Neuroplasticity - What is it?

Neuroplasticity is a real thing that is shrouded in mythology and used by companies to market products, but what do we actually know about neuroplasticity?

Please leave feedback at https://www.psydactic.com or send any comments to feedback@psydactic.com.

References and readings (when available) are posted at the end of each episode transcript, located at psydactic.buzzsprout.com.  All opinions expressed in this podcast are exclusively those of the person speaking and should not be confused with the opinions of anyone else.  We reserve the right to be wrong.  Nothing in this podcast should be treated as individual medical advice.

Transcript

Copy Transcript Copied

SPEAKER_00 0:04

Welcome to Psychactic. I am Dr. O'Leary, a fellow and child and adolescent psychiatry in the National Capital Region. This is a podcast about psychiatry, psychology, and neuroscience. I'm the creator and producer of this podcast, so everything here should be considered my own opinion and my responsibility. Nothing I say should be considered the opinion of the Department of Defense, the Defense Health Agency, the Federal Government, or the Galactic Empire. Today, I'm gonna struggle with the term neuroplasticity, which is one of those terms that can have very specific meanings, but can also be used in popular jargon, especially on social media, to sound fancy, but not really mean anything in particular. I hope by the end of this episode, you will be better equipped to understand when someone is talking about meaningful neuroplasticity, or they're just trying to sell you something by saying fancy words. Neuroplasticity describes the nervous system's capacity to alter its structure and its function in response to either internal or external stimuli. For most of the 20th century, the prevailing medical and social dogma supposed that the adult brain was essentially just a biological machine. It had fixed wiring. It couldn't really change. Once the brain solidified as an adult, there was little point in trying to change it because it was just stuck. Adults could learn things by maybe reinforcement of already existing neural pathways, but that was it. You can't really teach an old brain new tricks. In a very vague way, we can think of neuroplasticity as the ability of the brain to learn new tricks. Any brain, old or new. Neuroplasticity is not merely the creation of new neurons in the hippocampus, which could help us to reliably create new memories. It's not only that, it requires adaptations of each individual neuron. It requires adaptations among networks of neurons that support that neuron, and adaptations in larger brain networks that share and refine information, as well as the rest of the body's ability to create and maintain the states that are needed to execute that learning. So neuroplasticity really happens on different levels, all the way down to a molecular level and up to an entire organism level. It's not something that you can do by merely buying or using an app or playing a brain game. Neuroplasticity is not broadly generalizable. Whatever skill you want to learn requires practicing that skill and not how much you practice a game, unless the game is exactly the skill you want to learn. Many geniuses are really bad at doing anything other than being a genius. And those of us, like myself, who are really far from being geniuses, can make ourselves smarter by studying, but not by listening to Mozart or downloading an app. And that's because the way neuroplasticity works is to refine a very specific thing. It does not give you broad abilities, like that app who's promising to expand your mind and open your brain. Here are a list of some claims that I've read that attempt to use the idea of neuroplasticity in a way that's probably good to sell their products, but it's not at all based on actual neuroscience. Products often claim to be able to delay or prevent neurodegenerative diseases, like dementia. They may claim to rejuvenate the brain. They may claim to increase your IQ, improve your memory, unlock your potential, or even to treat conditions like ADHD and autism. Some of these unsupported claims even come from researchers who are working at prestigious institutions. And I hope by the end of this podcast you will be able to identify when a claim is likely overstated and improbable, and exactly why that is the case. Why is that claim unlikely to be true? Our brains need to be able to change, but they also need to be able to stay the same so they don't degenerate into chaos. Thinkers like William James, as early as the late 19th century, proposed that the organic materials in the brain necessarily need to change in response to experience, so the brain's actual structure has to change. However, too much change is bad, so there needs to be some mechanism by which change was managed. Other thinkers like Jerry Konorsky in the 1940s proposed that there must be a difference between innate reflexes, things that are resistant to change, and adaptive reflexes, things that change to or adapt to new situations. Konorsky used the term neuroplasticity to describe this. John Hebb, a year later, published a paper describing what is now one of the most famous terms in neuroscience, Hebian learning. Hebb proposed the idea that neurons that fire together wire together, and this is based on his reviews of previous experiments in the literature. Many experiments after this confirmed that stimulating neurons causes those neurons that were stimulated to connect more easily and to be stimulated together more frequently, and to fire together using less stimulus than was needed in the beginning. This is one of the processes that can help result in things like memory formation and learning. Many experiments since then have shown that experience can create novel and enduring relationships within the brain between different neurons. The problem for a model, as simple as the Hebian learning model, is that, well, mathematically, if that's the only way you learn, then it's not enough. Merely reinforcing