← New Books Network
New Books Network
Steve Ramirez, "How to Change a Memory: One Neuroscientist’s Quest to Alter the Past" (Princeton UP, 2025)
00:50:49
Loading summary
Transcript31 lines
- [00:01]
A
Welcome to the New Books Network.
- [00:05]
B
I'm Caleb Zakrin, CEO and publisher of the New Books Network. Today I'm speaking with Steve Ramirez about his new book, how to Change One Neuroscientist's Quest to Alter the Past. An associate professor in Psychological and Brain sciences at Boston University, Steve has published seminal research on the neuroscience of memory. While many aspects of memory remain a mystery, neuroscience research has discovered where memories reside, how they change, what activates them, how false memories are produced, and much more. Memory is such a profound topic that it's impossible to imagine our lives without it. As our understanding of its processes advance, we're learning how memories can be manipulated, artificially produced and destroyed to take us on a memorable journey. I'm pleased today to get the chance to speak with Steve Ramirez. Steve, welcome to the podcast.
- [00:52]
A
Awesome. Thank you so much for having me today.
- [00:55]
B
This is really a fun book to read in part because, you know, you really just incorporate so many great personal stories and I think oftentimes it's hard to translate, you know, hard nosed scientific research into a book that's meant for non experts like myself. But I think it's really important work. So, you know, I commend you for doing it and I think you've done a really excellent job of explaining memory and exploring it in a way that's engaging, fun. But at the same time too, you know, I did feel like I came away with a much better understanding of how memory actually works in the brain. And I was wondering if you could just start by telling us a little bit how you got interested in the scientific study of memory.
- [01:34]
A
Definitely. Thank you for the kind words because they, they mean the world. Especially for my first book where I didn't really know what to expect really. So the way that I got interested in memory was sort of through a series of accidents throughout college. And um, when I was an undergraduate at Boston University, I actually enrolled undeclared because I had no idea what I wanted to do with my life. Like, there's one parallel universe where I wanted to go and study at NASA. There's another where I wanted to become a chef. There's another where I wanted to play music my whole life, which was a long winded way of saying that I didn't really have it figured out yet. And one of the careers that I thought was interesting was that of a scientist, but I had no idea what that entailed at all. So I talked to my advisor and he recommended that I join a lab that studied Alzheimer's in particular and that first lab, unfortunately, I spent, like, four months just making the same chemical again and again and again, and it felt very rinse, wash, and repeat. And it didn't really feel like the kind of exciting world that I thought science was going to be. So I was a bit discouraged, admittedly. And the same advisor encouraged me to not give up and to just give it one more stab at it in another lab. And that other lab was Howard Eichenbaum's lab at Boston University, which studied the neuroscience of learning and memory. And I joined Howard's lab, and I'm convinced that if that lab studied algae or how paint dries, that's what I would be studying today, because everyone there was so supportive of just me as a burgeoning scientist, trying to figure out what the process was all about. So when I joined that lab, I was paired up with a postdoctoral fellow at the time, a researcher in the lab who was training me in doing experiments with rats and studying how rats process time, our perception of time in the brain. And that. That, to me, over time, really had me hooked because I worked with a group of people who were very like, the experiment worked, let's go out and celebrate. Or the experiment didn't work, let's sit down and go to the drawing board together. But it was that kind of camaraderie, plus the neuroscience of memory, that initially had me, I'd say, relatively hooked where I liked it, but I knew that it was something that I could maybe give it a shot and learn to love over time. And then the rest basically is leading up to where I am now. I never really looked back.
- [04:09]
B
One of the concepts that you explore in the book is that of the engram, which is a sort of a physical trace of a memory. This idea that, you know, your memory as a kid of fly fishing with a parent, or of going to a baseball game or of stubbing your toe really badly one time that. That there's literally like an etching somewhere in the brain with that memory. Can you talk about this idea of an engram?
- [04:34]
A
Yeah. So the engram, in a sense, is a mini holy grail of memory research, because it's the physical manifestation of memory in the brain. And if we understand what an engram is and how it works, then we understand what memory is and how memory works. So the way that I think about it is that all experience, whether you're awake or sleeping or daydreaming, mind wandering, idling, doing some activity, all experience leaves some kind of measurable change in the brain. And that Measurable change is sometimes transformed into the thing that becomes a memory of that experience. So what that measurable change is and what the physical correlates of that memory is in the brain, that's what we think of as the engram. Now, it's a little bit complicated in the sense that our brain is really dynamic and it's constantly on and it's constantly shuffling and reshuffling information, and basically nothing is static in the brain. So to try to identify an engram, let alone all of the physical properties and features of an engram, it's a little bit like trying to hit this receding, ever moving target where we can hit it and then we can get a snapshot of it for a brief moment. After that brief moment, everything changes once again. And then, and then the brain kind of that dynamism kicks in and the memory or the experiences that we have continue to transform and warp in the brain. So we understand a little bit of the rules of how that happens, but the actual nitty gritty of it is exactly what engram research is trying to figure out today.
