There is an hypothesis that I’ve been playing around with that I’m thinking this group is in a very good position to process…
In thinking about the fact that neurons communicate through both electrical impulses and through passing of neurotransmitters, I believe a brain is generating network constrained networks… this would work in get following way
The Imposition: Sensors create electrical impulses that generate a current map of reality in the modeling brain. This model includes a representation through cluster modeling of the world around the subject both in space and time and includes a simplified model of the subject itself, it’s capabilities, knowledge, skills, feelings, etc.
Between the Imposition and the Resolution there is process which I believe is a chemical communication using neurotransmitters (let’s call them colors) that represent different aspects of thought and are mediated by which neurons are lit up by the Imposition. So for example, Red might be possible next states of the world model that have been observed (or heard of) in the past. Green might be the action plans that have achieved those possible changes in state. Orange might be lighting up the level of desire for those next states based on current needs resolution… Black might be feared possible next states that need to be avoided, etc.
Once the various colors have resolved based on the Imposition, now they drive towards the Resolution which is when we switch back to electrical impulses that are sent out to the body to achieve the nearest next state that leads to the most desired next state of the world model.
There appears to be a tendency to think that neurotransmitters attenuate the electrical impulse throughput of the neurons, but I think the system is the reverse of that… the electrical impulses create the map on which the chemical thinking system is able to apply itself. There is no opportunity to wonder about crossing a street when the “street in front of you” neuron is not lit up.
Does the above make sense? Is this an area that Numenta is already considering? Constructive or critical thoughts?
In regards to the recent paper that was published, I would like to add that the brain seems to treat time in a very similar way to the way it treats space… something that is close to me specially is also close to me temporally, and so on.
I should add to the above that each color would have it’s own set of weights within a neuron and would have some ability to attenuate the performance of other colors within a type of neuron.
So for example, sending purple would reduce the black sensitivity of any black triggered neurons when a bad thing didn’t just happen.
Taking that thought a step further, what you have is neurons using the electrical system to generate a chemical map of the world that is sticky. Including things like changing the mind’s perspective within the chemical rendering. A sort of chemical holodeck where there would be a cluster of directional neurons linking to representations of the thing you think is in that direction at around some distance.
The thought/imagination/plan systems would just be watching the chemistry of the sensory system.
One can imagine that planning systems work in he same fashion… intention related neurons filling up with chemistry as they become more likely and then becoming the driving force until a subsidiary set of neurons has signaled completion.
The above also goes a ways to explaining how hormones and psychoactive substances impact certain brain functions. Messing with the speed at which a neuron can change its chemistry messes with the rate at which aspects of the mind are able to change relative to each other.
An easy guess as to how all this works is that a path lit up with serotonin says “this part of the map is interesting”, Dopamine “this part is good”, others for “be anxious about this part”, “fear this part”, “seek this part”, etc.
As far as I know (and I don’t know much yet), the chemical transmitters function as a temporal delay for the synchronisation of the electrical signals from different sources (in different synapses). In other words, the build-up of neurotransmitters enables the electrical signal long enough so that several electrical signals can join to form a strong enough spike in the soma.
In industrial computing this is done with a latch instruction. It is crucial to synchronise processes.
On the other hand, this does not explain why there are so many different neurotransmitters. One type should be enough I’d think. And I also wonder how certain neurotransmitters are associated with certain feelings. (This may not be important to explain our intelligence, but in my opinion it is most likely related to our consciousness). So you may have touched on something interesting here.
Imagine you are seeing a cross walk. The symbol for crosswalk fills with “present” chemistry which primes “intentions” to cross or to move along with “intention” desired state chemistry. “Present” chemistry that describes your current task would then push one or the other of them into an actionable state that would alter actions and/or new desired states.
Migraines tend to occur when there is a drop in blood pressure…
A drop in blood pressure in the head would change neurotransmitter concentrations for a while.
If neurotransmitter concentrations bridge networks within neurons it might stand to reason that migraine flashes are network crossover events.