Science Goes To The Movies: ‘Rememory’
In the recent film “Rememory,” an inventor has built a device that can extract memories from the brain. There’s more to the story—a murder, for one, and an amateur sleuth played by Peter Dinklage—but let’s pause here. How close are we in real life to being able to record our memories, Hollywood-style?
If you ask Boston University neuroscientist Steve Ramirez, in some ways we already do. “If you think about it, we outsource our memories already to things like social media, like Facebook and Instagram, Snapchat, the works,” he says. But as he explains, these objective recordings of our day-to-day lives don’t work the same way as our memories, which can change a bit each time we recall them.
“When we recall a memory, often times it can be susceptible to modification,” he says. “So a blue shirt will turn purple, we add people in and out, the point of view that we’re recalling the memory from changes. And I think that’s a good thing, because some hypotheses out there are that what we lose for having inaccurate memories, we gain for being able to recombine elements of our past. And some would argue that that’s a basis for things like imagination and creativity.”
In 2012, Ramirez and his colleague at MIT, Xu Liu, successfully created a false memory in mice. “We actually were able to go into the mouse brain and find the brain cells that held onto a particular memory and then artificially update that memory, so change the contents of that memory,” Ramirez explains. (Watch their famous TED talk about the research here.)
Since then, Ramirez has continued to study memory—including how in mice, pleasant memories can be “activated” to fight the symptoms of depression. As he described in an interview with National Geographic, memory manipulation could someday be used to treat post-traumatic stress disorder or depression in humans—although he acknowledged the ethical road there could be thorny.
But Ramirez’s findings about mouse memory don’t translate directly to humans—at least not yet. Why? He likens the mouse brain to a 1988 Toyota Camry, and the human brain to a 2020 Lamborghini. “We can learn a lot from mouse brains in the same way that we can learn about the principles of how wheels work or how engines work between these two cars,” he says. “But you know, there’s a lot of work to be done,” he adds. “We don’t quite yet know where a memory is localized in the brain, or how it actually works, but we have some rough ideas.”
Ramirez is confident that in the future, we’ll crack the mysteries of memory. “As a scientist, I believe that everything in the physical universe has an answer, it can be understood and it can be reverse-engineered,” he says. “And the human brain, for instance, happens to exist in a physical universe, so it can be understood and it can be reverse-engineered.”
But for now, just like the rest of us, he outsources his memory to technology when he can. “Right now, my entire life depends on my Google calendar, for instance,” he says. “Maybe that opens up new corners so I can dedicate more time to other things that are in front of me, which I think is a sign of progress—we’re never going to have less to remember, it’s just going to be different things that we have to remember.”
Steve Ramirez is a professor of Neuroscience at Boston University in Boston, Massachusetts.
[AUDIO PLAYBACK] [MUSIC PLAYING]
– The most spectacular science shocker ever filmed. Too real to be science fiction, now science fact.
IRA FLATOW: That sound signals another edition of Science Goes to the Movies. Stories of memory manipulation seem to be everywhere in Hollywood these days. I recently saw a film called Rememory, starring Peter Dinklage. And in the movie, Dinklage’s character visits a biotech company that’s developed a device for extracting memories from the brain in hopes of finding some relief from a traumatic moment in the past. But when he arrives, there has been a murder. He ends up using the memory device to piece together the truth using other people’s memories.
I’m not going to spoil it for you, except to say in the end, this film leaves you thinking about exactly who we are and what are we without our memories? Science fiction, of course. But is there any real world research that gets us even close to realizing these made up scenarios? Is it possible to record memories on a chip, for example? Joining me now to discuss how real science squares with the stuff of Hollywood is my next guest, Dr. Steven Ramirez, professor of neuroscience at Boston University. Steve, welcome back to Science Friday.
STEVEN RAMIREZ: Thank you, it’s good to be back.
IRA FLATOW: Did you get to see the movie at all?
STEVEN RAMIREZ: I actually got to see bits and pieces. I had one job to do, and instead, I actually only saw the trailer.
IRA FLATOW: Ha, well, I won’t give it away. Are we getting any closer to the day where we could possibly record memories that people have had over the years?
STEVEN RAMIREZ: You know, I think that yes. So I say yes because we kind of are already doing that, right? Like if you think about it, we outsource our memories already to things like social media, like Facebook and Instagram, Snapchat, the works. I mean, we upload objective pictures or videos of what’s actually happening in our day-to-day lives.