- [06:15]
B
Right. So could, could you talk about, you know, just as an example of, of where memory might reside, the difference between a short term memory, you know, something, my memory of a, you know, the meal I had yesterday versus you, longer term memory, something that happened when I was a little kid. Where would those be located?
- [06:34]
A
Yeah. So I'll start off with a bit of the dogma that was pretty much ubiquitous in the field about how these memories of recent events are changing and transform over time into old events, and then kind of give it a little bit of a facelift with where we are today because the brain threw a couple of curve balls at us. And it turns out our understanding is still pretty holy and complete when it comes to how memories age in the brain, for example. So short term memory we think of as this time limit, time limited snapshot of experience that we have that enters our brain. It lives throughout the brain. And at some point, though, kind of like a book taken out of a library, at some point that book's gotta be put back. So what's doing the putting back of that book? Is that book changing in its contents over time? Are people scribbling in it with new information and so on while it's in the bookshelves. What we used to think was that when you first have an experience and first form a memory, there's areas of the brain like the hippocampus. You have two hippocampi right above your ears, tucked Inwardly, that areas of the brain, like the hippocampus, light up and they help process experience and then file it away to the rest of the brain, which is the cortex, the outer mantle of the brain. So this theory was massively influential and it's called systems consolidation, which is the idea that as memories are stored or consolidated in the brain, they recruit various different systems of the brain to make that memory possible. But in the past, I'd say 15 years, there's been a pretty hefty challenge to that dogma that actually suggests that memories are located everywhere in the brain and it's not like they enter the hippocampus, and the hippocampus hands off a baton to the cortex so that that experience can be shuffled on later on. It's more of the. Any memory recruits the brain, high and low, left and right, ancient and modern, and it recruits a constellation of activity throughout the brain. And that constellation of three dimensional activity throughout the brain and is refined over time. It's kind of, it's chipped away at different clusters of cells start firing in different patterns, maybe even new cells begin to take over over time. And that makes sense because our experience of a memory changes, whether it's the memory of what we had yesterday or the memory of what we did 20 years ago, they kind of feel different. That I could probably recall yesterday's dinner in pretty vivid detail. I remember where I was, when it was, what I was doing and so on. To recall that Same dinner from 20 years ago is probably a little bit more challenging, presumably because some aspects of memories lose detail and become a bit fuzzier over time as that constellation of brain cells that makes that memory possible begins to become also warped in its activity and changing in its activity over time. So this is now just my speculation, but I think memory is just what the brain does and that it doesn't make sense to me to think of memory as it recruits one brain area on one day, in another brain area on another day. It's recruiting everything all at once. And that three dimensional constellation of activity is shifting and changing and morphing itself over time. And I think that's a little bit closer to at least where, how we think memory works nowadays.
- [10:06]
B
So as far as what is changing those memories, could you, could you talk about what is causing a memory to change? Is, is it every time we're recalling it, the memory is a little bit different. So, you know, if, if there's a particular memory that I've recalled, let's say, a hundred times in my life, it's going to be different than a memory that I've only recalled twice or three times. I don't know if that, that, that framing necessarily makes perfect sense. But, you know, how does the recollection of a memory change the actual, like physical instantiation of that memory in the brain?
