"The Astonishing Hypothesis as a firm background to my testable suggestions."
by Stuart Kauffman
The human brain has about 10 to the 11th power neurons, with many distinct regions from the brain stem to the thalamus to the cortex. A neuron can be thought of as having three main parts, a cell body, a descending axon which may or may not branch, and a dendritic "tree" with about 6000 synapses that leads into the cell body. Neurons have cell membranes with a transmembrane resting electrical potential which is about - 70 millivolts inside. The axon of one neuron terminate on one synapse, or if the axon is branched, each branch terminates on one synapse of the same or different neurons.
Each synapse is a small "bubble" of membrane attached by an anatomical "spike" to a very small region, less than a micron squared, of adjacent dendritic membrane. A synapse contains about 200 specific proteins on the "presynaptic" half of the synapse, and 1500 on the "postsynaptic" half of the synapse. The two parts of the synapse are separated by the synaptic cleft.
Axons propagate action potential spikes in "spike trains", from the cell body to their termini. An action potential arises if the membrane voltage in the cell body rises from -70 milli volts to about - 50 millivolts due to events in synapses and the dendritic tree leading to that nerve cell body. The action potential is a rapid change in membrane potential due to opening and closing of membrane channels that allow charged sodium, potassium and other electrically charged ions to cross the membrane. The action potential propagates as a membrane "depolarization" spike down the axon.
Where an axon terminates on a synapse, the electrochemical events cause vesicles in the presynaptic half of the synapse to open and release neurotransmitters such as glutamate or GABA. These diffuse across the synaptic cleft, and bind to neurotransmitter protein receptors on the post synaptic half of the synaptic membrane. These bound receptors, acting alone or in multi-protein complexes, open membrane channels in their local membrane region and allow charged ions to flow across that tiny region of the dendritic and synaptic membrane, either "exciting" (glutamate) that tiny region by making the transmembrane voltage more positive, ie -60 millivolts rather than -70 millivolts, or "inhibiting" (GABA) that tiny region by making the voltage more negative, eg - 80 millivolts.
The tiny alterations in membrane potential at the 6000 synpatic - dendritic membrane regions propagate to the cell body and are "summed", and if the resulting membrane potential of the cell body is -50 millivolts or more positive, an action potential initiates and propagates down the axon.
With this sketch: The 10 to the 11th neurons form a PARALLEL PROCESSING NETWORK in which action potentials are propagating down axons to dendrites to elicit synaptic behaviors, local dendritic membrane transmembrane potential changes, which may or may not induce further action potentials. Inputs from the eyes, ears, nose project to specific regions in the brain, which in turn is richly interconnected in highly evolved ways.
The Neural Correlates of Consciousness
The basic view of all neuroscientists and psychologists interested in consciousness assumes the following, which may well be true. There is some kind of neural spike train "code". The code might be, for each axon, the average rate of action potential spikes traveling down the axon, or some "bursty" time varying pattern of action potentials.
There must be specific "consciousness neurons" as Crick says. The aim is to find out which. Experience, ie qualia, are to be associated with the CLASSICAL PHYSICAL BEHAVIOR of these consciousness neurons via the neural code.
Based on psychological and neurophysiology, it is generally believed that that there are two essential further correlates of consciousness: 1) A very short term memory, about 60 milliseconds. 2) A shifting focus of attention, where the current focus of attention is required for consciousness. The focus of attention shifts in a SERIAL fashion, while the neurons are parallel processing.
The Binding Problem
The "Binding Problem" is crucial. An example is visual awareness. Visual "information" in processed in highly complex ways in different areas of your brain. Suppose you look at a blue circle and yellow triangle, and the "blue", "circle", "yellow" and "triangle" aspects of what you experience are processed in DIFFERENT REGIONS of the brain, eg regions of the cortex. How does "blue" become associated with, or bound to, "circle" and "yellow" bound to "triangle"?
The current hypothesis is that there is a rapid, 40 per second rhythm in the spiking of neurons, and neurons that spike at the same time, EVEN IF DISCONNECTED ANATOMICALLY, will somehow "BIND" their qualia from their neural codes into one experience. Then the hypothesis is that "blue" and "circle" neurons spike at one time in the 40 per second rhythm so are bound, while "yellow" and "triangle" spike together at another time in the 40 per second rhythm, so are bound.
It is hard to overstate the binding problem: As you look at your visual "scene" how many "features" do you see? These include not only the blue circle, but the relational feature of the distance between the blue circle and yellow triangle and all the relational features of your visual scene. It was hoped that "grandmother cells" each firing for specific combinations of features, might do the trick, but the brain would need more neurons than there are combinations of features we can see for grandmother cells to "work".