And that’s a good and a bad thing. It’s a good thing because it’s an objective recording of what’s actually happening, which isn’t how memory works. Memory isn’t, in fact, an objective recording of our experience. And it’s a bad thing because, well, our memories can be warped pretty easily, right? I have this memory of what I look like when I finished graduate school. And in my mind, I look like Jon Snow. But I go back and look at that picture and it’s like, age turned me into a White Walker.
IRA FLATOW: Well, that’s interesting, because I was just talking with Nobelist Susumu Tonegawa I know you know him.
STEVEN RAMIREZ: Yeah, he was actually my graduate professor.
IRA FLATOW: That’s right. And you both studied memory in mice and found that you could create and change memories in the mind.
STEVEN RAMIREZ: Yes, so that was actually back in 2013, which is when I first appeared in this segment. And that’s when in that lab along, with a colleague, Xu Liu, we actually were able to go into mouse brain and find the brain cells that held onto a particular memory, and then artificially update that memory, so change the contents of that memory.
IRA FLATOW: And do you know if the memory was actually erased or did you sort of pave it over a little bit?
STEVEN RAMIREZ: So I think we paved it over. So in terms of memory erasure, I always make this joke that if mice had Hollywood, then things like Eternal Sunshine of the Spotless Mind or Rememory would totally be a reality. And we actually were standing on the shoulders of giants there from labs from Sheena Josselyn, for instance, in Toronto, who was actually the first to be able to successfully erase a memory in rodents. And we wanted to ask the flip side. Could we go in and change the contents of the memory or artificially activate a particular memory?
IRA FLATOW: So we do know where in the brain, then, that the memories are located.
STEVEN RAMIREZ: We have a rough blueprint idea. If the answer to that was yes, then I would kind of be out of a job, because everything would be solved. But the reality is the mouse brain is sort of like a 1988 Toyota Camry, and the human brain is like a soon to exist 2020 Lamborghini. So we can learn a lot from mouse brains in the same way that we can learn about the principles of how wheels work or how engines work and stuff between these two cars. But there’s a lot of work to be done. We don’t quite yet know where a memory is localized in the brain or how it actually works, but we have some rough ideas.
IRA FLATOW: But you’re confident, if I understand it, that you think that someday, we may be able to capture memories and store them.
STEVEN RAMIREZ: Oh yeah.
IRA FLATOW: Offline and up in the cloud or something.
STEVEN RAMIREZ: Yeah, I’m pretty hard-headed about this, because for me, it’s like as a scientist, I believe that everything in the physical universe has an answer. It can be understood and it can be reverse engineered. And the human brain, for instance, happens to exist in a physical universe. So it can be understood and it can be reverse engineered. The thing is that it takes a lot of ingenuity and it takes a lot of time.
The same way that if you were to ask somebody 2,000 years ago, imagine getting to that silver orb up in the sky, you would probably get bludgeoned for witchcraft or something. And now, getting to the moon is something that happened like 60, 70 years ago. So I think that imagine 2,000 years from now in the future, that the kinds of questions that we’re asking right now that are a so-called moonshot, would also exist in the brain, and that we’ll get there at some point.
IRA FLATOW: Part of the plot of Rememory is trying to capture the memories of people because our memories are so bad. We remember things poorly, right? And maybe if we can get to the core of the real memory, we might be able to deal with our memories better. Is it possible though? I mean when we remember something, do we actually remember it the way it happened, or do we change it every time we pull it out?
STEVEN RAMIREZ: So this is something that, this is a hot button today in memory research. So if it’s one thing that we’ve learned from 60, 70 years of memory research is that when you recall a memory, it’s not at all like a carbon copy or like an iPhone video of the past where it’s this bona fide objective copy of what we’ve experienced.
It tends to be a lot more dynamic and a lot more fluid in that when we’re recalling memory, often times it can be susceptible to modification. So a blue shirt will turn purple, we add people in and out, the point of view that we’re recalling the memory from changes. And I think that that’s a good thing, because some hypotheses out there are that what we lose for having inaccurate memories, we gain for being able to recombine elements of our past. And some would argue that that’s a basis for things like imagination and creativity.