- [10:41]
A
Yeah, I'm glad you asked because I, I realized way after the fact that if I were to answer now just the title of the book, how to Change a Memory, it would have just been a single page and the answer would have been to recall it exactly like you intuited. So one thing that we've learned from maybe the last half a century of psychology and neuroscience is that memories aren't like a video of the past that you can record and then rewind and revisit again and again. I think of it more as a kind of reconstruction of the past. Some even think that memory is our brain's best prediction of what we think actually happened in the past. It's almost like your brain's forecast of what it thinks happened. So the way that I view this is that we know that memories are imbued with sights and sounds and smells and emotions and context and temporal information and so on and so forth. And that's going to recruit the different corners of the brain that are involved in processing the sights and sounds and smells and emotions and contexts and timing information and so on. So that's another way of saying that memory is what the brain does in terms of recruiting all of its expanse to be able to store a given experience so that we can later recall that experience. Every time we recall that experience, I think of it as if we were like on social media and we pull up a picture of ourselves and we want to apply different filters and contours and hues and modify the picture. That's kind of what we're doing every time we recall a memory where different aspects of the memory are shape shifting and changing and the information embedded in that memory is changing. Usually this is pretty inconsequential because if I remember the memory of my wedding day, for example, it doesn't really matter too much if I misremember what my mom in law was wearing, for example, or what my father in law said for a speech or anything along those lines. Usually it's like we storytell, we misremember some details collectively, we fill in those details and we're off to storytelling and we're good. It's a bit different, of course, when what the person said or what they're wearing Is the difference between someone being put away in jail or not and used as eyewitness testimony. So I think of memories malleability as this inherent function that memory. All memories have that whenever we recall them, we're constantly warping them because the brain is really good at creatively shuffling information around, presumably to try to create something new with that information, such as imagining what we're going to do tomorrow or imagining what we're going to do a year from now, and so on. But yeah, every time we recall a memory, it's susceptible to modification. And that modification can happen because of what we're experiencing in that moment while we're recalling that memory, what our mood is, what's happening in the outside world that might bias what we recall and how much we recall it and so on. Yeah, so I sometimes think it's a miracle that we all have some collective agreement on things like what we did for dinner yesterday and so on. But. But I think that this is an inherent function to the biological properties of memory is its inevitable changeability every time it's pulled out of the library shelves and recalled.
- [14:06]
B
The way that memory works is it's not necessarily like there's some directory in our brain where we can just go and look up past experiences and. And then recount them. You know, we need triggers. We need things like, you know, we need a scent or something like that. Someone. Someone has to be. Has to remind us of something. We have to be in a particular location for a particular memory to. To rush in. How does that operate? Why. Why is it that you need certain triggers to activate the memory?
- [14:34]
A
Yeah, I think of this like a domino set where you have a bunch of dominoes lined up, and that's the domino of experience. And any given domino can be knocked over to cause the chain reaction of recollection. One of those dominoes might be the odor, like the smell of a cupcake that might remind you of the memory of a bake sale, or it might be the smell of some musky cologne that reminds us of a parent, or so on and so forth. So we think that when you first have an experience, there's sights and sounds and smells embedded into that experience that recruits the constellation of brain cells that make that experience possible, literally memorable. All of those brain cells process some aspect of that experience and some aspect of what will become that memory. So we think that when you get some kind of trigger in the world, whether it's the smell of a cupcake, the odor of a cologne, a plane flying overhead, anything along those lines, that that information is entering the brain and hitting the corresponding cells that originally processed that aspect of the experience. And once that. Once those brain cells are activated, that's the domino that got flicked over that then causes the recollection of that given memory. And it gets pretty weird pretty quickly, because it can be because of external stimuli like a smell or a sound. Music is particularly powerful with recalling particular memories, but it can also be internally generated. And by that, I mean our thoughts can be triggers that are the gateways for worlds of memories and experiences to come flooding back while we're just kind of chilling in our chair. And. And I think that's really kind of the miraculous and magical properties of memory, that outside triggers can bring back worlds to life, and that our internal triggers, namely our thoughts and moods and perceptions and biases and so on, can also reawaken corresponding worlds of memories. And how exactly that happens, I think, is there's a lot of good research suggesting that there are some cells that get turned on that then turn on other cells in particular patterns, and that's the physical correlate of memory. But I think that we certainly have our work cut out for us in terms of trying to figure out what are the real rules of the game here, like, will the same trigger always reawaken the same memory? Or how come the same memory can be reawakened by many different triggers, or many different triggers can reawaken many different memories, or the same trigger might reawaken one memory today, another memory tomorrow, and so on. And I think that that's really a testament to the vast history of memories that our brains really store. Right? Yeah.
- [17:16]
B
It gets complicated really quickly, which I think is part of the reason why neuroscientists still have a lot of their work cut out for them, is that it was hard for me even reading the book, to keep track of all of the different ways that memory seems to operate. And it gets even more complicated when we bring in these topics of false memories or this idea that you can literally. Some of the research that you were doing, literally creating a memory to flash in a rat out of by. By shooting some light at it, essentially. Can you talk a little about some of your work and research around this idea of implanting false memories?