connections is not enough for learning. Because if you keep reinforcing connections, then these connections become basically exponentially larger or more reinforced over time, and then the system itself will become unstable. So that can't be the only way we learn. Hebbian learning is kind of a brute force method that is limited by the fact that we don't have an infinite number of neurons. And the problem with a brute force method like this is that you also run out of energy or computing power. It becomes less and less efficient as you scale it up. There are always give and take in the brain. Strengthening one connection might mean that other connections are weakened. A purely Hebian learning space would learn with absolute certainty and then basically outgrow the brain and collapse. So there needs to be more than just Hebian learning. What keeps the brain from falling over the edge of that chaos is an exquisite dance between neurons and their associated cells. They're all trying to create predictive, stable states. Now the brain is not a computer like the computers that we have. It is an association of billions of living, interacting computers, each of which has a very high computing capacity itself. So the dynamics of the brain are such that it is always teetering, basically on the edge of chaos. It can avoid chaos because all of the cells in the brain are primarily interested in predicting their own microclimate and ensuring their own survival. It's an ecological system where each player is contributing to the function of the whole by optimizing itself. Plasticity can happen by changing how various neurons function, even without adding any new neurons or new connections. And all of this is governed at a very high level, or you could say the very basic level, by an energy balance. If any one part of the brain becomes too selfish as using too much energy, the rest of the brain will suffer. And in the end, the selfish part will also suffer. There's only so much energy to go around. A neuron might be sending a signal, like glutamate, to a neighboring neuron, and that signal is saying, go, go, go, fire, fire, fire. But that neuron can only go so much before it's exhausted. So that neuron has to be able to dial down the volume that it's receiving so it can maintain a stable relationship with its neighbor. Having limited capacity and limited energy resources is also why the brain can't deal with all of the information that's coming at it at once. All of those signals coming from the inside and from the outside, your brain is not processing all of that. It's only processing a select bit of that. And the way that it constructs reality that is far more efficient than trying to process everything that's coming in, is to predict reality or extrapolate reality from imprecise data. If it tried to process all the data coming in at once, it could potentially get a better picture of reality than it does, but the energy cost would be so much that we could not sustain it. But I'm getting a little off into left field. I wanted to talk about this energy balance because I think it helps people to understand that the brain is a collection of different cells that are each trying to survive and optimize their own environments, and they don't necessarily care what's going on in another part of the brain. But let me get back to neuroplasticity. The first two types of neuroplasticity I'm going to talk about are referred to as Hebian plasticity, because they appear to play by that rule that neurons that fire together wire together. They also extend the rule a little to state something like: neurons that don't fire together unwire. So this is a kind of plasticity called long-term potentiation and long-term depression. So to explain this right, I'll need to review some neural anatomy. A neuron can be put together in different ways. There is a cell body. If you're reading about it, you'll see that called a soma, and that's usually an expanded portion of the cell, and it contains most of the DNA and other organelles required to keep the cell alive and functioning. It's basically the operation center of the cell. Projecting off this cell body is something called an axon, and it's like an insulated wire that sends signals to other neurons or sends signals to muscle cells or other cells in the body, dumping different chemicals into the space between cells. Axons can communicate with thousands of other cells. A single axon can communicate and split off and communicate with thousands of other cells, which means that a signal sent to one cell may then be sent to many others. Now, not all cells the axon communicates with will receive the same amount of signal. A firing neuron could send a slightly different signal to different cells, even though it's one neuron firing. Most of the time, the terminals where these axons end dump their chemicals into structures on the receiving cells called dendrites. Dendrites are like the small wires sticking out of the cell that are trying to receive signals, not send them. Dendrites themselves have little spines on them that receive the signals when they're near that axon terminal. Now the space directly between the tip of the dendrite and the axon is called a synapse. Synapse just means joining together. So the synapse joins an axon from one neuron with a dendrite from another. So let me quickly review that. A neuron has a cell body that has projections. The projections either send information or receive information from other neurons. Axons are the wires that send information, and dendrites are the wires that receive information. The place where the information is exchanged between the neurons is called a synapse. And signals here are sent via chemical messengers, collectively called neurotransmitters. The chemicals that the axon dumps include things you may have heard of, like dopamine, acetylcholine, serotonin, GABA, and glutamate, among many, many others. So these neurotransmitters either potentiate the downstream cell, that means to make it more excitable, or they can depress the downstream