Orch-OR

Orch OR (“Orchestrated Objective Reduction”) is the proposal that information processing in the brain involves complex computational processes within every neuron, that involve co-ordinated changes in the conformational states of tubulin proteins within microtubules. The proposal was put forward in the mid-1990s by British theoretical physicist Sir Roger Penrose and American anesthesiologist Stuart Hameroff.

Microtubules are cylindrical lattices of tubulin proteins that can act like "conveyer belts" inside cells to move vesicles, granules, organelles like mitochondria, and chromosomes to different locations in the cell via special attachment proteins. Structurally, they are linear polymers of a globular protein, tubulin - these linear polymers are called protofilaments. Microtubules are important components of cilia, flagella or centrioles . These structural components are abundant in all cells and are involved in many cellular processes including mitosis, cytokinesis, and vesicular transport. The basic Orch OR idea is that discrete MT tubulin states act as information bits in MT computers inside brain neurons. Orch OR suggests MT tubulin qubits switch coherently and performing quantum computations by entanglement with other tubulins, .

History and background

In the 1970s and 1980s Hameroff attempted to show that consciousness depends on computation within neurons in microtubules, self-assembling cylindrical polymers of the protein tubulin. Microtubules organize neuronal shape and function, e.g. forming and maintaining synapses (and help single cells like paramecium swim, find food and mates, learn and have sex without any synapses). Hameroff concluded that microtubules function as molecular-level cellular automata, and that microtubules in each neuron of the brain had the computational power of 10¹⁶ operations per second. Neuronal-level synaptic operations were regulated by these internal computations, Hameroff claimed, so attempts by artificial intelligence (AI) workers to mimic brain functions by simulating neuronal/synaptic activities would fail. Hence, as far as explaining consciousness – why we have inner experience, feelings, subjectivity – merely adding another layer of information processing within neurons in microtubules did not help.

Meanwhile Roger Penrose, famous for his work in relativity, quantum mechanics, geometry and other disciplines, had concluded for completely different reasons that AI computational approaches were inadequate to explain consciousness. In his 1989 book “The Emperor's New Mind” Penrose used Kurt Gödel’s theorem to argue that human consciousness and understanding required a factor outside algorithmic computation, and that the missing “non-computable” factor was related to a specific type of quantum computation involving what he termed “objective reduction” (“OR”), his solution to the measurement problem in quantum mechanics.

Penrose considered superposition as a separation in underlying reality at its most basic level, the Planck scale. Tying quantum superposition to general relativity, he identified superposition as spacetime curvatures in opposite directions, hence a separation in fundamental spacetime geometry. However, according to Penrose, such separations are unstable and will reduce at an objective threshold, hence avoiding multiple universes.

The threshold for Penrose OR is given by the indeterminacy principle E=ħ/t, where E is the gravitational self-energy (i.e. the degree of spacetime separation given by the superpositioned mass), ħ is Planck’s constant over 2π, and t is the time until OR occurs. Thus the larger the superposition, the faster it will undergo OR, and vice versa. Small superpositions, e.g. an electron separated from itself, if isolated from environment would require 10 million years to reach OR threshold. An isolated one kilogram object (e.g. Schrodinger’s cat) would reach OR threshold in only 10^-37 seconds. Penrose OR is currently being tested.

An essential feature of Penrose OR is that the choice of states when OR occurs is selected neither randomly (as are choices following measurement or decoherence) nor completely algorithmically. Rather, states are selected by a “non-computable” influence involving information embedded in the fundamental level of spacetime geometry at the Planck scale. Moreover, Penrose claimed that such information is Platonic, representing pure mathematical truth, aesthetic and ethical values. Plato had proposed such pure values and forms, but in an abstract realm. Penrose placed the Platonic realm in the Planck scale.

In The Emperor’s New Mind Penrose suggested that consciousness required a form of quantum computation in the brain.

Quantum computation had been suggested by Paul Benioff, Richard Feynman and David Deutsch in the 1980s. The idea is that classical information, e.g. bit states of either 1 or 0, could also be quantum superpositions of both 1 and 0 (quantum bits, or qubits). Such qubits interact and compute by nonlocal quantum entanglement, eventually being measured/observed and reducing to definite states as the solution. Quantum computations were shown to have enormous capacity if they could be constructed e.g. using qubits of ion states, electron spin, photon polarization, current in Josephson junction, quantum dots etc. During quantum computation, qubits must be isolated from environmental interaction to avoid loss of superposition, i.e. “decoherence”.

