Neurons that fire together, wire together… Ok, but how?
One of my little pet projects is a neurology book for psychologists, coaches and the like. What most people in these fields imagine about the brain is somewhere between perplexing, preposterous and potentially poisonous. (Seriously, I could tell you stories…) So I kind of wanted to write something like this for them.
Now I do consider it a tall order for me – I have some knowledge of neurology, but actually writing a SENSIBLE, understandable guidebook to neurology, with some practical applications too, that’s a challenge. (Senseless guidebooks to neurology are a dime a dozen, of course. In fact, I’ve walled in at least two people in one library, using only senseless pop-neurology books and as far I can tell, the rotting skeletons still haven’t been found. But then again the rotting corpse smell really doesn’t stand out much in most libraries I’ve been to, so I shouldn’t be surprised.)
But back to my main point. The great thing about trying to write a book like that, even attempting to prepare for doing it at one point in the future, when you know enough, is finding all the lovely little holes in how you’re telling the story to yourself. The teeny-tiny little jumps in understanding, the „lies-to-children”, the simplifications.
And one of these, for me, was how neurons get to connect.
And when I asked around, it turned out I wasn’t the only one missing this little bit.
I mean, yeah, we all know Hebb. „Neurons that fire together, wire together”, the neurology mantra in flesh. I can just imagine NeuroZen masters, walking around meditating neurologists, „fiiiiiiiiireeee toooooogeeeether” instead of „ommmm” going around.
That’s how neurons connect. When two fire at once, their connection becomes stronger. The synapse gains long term potentiation/LTP, by both an increase in the neurotransmitters produced, and an increase in the postsynaptic receptors for the neurotransmitters. Short-term it happens through utilizing some of the additional AMPA receptors available near to the synaptic membrane in the postsynaptic cell, long term by changes in protein synthesis and gene expression, in order to ensure a larger number of receptors. So far so good, in one way or another pretty much anyone who understands anything about neurology understands this bit. It might be more or less simplistic, you might have the AMPA receptors memorized or not, but you get the gist of it. (Although this begs the question: does Dale’s principle means that the increase in presynaptic neurotransmitter production influence not just this single synapse, but also the concentration of neurotransmitters in all synaptic connections of the presynaptic neuron? Or perhaps does a reverse mechanism happen, with other connections of the presynaptic neuron being stripped bare of their neurotransmitters? Something I need to check! )
Now here’s the kicker: how do neurons that AREN’T already connected connect? How does their “fire/wire” rule go?
I mean, that still applies, doesn’t it?
Actually, it doesn’t apply in the original Hebb’s law, since that required the neurons to be connected already and a couple of other things as well. But still, the general principle is how we tend to explain things like associative learning, conditioning, etc. The direct wording of Hebb’s law can’t apply there, because then we’d need every neural network to be connected to every other neural network at once, and that’d be a mess that wouldn’t really work all that well. But the general principle seems to apply.
And when I started asking around, no one could tell me how exactly.
How does a neuron know to extend new axonal or dendric connections towards another neuron, so that they might actually form a synapse between the two? Because this has to happen somehow, and this has to happen in a fairly structured way. If it didn’t happen, if we were only left to the remains of the synaptic pruning we got since birth, then our neural networks would become pretty much immovable after were what, twelve? I mean yeah, I’ve met these people, and so have you, but they’re in the small minority… well, large minority… well, OK, a small majority. But even so, there are quite a few people who aren’t that way. And if the neural growth was haphazard, then we’d get a far more random jumble of associations, then the fairly clean ones we tend to get in any conditioning/associative learning experiences.
So I knew there must be some kind of system for neurons to detect other neurons it should be „aiming for” in their development.
And – this was about two or three years ago as of writing this – no one I asked could tell me what it was. I’ve reached out to a lot of people I’ve considered very competent in the field, including some of my old Uni professors, and no one could direct me. Nor could I find the answer in any of the textbooks I’ve referred to.
As it turns out, the answer is fairly recent, as articles about it only started to appear in the 2000s. Since it took me some time to find it back then, I figured I’d make this article to share it now, making the whole thing easier to find for others.
We’ve known from about 1890 that the endings of both the axon and the dendrites are covered in something known as a growth cone. Like the tip of the branch, this is the part where the cell can further develop and extend, either in original growth, in regeneration, or as a reaction to some factors (including a decrease in overall electric activity which causes a form of synaptic scaling – hence explaining the „devouring” of unused neural pathways of amputated body parts by nearby active structures). But how do the growth cones know how exactly to grow? The brain is a complex three-dimensional web of interconnected neurons, so „hitting” the right point is definitely too hard if the neurons just kept on growing until they hit anything at all.
Now, the synaptic scaling process is still a piece of the answer – even if a neuron forms a new synaptic connection with a different neuron, if the target neuron already has too many connections, it will tend to remove the weakest ones, and this includes the most recent ones. The scaling goes both ways after all – it goes for more synapses when it starts with too few, but for less, if it starts with too many.
But synaptic scaling is not everything. As it turns out, the tips of the growth cone constantly produce structures called filopodia, and these react to specific chemical attractants and repellents. These chemicals are produced by both cells at the target area, and by so-called guidepost cells along the way. There are suggestions that the system for such targeting is fairly robust, especially in early development (and its limitations in later life might explain why spinal cord injuries and the like are so hard to fix).
(There is one other process for directing cells, the growth along the radial glia, but since it mainly refers to embryonic development, it’s really not all that interesting for us in this topic.)
Now, the more perceptive of you will already have noticed that while what I’ve given is AN answer, it is not THE answer to the question I’ve originally asked. Or at least, not necessarily.
It might be possible, that the interaction of the three systems:
– a pre-set network of attractor/repellent chemical pathways for guiding axon/dendrite growth,
– the synaptic scaling mechanism of homeostatic plasticity, which limits the max connections and promotes new connections if existing neural excitability falls below what is the baseline for the network,
– and the good old Hebbian mechanism for strengthening existing neural connections,
is in fact enough. That the interplay of these systems does answer our basic question, and simple enough rules result in sufficient flexibility to explain what is observed.
But, obviously, the process might be a bit more complicated. There isn’t that much literature about it, unfortunately, so what I’m doing here is more of an (un)educated guess, but since guidepost cells tend to be neurons which have yet to develop an axon – but already have dendrites and can receive signals, a slightly more complex – but better targeted – system can be imagined, where guidepost cells would be a part of a wider neural network, the stimulation of which „calls forth” further cells to build connections. This, in my understanding (Again, limited, so don’t take my word for it. I mean it.) would be a more effective explanation of how we can rapidly learn to connect fairly new information, even across huge (neurologically-scaled) swaths of neural space. A strongly agitated system produces a stronger „call for connections”, hence mediating the faster development of new and broader associations.
So there you have it, a quick summary of one part of neural connectivity I’ve yet to see described in a textbook about the brain, but which really should be given out there, along with the classic Hebbian principle. Hope it’s useful 🙂