cell, they can inhibit it, make it less excitable. Chemicals like glutamate are mostly excitatory, and some like GABA are mostly inhibitory. But it really all depends on what receptors these chemicals are hitting. Neurotransmitters like dopamine and serotonin can either excite or inhibit a neuron based on the kind of receptors that it's interacting with, so you can't tell whether or not it's going to be excitatory or inhibitory until you know what receptor it's hitting and where that receptor is on the cell. But I don't want to get too off into right field. Let's just move on to talk about how changes are made to neuronal connections. Some of the mechanisms of neuroplasticity include long-term potentiation. I mentioned that term before, and this is where a cell is maintained in a more ready state. But it can also be long-term depression where a neuron becomes less responsive to its neighbors. Long-term potentiation makes it easier to excite the cell after they receive inputs. So imagine a neuron dumps a whole lot of glutamate onto a dendrite. The dendrite's membrane has receptors. One type of receptors is called the AMPA receptor, and another is called the NMDA receptor. And these are basically channels that can either let ions flow into or out of the cell. Ions are just like charged particles. Before a neuron has been potentiated, it's just basically chilling there, conserving its energy, just listening for signals coming in. Its NMDA receptor channels are blocked by big old magnesium ions. When a cell starts to receive high-frequency signals from another neuron, this is because a massive amount of glutamate is being released from the presynaptic terminal from the axon, and this is activating these AMPA receptors. And it causes what's called a depolarization of the postsynaptic membrane. And all that means is a bunch of sodium ions begin to flow through the membrane and the cell fires a signal. This results in electrical signaling. When a cell fires, the change in the membrane potential is enough that it expels these magnesium ions out from these NMDA receptors and it allows them to open up. So a bunch of calcium ions can rush in through the NMDA receptors. Once the calcium ions rush into the cell, this triggers a cascade of signaling inside the cell, and this culminates in the insertion of more amporeceptors into the postsynaptic membrane, into the dendrite. So the dendritic spine, after it is stimulated a lot, will actually put more receptors on it to try to get a signal more easily. And the little spines that these receptors are on grow a little bit. So what this does is it increases the synapse's sensitivity to future glutamate release, so glutamate doesn't have to be released in such big amounts. It strengthens the connection and it makes the connection more efficient at sending signals. This is exactly what Hebb said. Neurons that fire together wire together. The downstream or postsynaptic neuron turns up the microphone that's listening for signals coming from the presynaptic neuron. So the presynaptic neuron doesn't have to be as loud in order to communicate. So this results in what we call long-term potentiation. And that means it becomes easier to make these two cells fire together. Another way to say this, if you're trying to use more technical terms, is that the cell receiving the signal increases the precision of the model it has that expects that a particular presynaptic axon terminal is going to talk to it. Now remember when I said that if the brain used only Hebian processes to increase connections, it would become unstable. So there are ways in which dendrites can be stimulated to increase their connections with their neighbors that don't require dumping huge amounts of glutamate and neuronal firing. This occurs when a group of these dendritic spines I was talking about are close enough together, kind of clumped up together, that even though they don't receive a big signal, they're all receiving a small enough signal that they depolarize, but only locally. So when they depolarize locally, they shoot those magnesium ions out of the channels, calcium flows in, and a bunch more amporeceptors are plugged in and they grow. But the neuron itself didn't fire. So if it's close to something it thinks it needs to be listening to, but it doesn't need to be firing, then it can grow. Cells can potentiate themselves without actually firing. So this is kind of a non-Hebbian learning process. But why do we need different methods of dendritic growth? Well, one reason might be that it's not always necessary for a neuron sending a message to cause the downstream neuron to fire. It could be sending messages that act more like feedback, saying to that neuron, you know, you're kind of onto something, keep up the firing. Or they could be sending error signals, calming a neuron and kind of pushing it back to a less excitable state. And if these signals are regular enough, then neurons need to listen to them. I mentioned before that neurons need to be able to turn down the volume of certain signals, not turn up the volume. And one way that we do this is a process they call long-term depression. So neurons can have axons from many different neurons talking to it at the same time, and it doesn't do it much good to strengthen and maintain connections between it and other neurons that are not really frequently or reliably sending it information. Neurons are going to function efficiently when they concentrate their resources on the places they're getting active information. Long-term depression is the process by which a neuron that's firing will actively decrease its connections to other neurons that are not active at the same time that it is. The way this happens is that some of the massive influx of calcium at the highly active synapse diffuses into less active parts of the neuron. So it starts to float around all this calcium, and it gets to places where there's not active signaling going on, but it gets there in high enough amounts that it can cause some changes. And high levels of calcium cause