Turing The Brain On Its Head
Crick states, just in passing, a stunning fact. There is a VAST LOSS OF INFORMATION FROM ALL THE MOLECULAR BEHAVIORS IN THE SYNAPSES AND ALL THE EXQUISITELY FINE GRAINED SPATIAL AND TEMPORAL CHANGES IN TINY LOCAL POST SYNAPTIC DENDRITIC TRANSMEMBRANE POTENTIALS TO THE ENORMOUSLY REDUCED INFORMATION CONTAINED IN THE ACTION POTENTIAL SPIKE TRAINS TRAVELING DOWN AXONS.
Crick is obviously correct. What if we "stand the brain on its head" and try to imagine that the "job" of the entire nervous system and sensory inputs from eyes and so on, is to MEDIATE THIS FINE GRAINED TIME SPACE MEMBRANE POTENTIAL VARIATION AT SYNAPSES THAT INDUCE SYNAPTIC BEHAVIOR? IN TURN THAT SYNAPTIC BEHAVIOR DRIVES FURTHER DENDRITIC AND ACTION POTENTIAL BEHAVIORS.
What if the SYNAPSES are the brain correlates of consciousness? If there are 10 to the 11 neurons, and 6000 synapses per neuron, and hundreds of neurotransmitter receptor proteins per synapse, there is enormously more information in the behaviors of those neurotransmitter receptors (or other aspect, classical, quantum or both) of synaptic behaviors, than in the firing patterns of 10 to the 11th neurons in the hoped for neural code.
Interestingly, neurotransmitter receptor proteins evolved before neurons, eg in choanoflagellates, which are single celled organisms thought to be the precursors of animals. Maybe single cell organisms can be conscious. We truly do not know.
The Synaptic Correlates of Consciousness
R. Penrose and H. Stapp, mathematician and quantum physicist before me, and I, now suggest that QUANTUM MEASUREMENT is the BRAIN LOCUS OF CONSCIOUSNESS. I suggest specifically that we might look for measurement events in post synaptic neurotransmitter receptor proteins or protein complexes.(Penrose and Stapp have been entirely ignored because quantum events in the brain have been considered completely unreasonable. The quantum coherent behavior of chlorophyll and its antenna protein now suggests that we reconsider quantum events in the brain.)
A first clue is that anesthetics bind to hydrophobic pockets in these receptor proteins. Were anesthetics to "freeze" receptors into classical behavior, no further measurement events could occur, no qualia would arise, and we would be anesthetized.
I further suggest that we consider QUANTUM ENTANGLEMENT BETWEEN SYNAPSES IN ANATOMICALLY CONNECTED AND NON-CONNECTED NEURONS AS A POSSIBLE BASIS TO SOLVE THE BINDING PROBLEM.
Recall that two entangled quantum degrees of freedom, even if separated such that light cannot travel between them in the time interval between measurements of each of the two particles, are NON-LOCALLY CORRELATED instantaneously. More, recent work shows that as the NUMBER OF ENTANGLED QUANTUM PARTICLES INCREASES, THE CORRELATIONS AMONG THEM INCREASE. The network of close correlations among synaptic measurement events is a different means to solve BINDING PROBLEM to achieve a "Unity of Consciousness".
But the emergence of these non-local correlations, hence the "unity of consciousness", requires that each particle or degree of freedom be MEASURED. But this precisely fits the hypothesis above that experience, qualia, are associated with MEASUREMENT EVENTS, perhaps in synapses.
More, there is new evidence that on a "quantum graph" of many nodes, local behaviors can alter which DISTANT NODES ON THE GRAPH ARE QUANTUM ENTANGLED.
Therefore I suggest that quantum entangled measurements events, perhaps in synapses, are the loci of bound experience, and that TEMPORALLY SERIAL SHIFTING PATTERNS OF ENTANGLEMENT MAY CONSTITUTE THE SERIAL SHIFTING FOCUS OF ATTENTION THAT IS REQUIRED FOR CONSCIOUSNESS.
Finally, the measurement events of different entangled quantum degrees of freedom need not happen simultaneously, so the spread in time of these measurement can constitute or contribute to the very SHORT TERM MEMORY required for consciousness.
These ideas are a new way of looking at the brain. The entire brain may be an enormous, highly evolved Trans-Turing System. And single celled organisms may be conscious.
RELATED VIDEOS