- [17:59]
A
Definitely. And, you know, I'll say at the outset that the. The whole reason why we wanted to study artificial memory activation and creating false memories and so on is because we really wanted to understand just how memory works and how far can we push our understanding of memory. Because I think of it like A car that the more we know about how a car works, the better equipped we are to predict and intervene with it when something breaks down. So it's the same thing with memory, that the more rules of the game and the more principles of memory that we can discover, the better equipped we are to figure out, well, what's happening when this memory here seems to be going away or is completely lost, or what happens when a memory seems to be subject to degeneration, like in Alzheimer's, for example, and is that memory ever really gone or not? So our argument was that, well, we want to try to control memory and to try to alter the contents of memory to see if our speculations and guesses of how memory works are actually correct. Because if so, then we have, like, a more or less a blueprint of how we think memories may work, which can be a powerful blueprint for predicting what happens when memory breaks down. So for those experiments, we were building on a previous set of findings. That was really the first finding that launched my career, was the idea that we could go in and literally see the physical basis of a memory by finding the brain cells that held on to just one memory. And we could trick just those brain cells to glow green, for example, so that we could put the brain under a microscope. Look at the brain, we see a bunch of green dots, and those are the cellular correlates of a given memory. And then the first big discovery here was that when we artificially activated just those brain cells, that was enough to bring that memory back to life. That artificially activating a small amount of cells that were involved in just one memory was enough to act like that domino that we flicked over that caused the chain reaction of recollection. The next step was, can we alter the contents of memories? Because in humans, as we mentioned, this happens all the time, that every time we recall a memory, it's susceptible to modification, and hues and colors and contours are changing, information is warping and so on. But could we do that artificially in the rodent brain to understand how that process works? And for that experiment, my late colleague and friend Shu Lu, that I talk about at length in the book, that experiment we cooked up in New York in a winter in 2011, where we were in Times Square out late at night, celebrating our first big discovery, and we had sat down to write down on a napkin. What's the experimental design that we need? Like, what's the experiment that would truly help us test whether or not we could create a false memory in the brain? And the experiment was relatively straightforward from neuroscience standards here, where what we had to do was we had to find the memory of one environment that the animal explored that it got to know and sniffed around and formed a memory of just a neutral boring box. And we could find the brain cells that held onto that neutral box and actually trick those cells to respond to pulses of light, meaning we could shoot lasers into the brain and activate the memory of that safe, boring context. The experiment was we would reactivate the memory of that safe, boring context, but while the animal was experiencing something mildly negative in real life, because that was the attempt to link the real life negative event that was happening with the artificially reactivated safe, boring memory. And if that worked, when the animals went back to that safe, boring environment, they should show some fearful behaviors to that environment, even though technically nothing negative happened there. And flash forward to Christmas Eve of 2012. I had an hour in lab to run a handful of mice to test whether or not that experiment had worked. And. And every single animal showed evidence of recalling a fear memory in an environment where, again, technically speaking, nothing negative had happened. So the experiment worked, and we were beside ourselves in excitement because it showed that we could not just artificially activate a memory, but we could activate a memory and rewrite the contents of that memory by giving the animal whatever kind of experience we wanted to rewrite that memory with. Because it works the other way, where we could reactivate a safe, boring memory and turn it into a positive experience by giving the animals a physical positive experience while artificially reactivating the safe, boring memory and so on. And that's really where the field of changing the contents of memory, I think, really began to take off. Because now we could do it with something as simple as shooting light into the brain.
- [23:11]
B
Right. And obviously there are, there are reasons to be concerned about how this could be used. I mean, with this example, you were making them afraid. So one could imagine a way that someone might use it as a form of manipulation to make people artificially fearful or artificially scared in certain contexts. Cinema too. You know, there's also the flip side of it, of making someone or helping people, especially those who suffer from things like ptsd. Could you talk about some of the research that is done, some of the clinical applications that people are looking into for this false memory creation?