dendritic spines to grow. But low levels of calcium actually activate a process that results in the AMPA receptors being removed instead of being multiplied, and then the spines decrease in size. They atrophy. So long-term depression is when a cell is engaged in long-term potentiation with another cell and it's not getting signals coming from somewhere else. So it basically, by the process of calcium diffusing in small amounts to another part of the cell that's not firing, tells that part, hey, why don't you just quit listening so much? Why don't you conserve some energy here? Let's reduce making a whole bunch of receptors that we have to constantly produce and put back up there, and let's just chill. So the connections at that part of the dendrite to other neurons are weakened. And this is long-term depression. Now, another adaptability feature of the dendritic spines that I've talked about is they're not necessarily stuck in place. They can actually kind of migrate around a little bit. There are proteins that kind of hold them into the cell membrane, and the cell membrane is a little movable. It's not stuck in 100% stuck in place. Now, the bigger that they get, the bigger these dendritic spines get, the more that they actually do get stuck in place, because there's a lot more stuff that has to hold them there. But those smaller ones, those thinner ones that maybe they started to shrink down because they're in a place where they're not receiving much signal, those are actually kind of mobile. They can slowly change places. So this allows the neuron to basically search around for signals and forage for more information without having to create entirely new dendritic spines. It can just have them move a little bit. So all of this is to say, if we were to apply this to the adult brain, that even if the adult brain could never grow a new neuron at all, it could still learn. It could still learn by processes that change how neurons are connected to each other. So far I've discussed how neuroplasticity can happen at the level of the dendrite of a single neuron. And this can explain things like how we can learn something new very fast and make it useful in the moment. It can create a kind of context. The cell can change very quickly, start listening to one place more than it is listening to the other places, and fire more readily, and it can do this in a Very short amount of time. For example, we have cells in the hippocampus called place cells, and they help us learn and navigate in a new environment. Once we've learned an environment, we don't actually need a lot of uh navigation. We know what to expect, and we can just move through that environment because we're predicting where the next thing is going to be because we already know. But if you're in a new environment, you have to learn quickly where things are, or else you like, for example, just walk into a wall. So there are place cells in the hippocampus, and these adapt quickly by some of these potentiating mechanisms that I've talked about. But navigating an environment is only one way we navigate the world. We also have to navigate like a social context, and our brain has to adapt within a few seconds or even a few minutes to new social environments so that our actions are socially appropriate in different environments. Now, the more familiar an environment is, the more our brain can basically already use pre-formed pathways to work. But the more unfamiliar it is, the more our brain has to rely on neuroplasticity to make different decisions than it would have in the past. And we also have other parts of our brain, I'm not talking about a lot here, like the cerebellum, that really help us to do this. So it can help turn up or down the volume of different neurotransmitters in order to help the cortex pick the best decision. So cells in our cortex are already potentiated to a certain context. They are expecting certain information in a certain way. So they have they basically have a model that can be applied quickly and efficiently. Think about walking up the stairs. Your cortex already knows what's next, what the next thing to do is to step. Now, why are all stairs by code the same height? I think it's about seven inches. They're that height because you don't have to do work to walk up stairs if your brain can just apply the same model over and over and over again. You don't need neuroplasticity. You're just doing the same thing. You're reinforcing the model. You take one stair in a set of stairs and you change the height by an inch, and just watch what happens. People are going to be tripping all the time. And people get sued when stairs are not built in a regular fashion. So for our movements that require adaptation to a new environment, we use parts of our brain, like the cerebellum, to make immediate changes in the physical world, to estimate the next best thing, and to potentiate different models than the brain would usually use. You have to make impromptu adjustments. You have to adjust the precision of your different models in order to not be clumsy, socially or physically or cognitively. Because the brain already has these pre-formed, reinforced models, there need to be processes by which we can make new models, new memories, slower processes that help us adapt in the long term. And this often requires making new cells, or it requires making new connections that weren't made before. So the next type of neuroplasticity is one where you actually have to, at a whole cellular level, make big changes. There's only one place in the brain that we really know continues to have this kind of neuroplasticity into adulthood, and that's the hippocampus. The hippocampus is basically your memory center of the brain, but I'd rather refer to it as like the memory coordinator and not memory center, because I think that's maybe a little bit more precise language. Memories require the entire brain to produce them, not just the hippocampus, so it's not like memories happen in the hippocampus. The hippocampus coordinates the recall of brain states, and that's