Penrose argued that quantum computation which terminated not by measurement, but by his version of objective reduction, constituted consciousness (allowing Platonic non-computable influences). Penrose had no definite biological qubits for such quantum computation by OR, except to suggest the possibility of superpositions of neurons both “firing and not firing”.

Hameroff read The Emperor’s New Mind and suggested to Penrose that microtubules within neurons were better suited for quantum computing with OR than were superpositions of neuronal firings. The two met in the early 1990s and began to develop the theory now known as Orch OR. “Orch” refers to orchestration, the manner in which biological conditions including synaptic-level neuronal events provide feedback to influence quantum computation with OR in microtubules.

The Orch OR model

For biological qubits, Penrose and Hameroff chose conformational states of the tubulin subunit proteins in microtubules. Tubulin qubits would interact and compute by entanglement with other tubulin qubits in microtubules in the same and different neurons.

It was known that the peanut-shaped tubulin protein flexes 30 degrees, giving two different conformational shapes. Could such different states exist as superpositions, and if so, how? P&H considered three possible types of tubulin superpositions: separation at the level of the entire protein, separation at the level of the atomic nuclei of the individual atoms within the proteins, and separation at the level of the protons and neutrons (nucleons) within the protein. Calculating the gravitational self-energy E of the three types, separation at the level of atomic nuclei had the highest energy, and would be the dominant factor. P&H calculated E for superposition/separation of one tubulin qubit at the level of atomic nuclei in all the amino acids of the protein. They then related this to brain electrophysiology

The best electrophysiological correlate of consciousness is gamma EEG, synchronized oscillations in the range of 30 to 90 Hz (also known as “coherent 40 Hz”) mediated by dendritic membrane depolarizations (not axonal action potentials). This means that roughly 40 times per second (every 25 milliseconds – “msec”) neuronal dendrites depolarize synchronously throughout wide regions of brain.

Using the indeterminacy principle E=ħ/t for OR, P&H used 25 msec for t, and calculated E in terms of number of tubulins (since E was known for one tubulin). Thus they were asking: how many tubulins would be required to be in isolated superposition to reach OR threshold in 25 msec, 40 times per second, corresponding with membrane-level brain-wide effects? The answer turned out to be 2 x 10¹¹ tubulins.

There are roughly 10⁷ tubulins per neuron. If all tubulins in microtubules in a given neuron were involved, this would correspond with 2 x 10⁴ (20,000) neurons. However, because dendrites are apparently more involved in consciousness than axons (which contain many microtubules), and because not all microtubules in a given dendrite are likely to be involved at any one time, an estimate of, say, 10 percent involvement gives 200,000 neurons involved in consciousness every 25 msec. These estimates (20,000 to 200,000 neurons) fit very well with others from more conventional approaches suggesting tens to hundreds of thousands of neurons are involved in consciousness at any one time.

How would microtubule quantum superpositions avoid environmental decoherence? Cell interiors are known to alternate between liquid phases (solution: “sol”) and quasi-solid (gelatinous: “gel”) phases due to polymerization states of the ubiquitous protein actin. In the actin-polymerized gel phase, cell water and ions are ordered on actin surfaces, so microtubules are embedded in a highly structured (i.e. non-random) medium. Tubulins are also known to have C termini “tails”, negatively charged peptide sequences extending string-like from the tubulin body into the cytoplasm, attracting positive ions and forming a plasma-like Debye layer which can also shield microtubule quantum states. Finally, tubulins in microtubules were suggested to be coherently pumped laser-like into quantum states by biochemical energy (as proposed by Herbert Frohlich).

Actin gelation cycling with 40 Hz events permits input to, and output from isolated microtubule quantum states. Thus during classical, liquid phases of actin depolymerization, inputs from membrane/synaptic inputs could “orchestrate” microtubule states. When actin gelation occurs, quantum isolation and computation ensues until OR threshold is reached, and actin depolymerizes. The result of each OR event (in terms of patterns of tubulin states) would proceed to organize intraneuronal activities including axonal firing and synaptic modulation/learning. Each OR event (e.g. 40 per second) is proposed to be a conscious event, equivalent in philosophical terms to what philosopher Alfred North Whitehead called “occasions of experience”.