- [23:46]
A
Definitely the thing that stems from creating false memories is not so much the false aspect of it, but the fact that we could rewrite the contents of a memory in a positive direction or in a negative direction, or really in any which direction to update the contents of that memory. And we think that that has tremendous therapeutic value. Now, to take a step back, you know, I, I give this example, or I think of this example. A lot of, you know, something as elemental and nourishing as water can be used to nourish ourselves and to give us livelihood, or it could be used to torture someone. So something that simple can be used for good or bad. Imagine memory manipulation, which inherently has this negative sci fi. This is going to be used for bad connotation. So what you and I thought pretty early on was that the way that we could move this forward in a way that's beneficial for people is to have an ethically bounded goal in mind for doing this exact kind of research. And that ethically bounded goal was that we want to be able to understand memory and to control memory so that we can restore health and well being back to an individual who may be suffering or afflicted by a particular disorder. For example. Because now memory manipulation is not the thing that happens in Hollywood to brainwash a population of people. It's an antidote. It's part of the Swiss army knife toolkit that we have to go in and try to restore health to the brain or to chisel out something that might be pathological in the brain and chisel in something that might be able to bring it back to health. So, for example, after our false memory work, the next project that we embarked on focused on this exact issue, which was to what end? Like why, why manipulate memories to begin with? Sure, we do this naturally when we recall memories, but having an external party do it seems a bit sinister. So we figured, let's use an example or create an example where this could be done with that ethically motivated goal in mind of restoring health and wellbeing back to an organism. So that project focused on artificially activating positive memories in the brain and in particular in animal models of depression and anxiety. Because oftentimes in depression and in generalized anxiety, positive experiences are the very thing that's affected. It's the very thing that begins to either deteriorate or to become inaccessible to the individual. So what we did was we have, there's several animal models of, of virtually any disorder here. And we chose an animal model of depression and anxiety. And what we did was we went and we gave that animal a positive experience, which can be anything from social rewards, like animals socializing with each other, since mice are very social, or giving them a food reward. Sugar water, condensed milk, and Nutella. Like they go bananas for all of this, and they, they really seek it out. So what we did was we gave them this positive experience. We found the cells in the hippocampus that held on to that particular positive memory. And then when the animals were showing a depression or anxiety related symptom, that's when we artificially activated a positive memory. So for example, an animal that shows the symptom that is linked to anhedonia, meaning the inability to experience pleasure. That symptom manifests in animals when they're, for example, when they're given the option to choose between regular water or sugar water. And usually like, we would choose water that's flavored and sugary over regular water as well, as would mice. An animal that shows the symptom of anhedonia does not choose the sugar water over the regular water, it chooses it 50, 50. So in other words, it has no preference for either or. And we take that as this indirect proxy that the animal is not seeking out the pleasurable thing that it normally would find pleasurable. Now, amazingly, when we artificially reactivate a positive memory in those animals, they immediately start seeking out the sugar water. So that symptom of anhedonia vanishes the second that we artificially activate a positive memory and it's eliminated. And we actually came up with some protocols to eliminate that symptom on the order of weeks to months for the animal. So it was a lasting intervention in humans. Right. We don't have to do any genetic engineering or invasive shooting light into the brain, because I can just ask you, like any positive memories come to mind that make you feel good or that you particularly savor or enjoy recalling. And there's a whole litany of research now suggesting that when we recall positive experiences and really sit with them, it activates our brain's reward circuitry, it increases our motivation, it increases our sociability. Some studies even suggest that it increases our cognitive flexibility and creativity, makes us more likely to get up out of bed and interact with the world and so on and so forth. So the things that positive memories in particular can do are pretty remarkable because they alter our entire biology, both from our brain to our physical and mental well being. And then we could successfully study that in animals by turning those kinds of memories on to see what kind of symptoms that it could alleviate. And that's what we hope is the blueprint that can be used now to say, okay, positive memories seem to be pretty powerful. What kinds of therapeutic interventions can we create now, or at least understand on a deeper, more mechanistic level in humans, given what we can do in rodents. So that's, that's the hope at least, is that this provides an example of manipulating memories to restore health and to alleviate symptoms of. From the pathologies that increase so much suffering in the world.
- [30:00]
B
Right, right. And I think what that example shows is the way that memories and the triggering of certain memories impacts behavior and impacts how clearly how the mice think about their, what they're going to do. So how do you think about how memories impact, how we think about future impact, how we actually go about our lives? Like, you know, almost, almost want you to get like philosophical in a way. Like, to what extent are our memories us and who we are and what we do?
