That Neuroscience Guy - What is EEG?

Episode Date: March 28, 2021

"Brainwaves" help us explain how the brain works, which is why I study them all the time. But what is a brainwave? In today's episode, I discuss what EEG is and how we use it to understand the brain.��

Hi, my name is Olaf Kregolsen, and I'm a neuroscientist at the University of Victoria. And in my spare time, I'm that neuroscience guy. Welcome to the podcast. There's a lot of talk about brainwaves these days, what neuroscientists call the electroencephalogram or EEG. You might have heard about neurofeedback, going out to have your brainwaves optimized, or you might have heard about the fact that your brainwaves change during meditation. Today on the podcast, we're going to talk about the neuroscience of brainwaves. Let's start with how EEG is generated. All of your thoughts, emotions, and feelings are the product of neurons firing in your brain. The human brain is comprised of billions of neurons,

and when these neurons fire, that's what gives rise to our feelings, our thoughts, and our emotions. to our feelings, our thoughts, and our emotions. Typically, EEG is associated with the firing of cortical neurons. These are neurons that are in the top level of the brain, closest to the skull. And typically, EEG signals come from a specific type of cortical neuron called a pyramidal cell. They can be aligned perpendicularly or horizontally to the surface of the brain. And like I said, they're in that top layer of the cortex near the skull. Although EEG does come from deep sources within the brain as well. It's just the signal from these deep sources isn't as clear as the signal that comes from the neurons that are closer to the surface of the brain.

Now, when these neurons fire, there is an imbalance of electric charge. So in brief, the way a neuron fires, if you don't remember from a previous episode, the neuron takes in input, which is essentially electric charge. That charge is summated within the nucleus of the neuron, and then the neuron releases an action potential, a signal that propagates down the axon to the axon terminal where a neurotransmitter is released. That neurotransmitter goes across the synapse and binds to the postsynaptic neuron, and little electrical signals called EPSPs and IPSPs, excitatory and inhibitory postsynaptic potentials are generated. Now, when the neuron is doing this, like I said, there's an imbalance in charge.

There's possibly more positive charge near the cell body and more negative charge near the axon terminal, or vice versa. There's more positive charge at the axon terminal and more negative charge at the cell body. Think of it like a magnet with a north pole and a south pole, but in this case, it's an electric magnet, if you will, or what's called an electric dipole. It has a positive end and a negative end. Now, EEG doesn't pick up just one little dipole from one neuron, but it picks up the summation of hundreds of thousands, if not millions of neurons that are all aligned and firing close to the same point in time. And these little positive and negative charge discrepancies, the dipoles, well, if the neurons

are all aligned in a similar fashion and they're all firing roughly at the same time, these little dipoles sum up to one large dipole, what's called an equivalent current dipole. And this is what we're seeing on the surface of the scalp. Now, as different regions of the brain fire at different intensity levels, regions of the brain fire at different intensity levels. Now as different regions of the brain fire, in other words the dipoles change, then you see changes in the pattern of electrical activity on the surface of the brain or the EEG signal. And that's why the EEG signal goes up and down if you look at the long squiggly line. It's the result of those little dipoles changing and the large dipoles moving across the brain. And that's what we see.

Well, how do we see it? Basically, if you think of the way it works, you put an electrode or electrodes on the surface of the brain, well, realistically, on the surface of the head, just like you would with ECG. Everyone's seen a hospital show where they put electrodes on the surface of the brain, well, realistically, on the surface of the head, just like you would with ECG. Everyone's seen a hospital show where they put electrodes over the heart, and you can see the heart muscle contracting. And of course, if things go poorly, you get the flat line and they break out the paddles and charge. But that little electrical signal you see with ECG, the electrocardiogram, is essentially the electrical activity in the heart muscle as it contracts. With EEG, you're measuring those dipoles. But how does it actually work? Well, that electrode is sitting on the surface of the scalp, and it's a lot like a voltmeter.

Essentially, what you're measuring is what physicists would call a potential difference, or a voltage, a difference between the electrode and a reference point, which is just another electrode. So if you think of a typical EEG setup, where people wear either 32 or 64 electrodes arrayed over the top of the head, you're basically measuring the voltage between all of those electrodes and a reference electrode, which is typically also on the cap. And those voltages change through time. They fluctuate up and down as the underlying neural activity changes, and all of those little dipoles shift and reorient, and the larger some dipole moves around as different parts of the brain are active. So that's how the electrical activity gets to the electrode. But what's actually happening from there?

Now, I don't want to really go too far into how the hardware works, but the first step is what's called analog-to-digital conversion, or A-to-D conversion. Essentially, the actual neural signal is analog, which means it's continuous. And at the electrode, you can only measure data so often. What's called a sampling rate and is typically measured in hertz. EEG is typically sampled at 500 or 1000 hertz, which means you're taking 500 to 1000 measurements per second. That's the first step. So you convert the analog signal to a digital signal through an electric circuit. The next step is amplification. Essentially, you want to make that little tiny electrical signal a lot bigger.

Even though there might be millions of neurons involved, what you're measuring is occurring in microvolts. And for comparison's sake, the ECG signal is in millivolts, orders of magnitude bigger. Just like you amplify music to get a louder sound, you amplify the EEG signal to help it stand out from the other electrical noise. For instance, the wires in the wall, a phone that's nearby, or even the computer you're running your experiment on. In modern systems, there's actually a whole bunch of amplifiers.

