The causal law involves two complementary modes of evolution that, at least at the present level theoretical development, are quite distinct. One of these modes involves a gradual change that is govern by local deterministic laws, and hence is matter-like in character. The other mode is abrupt, and is idea-like in {\it two} respects. The first of these idea-like aspects is that each abrupt change involves a Heisenberg choice that is naturally expressed in terms of experiences, per se. In Copenhagen quantum theory the Heisenberg choice is made by the mind of the human experimenter/observer. In the vN/W theory, as developed here, that choice continues to be made by the mind of the human observer, i.e., by the experiential aspect of the mind/brain/body subsystem that constitutes the observing human being. This Heisenberg choice is represented within the formalism by what von Neumann called Process I: the Heisenberg choice is represented by the transformation S \longrightarrow [PSP + (1-P)S(1-P)]. The second idea-like aspect of the abrupt change is the Dirac Choice, which picks out either PSP or (1-P)S(1-P). It corresponds to a reduction of the state to one compatible with the experiential increment in knowledge associated with P, or to a reduction to the alternative associated with (1-P). A key consequance of all this is that the idea-like aspects of nature do not lie outside the physical state. Rather the physical state of the theory represents in mathematical terms the structural properties of an intrinsically idea-like reality. Thus the vN/W ontology, as I am interpreting it, does not push human experience out of the quantum state when it places the entire physical universe on the physical side of the von Neumann cut. Rather it eliminates the von Neumann cut altogether, by bringing the part of the physical universe that represents (the structure of) our human experiences into the quantum state of the theory. This is in line with the principle of psycho-physical parallelism that von Neumann invokes: ``it must be possible to describe the extra-physical process of the subjective perception as if it were in reality in the physical world -- i.e., to assign to its parts equivalent physical processes in the objective environment, in ordinary space.'' This quantum parallelism is also in line with the ``Identity theory of Mind'', if the physical world in identity theory is understood to be the physical world as it is represented in quantum theory, rather than the physical world that was imagined to exist in classical physical theory. The identity theory, which asserts that each thought, or other mental reality, is identical to some aspect of the physical universe, is puzzling and unreasonable if the physical world is what classical-physics assumed it to be. Classical physical theory is ontologically and dynamically micro-local: the physical reality is defined in terms of simple microscopic realities, each of which evolves in a manner that is controlled exclusively by the prior properties of its immediate neighbors. But a thought is a reality that grasps as a unit certain aspects of a macroscopic collection of these micro-realities that have a future-directed intentional aspect that the individual micro-realities lack. Since, according to classical physical theory microscopic causal connections from the past are sufficient by themselves to determine the causal evolution there is no dynamical or ontological need for any realities that grasp macroscopic aspects as wholes: the central thrust of classical physical theory was precisely to eradicate any logical need for realities of that kind. The situation is completely altered if the physical world is taken to be what vN/W quantum theory asserts it to be. Then the basic reality is idea-like to begin with, by virtue of the way it behaves, and the future-directed thoughts play an essential dynamical role. In this theory the evolving idea-like reality can be represented naturally also within the parallel description in terms of the physical state S, which represents, at least within the theory, the structural properties of the intrinsically idea-like reality. End [Stan} There are many interpretations of Orthodox QM. Some interpretations have very visible cuts (vN) some have hidden cuts (vN/Wigner, Stapp, Bohm, Everett). The choice of interpretation depends on one's esthetics rather than on anything measurable. A few days ago on q-mind Henry expressed a distaste for my esthetic of placing the cut by the neural correlates of qualia. He thought there would be too many feedforward and feedback signals flowing for a clean cut. I disagree, but that is for a future posting. The important point here is a point that I keep stressing, that as far as I know, all these different interpretations produce the same predictions of *behavior*. ... All my reading of the literature suggests to me that ..., regular Orthodox QM is sufficient to account for human behavior. [Stapp] The important point, rather, is that the issue is not just a matter of taste and aesthetics because the predictions about human behaviour are different in principle. Also, internal consistency of theories is at issue. Thus the issues here are primarily scientific, in the sense of being about differences in theories that lead in principle to differences in predictions about outcomes of possible experiments, and about the internal consistency of such theories. A first question is: What does "regular Orthodox QM" say about brain dynamics? Bohr excludes biological systems from the realm of systems covered by his (Copenhagen) formulation. [This exclusion stems from the practical difficulty in preparing (without killing them) biological systems in states that are sufficiently well defined to allow quantum theory to make useful predictions.] So in order to construct a quantum theory of mind/brain one seems to have to introduce something like the von Neumann state S of the universe (or at least a state S' of the brain of the observer) in order to have