- [30:34]
A
Yeah. So I think of it this way, that memories, I think biologically, literally are the thing that threads and unifies our overall sense of being, our, our identity of ourselves. Like the story that we tell ourselves about ourselves over time is built memory after memory after memory. So we see how shockingly real that is when memories are now absent. So this could be due to a traumatic brain injury or due to a particular disease that when memories deteriorate, we really tragically see just how much our sense of not just well being, but our sense of self and identity and ability to interact with all that we know and love around us also begins to deteriorate. And that's not an accident. I think it's because we use memories to make decisions in our day to day lives. We use memories to either savor the past or to bemoan the past, or to learn from the past. But we use them with some utility that seems to be biologically adaptive. And now one of my favorite theories out there, and this is where it gets both scientific and a little philosophical, is that areas like the hippocampus are active when we form memories. And areas like the hippocampus are also active when we make false memories. And it turns out that areas like the hippocampus are also active when we're imagining ourselves in the future and projecting ourselves in some future event. And the theories go that we use memories as building blocks to imagine what an uncertain tomorrow might look like. So we combine and recombine different aspects of our past and sometimes in like completely nonsensical ways, like we probably experience this firsthand when we're dreaming that it seems like it doesn't make sense. But dreamscapes, or when we're sitting in our chair and imagining the future, that imaginationscape is built from the building blocks of our previous experiences and remixed in some fantastical New way that they've never been remixed before. Because the. Presumably the more we can predict, or the more accurately we can predict what may happen tomorrow, the more ready we are for tomorrow and the more adaptive that might be because we're prepared for that uncertain future. So if we, if we take that to its logical extreme, it means that memories aren't just looking backwards. Memories are our brain's way of looking forward in time as well and projecting ourselves into the future. So in that sense, memories can go in any which direction that we want, whether it's going back to the past, to revisit it and learn something about an experience that we may have had, or imagine what tomorrow might be like and play out in kind of a simulation zone in our brain, play out all of the different events that might happen tomorrow. What if the restaurant's closed? What if they don't have the thing that I want to order? What if my friend cancels? What if it rains? And so on and so forth. And we think that all of that, like the amazing cognitive human faculty of imagination is built on the memories that we've had of the past. And likewise, because we're the ever imaginative species that we are, we can also remix and recombine elements of our past by imagining them differently as well too. So memory, in a sense, is directionless when it comes to time. It's just this like omnipresent thing that exists that can be used to recall the past or to time travel forward into an imagined future as well.
- [34:10]
B
Right. No, I want to get a little bit into. This is maybe a little sci fi territory because, and I always to a certain extent hate asking scholars or scientists to predict where, what are we going to learn in the future, because in a way, like, that's not what you're trying to do. You're trying to. I mean, if you make a prediction, you're going to do it in an extremely constrained way. But, you know, where do you see the science of memory developing towards? I think especially there's, you know, a lot of discussion, you know, about implanting chips into people's brains so that they can like literally have a computer in there, you know, that they can recall like, you know, as if it's a memory. They can go in and essentially search, search Wikipedia in their brain. Like, how do you think about the future of this development in terms of the ability for us to actually have almost super memories?
- [35:03]
A
Yeah. So I think we're a ways off from the chip in the brain being our gateway to all of the information of the Internet and so on. It's fun to think about though, because I don't think it breaks any law of physics. It's just an incredibly complicated technology meeting an incredibly complicated organ. But I mean, you bring your cell phone back 2000 years to like the Greco Roman days and it would make zero sense what your phone can do. And yet now it's like, oh yeah, it's just this casual supercomputer that we have in our back pocket kind of thing. So I, I definitely don't like to put any artificial speed limits on science and innovation because we may be surprised on where we, we, we may not be able to even comprehend the technology of 2000 years from now. The way that a phone that can accidentally butt dial halfway across the world that we don't Even know exists 2000 years ago and the way that that's possible. So where I see this going, I think is there's a couple of levels to it where on the basic science, just the understanding of memory realm, I think that we'll continue to really understand what memory is within the brain and its ability to influence our brain and cognition and behavior with remarkable detail. Because for, for example, we know that DNA is the building block of life. I'm convinced that we don't yet know exactly what's the building block of memory. Is it the DNA that's in brain cells? Is it the physical structure of brain cells? Is it like a symphony? And it's more of just the brain cells firing in particular patterns and there's different kinds of brain cells just like there's different instruments. So how are those working together to make the symphony of a conscious experience or of a memory possible? What happens when one of the instruments is not tuned as well as it should be? What does that mean for the memory? I actually think that we're going to be able to understand memory exactly at that level where we can go into the brain and say, here's a map of what the constellation of cells that make this memory possible look like. It seems like maybe there's some cells here that are susceptible to stress or susceptible to some genetic predisposition to degenerate. For example, can we do something about it? Maybe there's some kind of either cognitive behavioral approach or drug based approach to boosting the activity of those susceptible cells to try to make them resilient to degeneration or to try to make them resilient to becoming quieted