They're typically on the electrodes themselves in what's called an active system. And that helps get a cleaner EEG signal. In the old days, if you look at Google Images or something, EEG chambers typically were built inside what's called a Faraday cage. A bunch of wires that are used to basically soak up all of the surrounding electrical interference and keep it away from the EEG system itself. Then that EEG signal, which is the result of all those dipoles moving and summing, is recorded on a computer. So what's next? What do you do when you get the squiggly lines? So if you hear about EEG, typically you hear about things like delta power, theta power, alpha power. What is that? One of the most common techniques for analyzing an EEG signal is to submit it to what's called a fast Fourier transform. And the fast Fourier transform takes the electrical signal, which is

a voltage through time, and essentially what it's doing is it's breaking that down into the frequency of the brain oscillations. Within one segment of EEG there are oscillations at different frequencies. One way to visualize this is that slow oscillations look a lot like a long slow rolling wave on the ocean and these are very low power oscillations, typically between half a hertz to about three hertz, what's referred to as delta power. And there's neural activity up to about 80 to 100 hertz. There's still some debate. As the oscillations go up and down a bit faster, then you've got a higher frequency of oscillation. So what we typically see, as I said, are delta oscillations between about half a hertz and three hertz, theta oscillations between about four hertz

and seven hertz, alpha oscillations between eight hertz and 12 hertz, beta oscillations between 13 and 30 hertz, and gamma oscillations which go from 31 hertz up to about 100. Like I said, there's still some debate about what's going on in the gamma range, but a lot of exciting new research is showing that the brain does have oscillations in that 31 to 100 Hz range that are interesting and do convey specific neural information. In terms of what you typically see with EEG oscillations, perhaps the most common one is called alpha suppression. If you're looking at the human EEG signal and you have someone with their eyes open, you typically don't see a lot of alpha power,

so the oscillations are lower in the 8 to 12 hertz range. But as soon as they close their eyes, you see an increase in power, or you see greater oscillations in that 8 to 12 hertz range. And those oscillations in the alpha range are associated with the allocation of visual spatial attention. What you need to understand about it though is it's actually low alpha power or reduced oscillations that are associated with focusing your visual attention and it's actually an increase in alpha power that is associated with you not focusing your attention. Another common example you see of oscillations is what's called

frontal theta power, or oscillations over the frontal part of the brain between 4 and 7 hertz. Theta oscillations are typically associated with the engagement of the prefrontal cortex. are typically associated with the engagement of the prefrontal cortex. So think of engaging working memory or concentrating or making decisions. A simple way to evoke frontal theta power is to do the following. Imagine you had an EEG system on your head right now and I said, okay, starting from 10,000, count down by 13s as quickly as you can. When you do this, you see an increase in frontal theta power as the brain engages to take on this challenging mental task. If you have read about meditation, you'll see that there's also changes in the beta and gamma range during meditation.

So the meditative state is associated with a different pattern of oscillations than a state when you're just sitting there quietly with your eyes closed. Just so you know, you also see differences in delta, theta, and alpha power when people are meditating. And you can try this out pretty easily for yourself at home. Right now, there's a really cool device called the Muse, which is an EEG headband made by a company in Canada called Interaxon. And the Muse helps you be mindful. The Muse is a very low-cost EEG system that you can buy at Best Buy for about $200 to $300. And what it's doing is it's measuring EEG power in the delta, theta, alpha, beta, and gamma ranges. And it's basically giving you a form of feedback that

helps you train your mindfulness. The Muse is trying to capture the oscillations indicative of a mindful state. You also hear about EEG in terms of sleep. We mentioned this on a previous episode. If you go into a sleep lab to have your sleep cycle evaluated, they'll measure EEG and they can characterize the different stages of sleep as changes in power in the delta, theta, alpha, and beta and gamma bands. The last thing I'll mention as we wrap up our talk about the neuroscience of brainwaves is what's called an event-related potential. I like these in particular because a lot of the research we do in my laboratory involves them. Essentially what an event-related potential is, or what's known as an ERP,

it's your brain's response to an event. Right now, for instance, imagine you heard a series of tones. Beep, beep, beep, beep, boop, beep, beep, beep, boop. Your brain would respond slightly differently to those tones because they sound slightly differently. And if you were measuring EEG and you looked at the little bit of EEG right around the time of each of those events, you'd get an ERP or an event, the tone, the beep or the boop, related because the response is related to that tone, potential or a change in voltage. These ERPs are really cool. You can use ERPs to study all sorts of cognitive processes. Quickly, I'll tell you about the first one that I began studying when I began my research career. If you have people learn something and you give them feedback, positive and negative, so for instance, I teach you a new language and I show you a word in a foreign language and I ask

you what it means, and then I say right or wrong, you see a difference in the ERPs, what's being called a reward positivity. Essentially, it's your brain processing positive feedback. What's really cool about the reward positivity is that early in learning, when you're not sure of the answer, you get a large reward positivity when you're told you're right. But once you've learned the correct association between a word and a meaning, if you stick with my example, then the amplitude of the reward positivity changes. In other words, it's sensitive to how well you know something. The reward positivity is also really cool because you see differences in it depending on different states. For instance, people with Parkinson's disease

have reduced reward positivity, as do people that are experiencing depression. So what is EEG? EEG reflects the neural activity of hundreds of thousands, if not millions of neurons close to the surface of the brain. And because there's a little imbalance in charge, they're more positive on one side and more negative on the other. It creates a little electric dipole. And by placing electrodes on the head, we can measure changes in large groups of neurons that are aligned and firing in sync. And in terms of interpreting EEG, we can use a little bit of fancy math and look at the evoked power in the delta, theta, alpha, beta, and gamma ranges to draw some insight into brain function. Or we can look at little chunks of it associated with events in the world and measure what are called ERPs. My name is Olof Kregolsen, and I'm that neuroscience guy. I hope you enjoyed this

episode. You can check me out on my website at www.olivkrigolson.com or on Twitter at That Neurosci Guy. Thanks for listening, and I'll see you on the next episode.