What is a memory?

Tiny Matters

Have you ever had such a clear memory of something and then found out it never happened? We tend to think of a memory as a photo in an album that we can open up and access whenever we want. But it’s more like the pieces of a photo scattered in a stack of papers and you might only be able to access a couple of those pieces at a time. In this week's episode we're asking, "How do our brains form and store memories? And why do we lose them?"

Transcript of this Episode

Sam: Have you ever had such a clear memory of something and then found out it never happened? Or that it’s not real?

I know I have, and I bet you have too. Let's try something.

You know those delicious, flakey bright orange crackers? Cheez Itz? Imagine that box. Did you know it actually says Cheez It? Singular.

What about the Flinstones? Oh excuse me... Flintstones. There’s another “t” in there. Did you know that?  

Or the BerenSTAIN bears. Not BerenSTEIN bears.

It was never “mirror, mirror on the wall” it was “magic mirror on the wall.” And Darth Vader didn’t say “Luke, I am your father.” The line is “No, I am your father.” [audio from Star Wars: Episode IV – A New Hope].

Deboki: These are all examples of the Mandela effect—when a group of people have a false memory of something, whether it be the name of a cheesy cracker, a line in a movie, or a major global event. The Mandela effect got its name from a woman who said she remembered anti-apartheid leader Nelson Mandela dying in prison in the 1980s. She said she remembered news coverage of it, and even a speech from Nelson Mandela’s widow.

Nelson Mandela was imprisoned—for over 25 years—but after he was released in 1990 he went on to become the first president of South Africa, serving from 1994 to 1999. And he lived until 2013.

Sam: So the woman remembered wrong. But …she wasn’t alone. She found that a lot of other people shared those false memories. Conspiracy theorists might say this is proof of a multiverse—that alternate realities exist. But psychologists say it’s probably that our brains are suggestible—constantly influenced by the world around us and trying to fill in the blanks when pieces of a memory are missing.

We tend to think of a memory as a photo in an album that we can open up and access whenever we want. But it’s more like the pieces of a photo scattered in a stack of papers and you might only be able to access a couple of those pieces at a time.

Deboki: Although the concept of false memories and suggestibility is incredibly interesting, this isn’t a psychology podcast. Sam and I want to go smaller—tinier if you will. Because to even try to hypothesize as to why our memories can be so error-prone, we need to understand how we form memories in the first place.

Welcome to Tiny Matters. I’m Deboki Chakravarti.

Sam: And I’m Sam Jones. Today on the show we’re asking, “on a micro level, what is a memory?” We're talking with neuroscientists about how we form and store memories, and also how we lose them.

We started today’s podcast with the concept of false memory because, to me at least, it feels like the perfect illustration of how complicated memory is. And it calls into question the whole concept of reality, something that was a big motivator for neuroscientist André Fenton, a professor at New York University.

André Fenton: I have always been fascinated with how I understand what's real. From my early teens, various experiences of mine led me to wonder, like, how do I actually know this is an actual experience that I'm having? Because people around me and at another moment I could have what looked like the same experience, but a very different interpretation of it. And so I've always wondered, like, how do you know what's real?

Everyone thinks they know what a memory is. We all have the experience of memories, and we definitely have the experience of not having as good a memory as we might like most typically. But what a memory actually is, is up for debate.

Deboki: What’s not up for debate when it comes to memory is the importance of cells in the brain called neurons. No neurons, no memories. Actually, we wouldn’t have a lot of things—like even something as fundamental as breathing—because neurons coordinate things all around our bodies.

So here’s a little crash course on these amazing cells: To me a neuron kind of looks like a fried egg with a tail. You have the yolk, which is the neuron’s nucleus where its DNA lives. Then around the yolk is an egg white, which in the neuron is the cell body or the soma. And coming off the cell body are spiky projections called dendrites.

Then you get to the tail, which is called the axon, which will split off at the end into axon terminals.

Sam: Neurons are messengers of information, using electrical impulses and chemical signals to exchange information with other neurons. When a neuron sends an electrical signal down its axon, that triggers the release of molecules from the axon terminals. Those molecules are called neurotransmitters. So the neurotransmitters travel out of the neuron, into the space between neurons—called a synapse or synaptic cleft—and then bind to receptors on the dendrites of other neurons, triggering the next electrical signal, and then from there the cycle continues.