any hope of applying QM to brains and conscious experience. That is, one must be able to speak of a quantum state of the body/brain --- perhaps up to some vN cut associated with a collapse --- and either speak of some non-Schroedinger type of process that will produce a collapse, or fill in the missing details needed to make rationally coherent some many-minds type of interpretation of the (von Neumann) state S of the universe. I think by "regular Orthodox QM" one might mean, in the context of brain dynamics, that the collapses are assumed to occur at some mesoscopic level probably much smaller than entire neurons, but in any case very small compared to the macroscopic portion of the brain that, (1), is needed to represent the information that seems to be present in a typical thought, and that, (2), is presumably also the part of the brain that is directing the brain activities that will eventuate in the body/brain's doing what the thought "intends". More generally, I think the "regular Orthodox QM" probably means a model that is basically classical already at some very small scale, far smaller than the scale of the macroprocess that is effectively directing the brain/body to do what the thought intends. Perhaps a specific form would be that each release of a vesicle at a synapse can be considered a classical event that either occurs or does not occur: the switch from quantum description to classical description, and the effective collapse of the state of the brain that it induces, would occur at vesicle release. Or perhaps one might go down even deeper, to the opening and closing of the ion channels through the cell walls of the neurons. The problem, however, is that at the quantum level of description these events could be occurring a little earlier or a little later: the quantum description is continuous. So how does one separate cleanly the partially overlapping possibilities? These partially overlapping possibilities can string together join disparate possibilities. This is not a trivial problem. Tiny changes in billions of micro-channels or billions of vesicles can add up to possibly significant changes. To determine this significance one must consider what questions are put to Nature: one must determine the nature of the Heisenberg Choices. The issues, then, is whether Heisenberg choices that are made at a high level on the basis of properties of the entire quantum state of the brain can influence brain dynamics, or must the behaviour of the quantum brain be, within the "orthodox" rules independent of such macroscopic variables? A superficial glance at von Neumann's work might incline one to the view that the final behaviour of the brain cannot depend on such matters. But a little closer look shows that he proves no such thing. His proof was in a context where the system under study could be decomposed into a sequence of disjoint "good measuring devices" each of which had a disjoint set of ``pointer states'' (in the new jargon) that corresponded to distinct human observations. But there is a no assurance that, and no reason to believe that, these micro-channel or vesicle states have these critical properties. The issue devolves to the issue of the nature of the questions put to nature. If these questions, the Heisenberg choices, depend upon the whole state of the brain, then the dynamical evolution of the brain can depend upon the whole state of the brain in a way that apparently cannot be reduced to a behavior constructible from classical description from the micro-channels, or vesicles, on up. The example of the Quantum Zeno Effext shows how the nature of the questions can influence the BEHAVIOUR of the system. If the questioning is determined by macro-properties, as wholes, then the behaviour can depend on how all the parts combine to make a whole: this dependence would apparently be lost if the transition to the classical description were determined microlocally. So the key point is that the behaviour can depend on how the theory specifies what question are put to nature: different ways of resolving this issue will have behavioural consequences. Given that cut placement does make a behaviourial difference I suggest that the least problematic placement is the unique one in which the unified physical system is not divided artificially into parts that are treated differently Date: Sun, 7 Feb 1999 11:35:45 -0500 From: chrisnunn To: Stan Klein Cc: Henry Stapp Subject: Stapp and Bierman [Chris] Stan, ... Am still doubtful about what Stapp's scheme permits. He may not predict pre-sponses but I suspect that they nevertheless might be allowed, regardless of whether the picture selecter is random or pseudo-random, depending on the 'fine print' of his theory. ... Thus it may be possible for Stapp to say:- (1) It is conceivable that pre-sponses could be observed. (2) Pre-sponses will never be observed when an 'experimenter's friend' is present. (3) Pre-sponses are more likely to be seen when the experimenter expects them, even if his methodology is irreproachable. But maybe this is wrong and you know why. Best wishes. Chris [Stapp] Let Q be the operator corresponding to the experience of the experimenter in eventually observing a `pre-response'. Let P and (1-P) be the operators corresponding to the observer's possible responses if picture #1 is shown to him, and let P' and (1-P') be the operators corresponding to the observers possible responses if picture #2 is shown to him. The question is whether the probability of Q (the experimenters observing a pre-response) can depend on which picture is shown to the observer. The orthodox theory says: In case #1 = Tr [QPSP +Q(1-P)s(1-P)]. In case #2 ' = Tr [QP'SP' +Q(1-P')s(1-P')]. The experiences of the observer and of the experimenter are supposed to be ontologically independent: Q commutes with both P and P'. But then the properties PP = P, P'P' = P' and Tr PX = Tr XP entail = ' = Tr QS. So there will be, in any orthodox theory, no predicted dependence of upon which picture is presented.