a memory. And it can only make new memories if it can form new connections, and potentially getting rid of old ones. Now, as we age, the hippocampus ages, and so it's not making memories as easily, it's not saving as much information, you might say. It's not as efficient at learning as it was in the past. And so there's just functionally a breakdown, because one, there's already a whole bunch of models built and memories built, so building new ones requires new space and more energy. But also cells themselves start to become like an old building. They just start to have lots of problems and can't really fix themselves. They just age. So far, I've talked about two kinds of neuroplasticity. I talked a lot about something called long-term potentiation and long-term depression. And that requires changes right at the junctions where cells communicate, and those changes can happen really fast so that you can adapt to new environments within some seconds or minutes of being there. But there's also memory formation, and this happens because the hippocampus is able to make new cells that can stimulate brain states that you have had before. Stimulating brain states you've had before does not generalize to creating lots of other brain states that you've never had before. So let me get back to talking about why brain training apps are probably just a waste of money and a waste of a lot of time. The changes that happen in the brain that create memories and that create potentiation are very, very specific. They're specific for what is being learned. The laws of physics do not allow them to generalize. If you do learn a broad set of different kinds of things, which is what a liberal arts education is supposed to do, expose you to many different kinds of things, this will result in the brain being able to have insights that it could not otherwise have. But there's no real mystery here. The mystery is why learning to play a game or watching maybe a swirly screen or meditating would help with anything, other than helping you potentially pay more attention to something, put more of your attention in one place. The only thing you ever get better at when you do something is that thing that you're doing. You don't get better at other things. Unless those other things are basically the same thing. So if I'm hammering nails, I could hammer something else with a hammer. But it's because I'm doing basically the same thing. But I don't get better at riding a bike by hammering nails. And I don't become smarter by playing games. On my phone. I become better at that game. Ask yourself, why would the brain waste uh time making a whole bunch of different things better when you're practicing a very specific activity? How would it even know to do that? It doesn't make any sense on any logical or neuroscientific level. The most helpful thing about some of the tasks that some of these games might have you do is that they help you to be cognizant of paying attention to certain things, whether it's details or memorizing lists or spatial arrangements or task switching. While you're only going to be getting good at the things that you're practicing, and you're not going to unlock anything exceptional, you can become good at trying to get better at things. So in that way, they might help. But don't think that you're going to get better at math by memorizing lists of numbers. Some future episodes I want to discuss another way that the idea of neuroplasticity is being used both for the benefit of humanity, but also maybe to just disabuse gullible people of their money over time. I'm going to discuss various chemicals that we know have some effects on remodeling synapses or promoting the formation of new connections or new cell growth in the brain. One set of these chemicals, which are in the news a lot right now, are called the psychoplastigens or the neuroplastigens. So a psychoplasen is something that causes rapid significant changes in the brain. Psychedelics are psychoplastigens. So things like LSD, MDMA, ketamine, and psilocybin can cause a cascade of different chemical reactions in the brain, which enhance the brain's ability to make certain connections. In fact, almost all the medications psychiatrists normally give, antidepressants, anti-anxiety medications, and antipsychotics will change the brain's synapses and their wiring. Changes usually happen over longer periods of time, whereas with psychoplastigens, they can often happen very, very quickly. If you want to take a really broad view of the term psychoplastigen, you wouldn't apply it only to psychedelics or even to other medications, but you could apply it to electroconvulsive therapy or TMS, because these have shown to change connectivity, functional connectivity in the brain. So you could call transcraniomagnetic stimulation a psychoplastigen, an electromagnetic psychoplasgen. You can even call different kinds of therapies psychoplastigens, because making you do something new changes your brain. So therapies like cognitive behavioral therapy help you to intentionally rewire your brain by challenging you to act in ways or to think in ways that are contrary to the current model that your brain has. So you could call therapies therapeutic psychoplastigens. Now I say all these things just to get the point across that your brain changes in response to lots of things. It changes in response to information that it gets from your environment, from your therapist, from chemicals that you put in your body, from electromagnetic energy. But it's impossible to just in general unlock your potential. You have to actually do things, very specific things, to get better at them. You're not going to become a genius because you have an app. I hope to do more episodes in the future, specifically discussing things like psychedelics and how they cause the brain to change. So stick around for some of those. It may take me a while, but I usually get there. I appreciate you spending time with me today, and you're welcome back anytime. I'm Dr. O'Leary, and this has been an episode of Psidactic.