This is a very basic explanation of how neurons communicate, which is not by any means a one-size-fits-all scenario. Some neurotransmitters will actually block neurons from sending an electrical signal. Also, because we have billions of neurons, things can get complicated.

Richard Huganir: The brain works through electrical and chemical signaling between a hundred billion neurons. And so each of these neurons connect with each other about 10,000 times. So there are trillions of and quadrillions of synaptic connections in the brain and these form millions of circuits in the brain that underlie what we do—behavior or movements or emotions.

Sam: That’s Richard Huganir, a professor of neuroscience at Johns Hopkins University. Richard has spent his career asking…

Richard: How do you encode a memory that lasts decades in a squishy three pound organ?

Deboki: It turns out that forming or encoding a memory doesn’t just have to do with the connections between neurons, it has to do with the strength of those connections. Stronger synapses means more signaling between neurons and is called long-term potentiation. Long-term potentiation is what’s thought to make learning and forming memories possible.

Richard: When you learn something new, you’re forming these new connections and you're strengthening certain connections between neurons and weakening other connections, and it's sort of sculpting a new circuit in the brain that actually physically encodes the memory. So when you recall a memory, you actually reactivate that circuit and that brings forth the memory.

Sam: At this point you might be wondering, what does it actually mean for a synapse to be “stronger” or “weaker”? Richard and his lab study exactly that—synaptic plasticity, the process where that strength actually changes.
Richard: What we found and many other laboratories have contributed to this is that when a synapse gets stronger during learning, it's because the neurons add more receptors to the synapse. You have more receptors and so the signal is stronger.

And when synapses get weaker, the neuron removes receptors from the synapses. And so this process of dynamically adding and removing receptors is really key to learning and memory. And so one of the coolest things we're doing now is that with incredible advances in microscopy, we can actually image molecules inside the brain of a mouse as it's learning something.

Sam: So they’ve genetically engineered mice to have neurotransmitter receptors that have a fluorescent green protein incorporated in them. These receptors actually glow green!

Richard: All those billions of synapses I talked about are all glowing green in this mouse. And so we can zoom in and image about a million synapses at a time and watch the levels of receptors at each of these synapses while the mouse is learning something. And so we can look for changes that may underlie the learning process. So it's really watching learning occurring in real time in the mouse, in the brain at a molecular level.

Sam to Richard: So I'm curious, what kinds of tasks are you having these mice do that’s, like, a learning task?

Richard: We have lots of simple learning tasks. This sounds funny, but teaching a mouse to reach out and grab a pellet and bring it to its mouth. That's not how they normally eat something, they just chew on a pellet. But in this case you make them reach out and grab it, and this is a motor learning task. So the learning is occurring in the motor cortex and so we can image their motor cortex and see what’s changing.

Deboki: Your motor cortex is a part of your brain that’s used to coordinate movement, and also to learn tasks that involve movement. It’s located right behind your frontal lobe—the part of your brain that starts at your forehead and ends around your ears. If you put a headband on the top of your head, it pretty much perfectly rests along your motor cortex.

In addition to looking in these mouse brains as they’re learning the pellet grab, Richard and his lab also do a fear conditioning task with the mice where they’ll play a sound and then pair it with a shock. The shock is mild—the mice are not harmed by it—but they learn to associate the sound with the shock and soon the sound alone is enough to make them freeze.

Richard: We can watch in different regions of the brain that we think are involved in the fear conditioning process and watch synapses get stronger or weaker and inform novel circuits.

Deboki: So that’s pretty wild. They’re basically seeing the brain rewiring itself in real time.

Sam: Yeah. And it is incredible to think that that rewiring happens throughout our entire lives.

Richard: You're learning even when you're 80 years old. One of the amazing things is the brain is so plastic, but then when you're 80, you still remember, you know, when your mother tucked you into bed at night or something like that. That's one of the more fascinating things about memory, I think, is how do you make that memory stable in such a dynamic brain?

Deboki: Some memories don’t stick around long, but some stick around for what feels like forever.

Sam: Yeah, the number of boy band songs I learned when I was ten and can still sing every word to is… a large number that I will not openly admit.