and maybe having that experience be forgotten forever and so on. I actually think that there's some Sneak previews of this happening in contemporary neuroscience as we speak. So, for example, one of the biggest success stories, in my opinion, in the last 15 years of neuroscience, Is the ability to artificially bring back memories that were once thought to really be lost, like, lost forever. Here we know this because there's many ways of inducing amnesia, and in rodents and in humans as well as. And this could be through a traumatic brain injury. It could be through sleep deprivation. It can be through drug addiction. It can be through a neurodegenerative disorder like Alzheimer's or cognitive aging that happens over time, inevitably, and so on. There's many different ways to make a memory inaccessible and give rise to amnesia. And in every single one of those cases, it's amazing, even in the case of the one amnesia we all share, which is infantile amnesia, Our inability to remember what we did, like when we were three years or less, that in every single one of those cases, we're able to artificially restore the memory that was thought to be lost. So if we have subjects that are sleep deprived and they show amnesia for a memory, we can find the brain cells that hold onto that memory, reactivate them, and bring that memory back. And the same thing with cases of Alzheimer's and drug addiction and traumatic brain injury. And what makes me so optimistic about that Is that it means that memories might actually be a lot harder to break than we think. They might actually be relatively easy to make inaccessible in the brain, because we don't need to be recalling everything all at once. We usually recall one memory at a time here. But all of the instances where we thought those memories were gone and give rise to amnesia, We've been able to artificially restore those memories and bring them back. And in some cases, a handful of clever, very clever studies Were able to bring the memories back and do some kind of cognitive behavioral intervention to the animal to make that memory stick, to make it stay back so that it doesn't just go back and become amnestic. We say again, so that that memory is now accessible naturally in the future, Whenever the animal needs to naturally recall that memory. My hope is that the same thing would begin to happen in the human literature where we would be able to. We have, of course, like, cases upon cases of amnesia. And if we could find out ways of identifying the. Either the physical basis of particular memories or different intervention strategies that can help make those memories that were thought to be lost come back, then we're in business. And that's kind of my hot take in all this, Is that I think that memories are really hard to break and that the brain stores a remarkable amount more than we think it actually stores. And that in many instances of amnesia, it's the librarian that's checked out of the library, but the books are still there. We just need. We need to go in and hire this new artificial librarian that can help us get those books back. And that's one example, at least, that I am very hopeful that we'll get there in terms of relevance to humans in particular.
- [41:02]
B
You've been intensively studying memory for more than a decade. So you've made a lot of memories about memory. And I'm wondering if there's any way that you think about your own memories as you're going about your day. Let's say you have a particularly great experience and you think, okay, I really want to remember this well, or you have a bad experience. You're like, I want to forget this. How do you think about your own memory, having studied it for so long?
- [41:28]
A
That's a great question. And I admittedly think it depends on the kind of memory for the most part. You know, weirdly enough, I've. When writing this book, I basically lived in memory lane to try to come up with all of the different kinds of memories that meant something to me, that I think were relevant to my journey as a scientist. My relationship, friendship with Shu, my time in grad school, and so on and so forth. And I think of those memories very fondly. Bittersweet, if anything. So those memories, for me, what's transformed is the purpose of those memories. I still remember the first day I met you. I still remember us celebrating in New York, coming up with our false memory project. I remember the last day that I saw him when we were saying goodbye at a bar in Cambridge. And. And I think of them really fondly in the sense that they're providential now. Like, I think of them, and they kind of have become this moral compass of decision making in my life, such as, how do I want to mentor my lab? Well, I want to mentor my lab the way that Shu mentored me, which was with kindness and scholarship. Right. It was like a big brother in the lab, for example. I think of the different memories that I've had in the past years, and. And I think I do one of two things. I either say, let it let the memory enter my conscious experience, and then let it also flee from my conscious experience and let me just sit with it and live with it and take all of the experiential glory with it, whether it's Good or bad, just let, like, literally sit with it and in a sense, try to relive that moment if necessary. Or I'll bathe, mentally rewind and try to learn or extract something from it. Like, what does that memory mean to me now? Maybe I think of it and it's one that I'm really excited or proud to share. Or maybe I think of it and it's a little bit more difficult and I get a little bit more blue and somber. But that's okay, because then I say, well, let me sit with that feeling too, because it's just as important as the good ones for me to understand the why, like, why I have it and why I feel that way. So in its own weird way, my own memories have just helped me understand my own self literally that much more. But it's also imbued a deep sense of gratitude because I think I've been a lot better in the past years. And this was my own goal of trying to sit with experience and sit with memories and really savor them for what they're worth instead of either trying to chase them away or trying to, you know, rewrite them. In a sense, it's more of just letting them be.
- [44:05]
B
That makes a lot of sense and is interesting to think about in that way. I'm wondering. So much of neuroscience has influenced the current research going on with artificial intelligence. And there's been a lot of debate too about these LLMs, specifically in regards to how their memory works and why their memory seems to work really well sometimes and not others. And I was wondering if you could comment at all on memory in relation to artificial intelligence and neural networks.