Thus one implication of the Orch OR model is that consciousness is a sequence of discrete events, rather than a continuum. Yet conscious experience is subjectively uninterrupted, analogous to a movie appearing continuous to observers despite being a series of frames. The difference is that in Orch OR, each conscious event is itself an intrinsic, subjective observation. Moreover the frequency of conscious events may vary, 40 Hz being an average. If someone is excited and conscious events occur more often, (e.g. at 60 Hz), then subjectively the external world seems slower, as great athletes report during peak performance. By E=ħ/t, more frequent conscious events correspond with greater E, hence more tubulins/neurons per conscious events and greater intensity of experience. Thus a spectrum of conscious events may exist, similar to photons. There exists a spectrum of conscious quanta-like events ranging from longer wavelength, low intensity events (large t, low E) and shorter wavelength, higher intensity events (small t, large E).

Questions

Orch OR was developed in 1994, and first published in 1995 and 1996 with followup articles in 1998 and 2001. It has been greeted with a mixture of skepticism and enthusiasm, with those aligned with A.I. being particularly critical. There are some obvious questions:

References

• Hagan S, Hameroff S, & Tuszynski J, (2002) Quantum Computation in Brain Microtubules? Decoherence and Biological Feasibility. Physical Reviews E, 65: 061901.

• Hameroff S.R., & Watt R.C. (1982) Information processing in microtubules. Journal of Theoretical Biology 98: 549-561.

• Hameroff, S.R., & Penrose, R., (1996) Orchestrated reduction of quantum coherence in brain microtubules: A model for consciousness. In: Toward a Science of Consciousness - The First Tucson Discussions and Debates. (Hameroff, S.R., Kaszniak, and Scott, A.C., eds.), pp. 507-540, MIT Press. Also published in Mathematics and Computers in Simulation (1996) 40:453-480.

• Hameroff, S.R., & Penrose, R. (1996) Conscious events as orchestrated spacetime selections. Journal of Consciousness Studies 3(1):36-53

• Hameroff, S. (1998) Quantum computation in brain microtubules? The Penrose-Hameroff "Orch OR" model of consciousness. Philos. Trans. R. Soc. London Ser. A 356, 1869-1896.

• Hameroff, S. (1998b) "Funda-mentality": is the conscious mind subtly linked to a basic level of the universe? Trends in Cognitive Science 2:119-127.

• Hameroff, S. (1998d) Did consciousness cause the Cambrian evolutionary explosion? In: Toward a Science of Consciousness II: The Second Tucson Discussions and Debates. Eds. Hameroff, S.R., Kaszniak, A.W., & Scott, A.C., Cambridge, MA: MIT Press, pp. 421-437.

Grush, R., Churchland, P.S. (1995). "Gaps in Penrose's toilings". Journal of Consciousness Studies 2 (1): 10–29. ^[e] </ref>

• Hameroff, S., Nip, A., Porter, M., & Tuszynski, J. (2002) Conduction pathways in microtubules, biological quantum computation and microtubules. Biosystems 64(13):149-168.

• Penrose R, Hameroff SR (1995) What gaps? Reply to Grush and Churchland. J Consciousness Studies 2:98-112

,

</ref> In 1996 Penrose offered a consolidated reply to many of the criticisms of 'Shadows'. "Cortical dendrites contain largely A­-lattice microtubules" is one of 20 testable predictions published by Hameroff in 1998 and it was hypothesized that these A­-lattice microtubules could perform topological quantum error correction. The latter testable prediction had already been experimentally disproved in 1994 by Kikkawa et al., who showed that all in vivo microtubules have B-lattice and a seam. Reimers JR et al. (2009). "Weak, strong, and coherent regimes of Fröhlich condensation and their applications to terahertz medicine and quantum consciousness". Proc Nat Acad Sci 106: 4219–24. </ref> an error in the calculated number of tubulin dimers per cortical neuron, McKemmish LK et al. (2009). "Penrose-Hameroff orchestrated objective-reduction proposal for human consciousness is not biologically feasible". Physical Review E 80: 021912–6.

</ref> Spier, E. & Thomas, A. (1998) Trends in Cognitive Sciences, 2, 124-125.

A Quantum of Consciousness? A glance at a physical theory for a mind ref name=Tegmark2000>Tegmark, M. (2000), "Importance of quantum decoherence in brain processes", Phys. Rev. E 61: 4194–4206, DOI:10.1103/PhysRevE.61.4194</ref>