Deboki: I mean, who among us can judge? But it’s funny because there are things like Backstreet Boys lyrics that feel so vivid and permanent…but could I tell you about the first time I heard a Backstreet Boys song? No, definitely not.

So we asked André what’s happening in our brains when it comes to forming memories that don’t last all that long versus ones that you feel like you couldn’t forget even if you wanted to.

Sam: For the record, I’m perfectly happy knowing all of the words to I Want It That Way.

André: People who study memory recognize that memory is actually not a thing, it's a process. And so if you're going to study that process, you have to name the different parts of the process. In terms of the vocabulary, one way of classifying memories is in terms of what we say is long and short term, and there's even intermediate term memory. What you have to recognize is, when we gave it these names, we did it without measuring very much just sort of through introspection and observation.

And so, you know, it's clear, you can learn something like, you know, what you had for breakfast today. But you don't remember what you had for breakfast a week ago. And you probably remember your first kiss or when you learned to ride a bicycle. When I talk about a short term memory what I’m thinking about are the protein modifications that underlie the storage of the information. So your neurons, they do something, that means you have an experience, and if you're going to remember for a short amount of time, that experience, what we understand is that proteins that already exist, that already are there at the time of the experience, become modified. They get phosphorylated or they get dephosphorylated.

Sam: There are tons of different proteins hanging out in our neurons and they’re used for a bunch of things, including the essentials like cell growth, not just memory. Phosphorylation or dephosphorylation of some of those proteins make encoding a short-term memory possible. When a protein is phosphorylated or dephosphorylated that means a tiny ion called a phosphoryl group is added or removed from it. In the case of short-term memory, addition or removal of this tiny group to certain proteins in the neuron can change the shape of those proteins. Might not sound like a big deal, but it can actually cause a big increase in the number of neurotransmitters being released, increasing signaling between neurons and making that synapse stronger.

Think of it like the neurons are talking to each other more and forming a tighter bond in the process—kind of like people do.

André: But it's not very long lasting because proteins don't last very long, they get broken and they get degraded and they get, you know, modified again. And so we think of short term memory as those proteinaceous modifications. Then we can talk about long term memory in the other extreme.

Deboki: So short-term memory is like that person you talk to for hours and spill all your life’s secrets to at a party, but then you never see them again.

What’s important is that short-term memory involves changing proteins that already exist in the neurons. With long-term memory, instead of working with proteins that already exist, our neurons are actually creating new proteins. Researchers are still trying to work out all of the different proteins important for forming long-term memories, but one that's been a big focus of André’s research is PKM-zeta.

In the mid-2000s, he and fellow neuroscientist Todd Sacktor figured out that PKM-zeta shows up when neurons begin signaling more to each other.

Sam: PKM-zeta stabilizes neurotransmitter receptors called AMPA receptors, keeping them at the synapse between neurons. More AMPA receptors at the synapse means more signaling from one neuron to another, which means more opportunities for that memory circuit to be encoded and stick around.

Unlike the proteins involved in short-term memory, PKM-zeta seems to last a while in the neuron because it’s stabilized by other proteins. So the big difference between short term memories and long term memories is actually a difference of short lived proteins versus the production of new, long lasting, stable proteins within the neurons encoding those memories.

Deboki: So now let’s get a little philosophical for a second. Yes, memories are bits of information that you’re able to recall because they’re encoded through neurological changes to your brain. But they’re also much more than that.

André: Memories not only are items that you can recollect, they're items that change how you will be in the future and how you will process the next experience that you have. And the reason for that is because the biological process of storing that information, to the best of our knowledge, is a modification of the connections between the neurons that you use, not only for storing memory, but for having experiences.

Everyone surely is aware of some of the memories that they have, that they almost never recollect, but they know have guided them in a surprisingly strong way for a lot of their lives. Very often this is easy to recognize with a traumatic memory. But there are also very joyous memories, wonderful memories that one can have from early in one's life that you don't actively recall and think, oh let me enjoy the reminiscence. Rather those memories guide you and have helped determine who you happen to be, what kinds of choices you make, in a very wide variety of things.

