The following is neither credentialed nor peer reviewed, but intended to offer a perspective on the brain activity which produces cognition.
In The Other Brain, R. Douglas Fields writes about his research into glial cells, which comprise the majority (~80% depending on the source) of cells in the brain. These were traditionally thought to provide support functions for neurons, and Fields (who was an NIH researcher) + others discovered that glial cell function has a correspondence with neural activity.
My thoughts around the implications of his work is that while realtime activity[1] happens over neural networks,
- Astrocytes are responsible for the reuptake of neurotransmitters. As a synapse is activated based on a threshold concentration, these cells may have a role in regulating synaptic strength (could they be one of the mechanisms involved in Hebbian learning?). What’s more, in the same way that a neuron can connect to ten thousand other neurons[2], one astrocyte can connect to tens of thousands of them.
- Oligodendrites connect to dozens of neurons and form the equivalent of myelin sheaths (myelin refers to material in the PNS) along the white matter tracts connecting regions of the cortex[3]. This affects individual neurons’ signal propagation speed, which, if we are looking at neurons not like transistors (on/off switches) but carriers of activity that is spatial+temporal (IE something dynamical in nature), may affect the pattern.[4]
- Microglia are the brain’s immune cells (the blood-brain barrier forms before the immune system develops, so there are no immune cells in the brain) and are mobile, moving around like amoebas. Besides fighting infections, they also prune synapses.
If these glial cells have functions that affect the spatio-temporal patterns of activity that generate ‘mind’, do they play a role in cognition? Are they proxies that carry out tasks based on chemical signaling from neurons, or do they act independently?
100 different types of cells, amoeba-like and octopus-like creatures (maybe entities, rather, since ‘creature’ implies a stand-alone organism) adjusting the neural networks?
To me, this makes the brain sound less like an object and more like a civilization of individual cells, every mind a society (as Minsky puts it) or an economy.
[1] originally this was “information processing” although one of the things Fields cautions against is using computers as a metaphor, since in every era people will use whatever metaphor they believe best fits — pipes, mechanical pumps, and valves during the Victorian era (where before Maxwell people were attempting to model electricity as a fluid), for instance
[2] I’m not clear if this means a count of adjacent neurons or dendritic spines + axon terminals — there is a topological (? this term or whatever the dynamical equivalent is) difference if two neurons will connect at more than one point
[3] I was wondering recently how a region of the cortex connects to other regions, and the nature of those connections. It turns out that each region is not connected to every other region
[4] Although I wonder if it makes sense to think of neural activity as pattern-like… the way Walter J. Freeman puts it: “For centuries the brain has been described as dynamic systems, beginning with Descartes, who conceived of the pineal as a kind of valve used by the soul to regulate the pumping of spiritual fluid into the muscles… this was followed by metaphors of clocks, telegraph and telephone systems, thermodynamics, digital computers, and holographs. Yet brains are not ‘like’ any artificial machine. If anything, they are like natural self-organizing processes such as stars and hurricanes.” < — does it make sense to think of the activity of air molecules in a hurricane as a pattern? A question that Freeman raised is whether the brain operates off of, or there is such a concept as ‘representations.’ The model of cognition traditionally used in artificial intelligence comes from the field of psychology, which (AFAIK) models thinking that way