- [44:39]
A
Totally. It's funny because I used to think of memory when I first joined Howard Eichenbaum's lab. Memory to me was like the most boring thing I could imagine because I'm like, memory means memorizing textbooks and regurgitating it back on the exam. Like it's. It's pretty, it's a. It's homework. Like, it's a hassle. And what I learned is that not only was I wildly incomplete in my understanding of memory at that time, it totally dismissed the very real conscious mental time travel back to any moment of the past or future that memory enables. And that, that facet of memory is exactly what I fell in love with. And it's that facet of memory that I think is largely missing from modern day LLMs. Modern day LLMs are basically like, it's the. It's the smartest kid in the room that can memorize everything in the book and regurgitate it back in exquisite detail and maybe even remix it a little bit. But what's missing is understanding. And what's missing there is like the. The understanding of the thing that it is that you're, you know, regurgitating on the exam, for example. So I, you know, it's promising in the sense that with really all of the breakthroughs with. With AI and LLMs nowadays. I mean, I'd like to think that we'll get to a point where understanding, like true understanding can happening can happen. Because right now it seems to be very brute force that we can have these different LLMs give us a summary of an interview or a summary of a project or a summary of a paper, and it does a fantastic job at it. Like, it makes life so much easier. Now, if we were to then begin asking questions that require a deep level of understanding so that we can cook up the next big experiment or a deep understanding so that we can intervene with human behavior in some ways in like, a psychosocial way that really requires a kind of social IQ and this deep understanding of how people work. I don't think we're there yet. I also, I don't think we won't get there. Like, I think we'll get there. It's just that doing it by brute force is not necessarily how the brain does it. Our brain is really, like, amazingly resourceful. Like, if I wanted to recall what I did a week ago, I don't have to retrace everything that I did yesterday and the day before and the day before and the day before. I can kind of just like take a shortcut there somehow in my brain and get to that experience and land on it immediately and then relive it in all its experiential glory. I think that AI will get there at some point. Whether it looks and behaves like a human brain, I think depends on whether or not we design it that way. But it is, it is interesting because I like to think that hopefully it becomes a kind of companion that makes life easier and healthier in the long run. But it's interesting to see where we are now in the. In the early days.
- [47:26]
B
Yeah, obviously our brain has had the benefit of over 2 billion years of evolution to lead to its creation. So, you know, it'll probably take a very long time before we fully understand how it works. You know, I also wonder too, like, you know, neuroscience focuses such a focus on the brain and. And, you know, it seems to me even in this research that, you know, where memory resides. For researchers, it expanded from, you know, just memory occurs in the hippocampus to memory is, is in the brain stem to memory to. It could. Memories are located all throughout aspects of the brain and we don't exactly know why it might be firing one neuron versus another. So you know, I wonder if, you know, we'll, we'll discover that there's, there's you know, memories in our pinky toes and you know, in our hair follicles.
- [48:19]
A
We're already discovering that there's different aspects of memories processed in like the gut microbiome and that you can, there's amazing studies actually coming out showing that you can reactivate brain cells that were active when you were having an immune response. Speaking of pinky toe, like somewhere in the periphery of your body. Let's just say you're a pinky toe and that if you reactivate the brain's memory of that immune response, you can recapitulate some of that immune response without the actual allergen presence present, for example. So I actually think that, you know, we, we kind of like say it tongue in cheek that we might find evidence of a memory or an engram in our pinky or in our like kneecap or something. But I think that as a biological organism, our brain never evolved independently of the rest of the body. They were in concert with each other throughout all of evolutionary time for as long as they've existence. So I would be surprised if we didn't find traces of a memory in our, the peripheral of our body too. And like that's a, that's a whole other now frontier of where neuroscience can go.
- [49:25]
B
It really is such an exciting field. There's so much, so much incredible research that's, that's ongoing. So I'm sure in the future you'll, you'll have to write another volume that, that further updates where we are and in the research of memory. And I'm sure while the, you know, while we know so much already, I'm sure in 10 years time or even less, we will be probably blown away by what the new research reveals. So, you know, I hope you'll be able to write another book for the non experts like myself who are interested in learning how our memory is evolving and adapting.
- [50:02]
A
Oh, absolutely. I already have, I already have pages full of ideas for the sequel. So I'm already, I'm already on it.
- [50:09]
B
Incredible. Well, Steve, it was really such a, so, so wonderful to have you on as a guest. I really enjoyed talking to you about your book. I really do recommend people go and read how to Change a Memory. It's a. It's a really fun read. It's interesting, and I think, you know, really just well suited for. For people who are trying to understand a little bit more about how our brains work. So thank you so much for being a guest on the podcast.
- [50:35]
A
Thank you. This was really such a blast.