Sam: Our memories make us… us. They shape our understanding of the world and help us make decisions. But what about when we start losing them—when we start to forget?

Deboki: Until about a decade ago, scientists were—for the most part—convinced that forgetting was a process that just happens to memories over time, a kind of “use it or lose it” scenario where if you don’t recall a memory enough it decays.

But then researchers started uncovering evidence in different animal models like fruit flies and rats that brains actively forget memories by destroying AMPA receptors—the receptors we just mentioned are so important for encoding a memory and helping it stick around.
André: You can think of forgetting as a loss of your memory, but forgetting is actually a great way to refine the information that you have so that you can use it in a purposeful way.
Sam: Forgetting could also be incredibly helpful—even life-changing—in the case of something like a traumatic memory. In 2010, Richard and fellow neuroscientist Roger Clem found that in mice undergoing fear conditioning—that training that pairs a mild shock with a sound—there’s an increase in signaling in the mouse’s amygdala.

The amygdala is the part of our brains that controls our emotions, including fear. Richard found that if they actually retrained the mice within a certain window of time by no longer pairing a shock with the sound, the mice would actually stop freezing. Meaning they weren’t afraid anymore. The scientists were able to see that the mouse amygdalas were rewiring themselves, which stopped that fearful memory from being encoded long term. That’s great for the mice, and it’s also potentially great for humans too because it opens up the possibility that treatment, whether behavioral therapy or medication, very shortly after an incident could mean that traumatic memory won’t stick.

Deboki: So forgetting is actually a very normal and very important thing that our brain does. But there are other types of forgetting that are harder to accept, like the kind of forgetting that happens in neurological diseases like dementia and Alzheimer’s.

Richard: What's known that occurs during Alzheimer's is one of the first things that happens is a sort of a dysfunctional regulation of synapses. And so the synaptic communication is compromised. And synaptic plasticity, the ability of synapses to change and to get stronger and weaker, is also compromised in Alzheimer's disease.   

Deboki: People with Alzheimer’s don’t have the synaptic plasticity they once did, and that affects their ability to both form and retrieve memories.

In later stages of the disease neurons start dying, which of course severely disrupts brain function well beyond just memory.

Richard: One of the more rewarding parts of my career has been, you know, we've been studying how learning and memory occurs and what genes and proteins are involved in controlling learning and memory, and what's become apparent in the last 10 years is that these same proteins and genes are involved in neuropsychiatric diseases. Not only Alzheimer's disease but intellectual disability, autism, and schizophrenia. And the proteins we’ve been working on for 20, 30 years are really key to this. So it's really exciting to try to now, since we know so much about those proteins, use that knowledge to develop therapies for these disorders.

Sam: Although memory and so much of the brain remains a mystery, it’s really incredible to think about how much things have progressed over the last 50 years. Neuroscience is a pretty new discipline within the sciences. I mean yes, people have been curious about the brain for millennia, but the field didn’t really take off until the late 1900s. And now there are so many tools available to neuroscientists that didn’t exist before. Things like fMRIs, that measure changes in blood flow in the brain, allowing scientists to draw some conclusions about brain activity. I mean, fMRI machines weren’t invented until 1990. You also have things like cell sequencing to identify new types of neurons in the brain. You also have new imaging technologies…

Deboki: The glowing mouse brains.

Sam: Yes! Glowing mouse brains! So I guess what I’m saying is that I’m excited to see what scientists figure out next.

Deboki: I can tell.

Sam: Yeah it’s pretty obvious.

Thanks for tuning in to this week’s episode of Tiny Matters. I’m your host, Deboki Chakravarti and I’m joined by my co-host Sam Jones who is also our executive producer.

Sam: This week’s script was written by me, edited by Andrew Sobey, and fact-checked by Michelle Boucher. The Tiny Matters theme and episode sound design are by Michael Simonelli, and our artwork was created by Derek Bressler. Thanks to André Fenton and Richard Huganir for chatting with us.

Deboki: If you haven’t rated and reviewed us on Apple Podcasts, Spotify, Stitcher, Audible, or wherever else you listen, please do! And if there are some tiny things that you think matter and that you’d love us to explore in an episode, please shoot us an email at tinymatters@acs.org.

Sam: We’ll see you next time.


Related Content