IRA FLATOW: Mm hmm. So then we have to just deal with the fact that our memories are imperfect?
STEVEN RAMIREZ: Exactly. And I think if you want to have a real existential crisis at night, you think about every time we recall our memories and we slightly change it and memories are what thread our identities over time, what does that say about our identities? It means that our identities also can be updated over time. And I think that that makes sense, and I think that that’s a good thing for us as a species.
IRA FLATOW: Do we know what we decide to make a memory of? How we decide something should be a memory?
STEVEN RAMIREZ: Yeah, I mean I guess I honestly don’t know the answer to that, other than I think that we’re almost constantly recording what we’re experiencing. And where we have this underappreciated bandwidth of what the brain is capable of remembering. So if you think about it, I’m sure we all have instances where we’re out with some friends and maybe grabbing a round.
And then for some reason, I don’t know, somebody says something that reminds you of a memory that you had from like 20 years ago. And that’s an amazing feat of mental time travel. Where was that memory for 20 years? It was dormant. And then you got the right cues in the world and then you remember it. So there’s these worlds of recollection that exist in our brains that sometimes go dormant for decades. So I don’t know what the storage capacity of the brain is, but I think it’s an underestimated number.
IRA FLATOW: Talking with Steven Ramirez, professor of neuroscience at Boston University on Science Friday from PRI, Public Radio International. Steve, if you were to write your own movie or maybe you’ve seen a movie, is there something that comes closest to what you would like to see, something that you would imagine?
STEVEN RAMIREZ: Yeah, so there’s two things. If I were to write a movie about false memories, it would be about myself and implanting a false memory of the Patriots having a perfect season in 2007. So that’s the Oscar-nominated film about false memories. But in terms of actually being able to accurately depict how memories work in the brain, I mean, sometimes Hollywood actually does its homework.
Movies like Inside Out actually consulted neuroscientists to try to get an accurate depiction of memory. And then if you’re a mind-bending filmmaker like Christopher Nolan, then you know you point to a movie like Memento, which actually has you experience what things like amnesia actually look like. So I think I would take a page out of those two books to write a script.
IRA FLATOW: Do you feel like artificial intelligence is going to make us not need to remember things?
STEVEN RAMIREZ: I mean, I guess this is one of those yes and no things too, right? Like right now, my entire life depends on my Google Calendar, for instance. So I try to outsource my memory to that as often as I possibly can. But who knows, right? Maybe that opens up new corners so I can dedicate more time to other things that are in front of me, which I think is a sign of progress. We’re never going to have less to remember, it’s just going to be different things that we have to remember. So maybe artificial intelligence will just help us remember less in one aspect of human knowledge, but that will give us more time to remember other things, again, as we progress forward.
IRA FLATOW: So I’m going to give you the $64– well, it’s the blank check question that I gave a lot of my guests if you’ve been listening. If you had an unlimited amount of money and you wanted to create a resource, a technology, a device, what would you like to have? What kind of thing would help you?
STEVEN RAMIREZ: So I’ve thought about this a lot. And I think that right now, for instance, we have tools like functional magnetic resonance imaging, where we can go in and we can see patches of brain as they turn on and off throughout the brain when you’re recalling a particular memory, for instance. But in those patches of brain can be anywhere between 10 to 100,000 brain cells.
And what I would love to see is this technology where if any listeners seen the movie, I think it’s Iron Man 3, where the bad guy in the movie puts a little chip behind his ear. And he’s in front of Gwyneth Paltrow and there’s a projection of his brain that happens in real time. And he’s trying to woo her, so he’s trying to say, well, look at these love centers that are lighting up in the brain. You know that I can’t fake my feelings for you.
If we had a technology where instead of looking at just patches of brain, we could see the activity of single brain cells in an alive human being, in this case in the entire brain, I think that would be amazing. I mean, we’d be able to see what happens both in healthy states, like recalling a positive memory, for instance, or in more pathological cases, like what happens in things like depression or things like Alzheimer’s.
IRA FLATOW: Hmm. Well, I wish I had that check in my back pocket for you.
STEVEN RAMIREZ: I do too.
IRA FLATOW: I don’t have it for you, but I think you’re going to get there someday. You know, it doesn’t sound impossible.
STEVEN RAMIREZ: I think so. I mean, if anything, this is something that my dad always tells me, that it’s not a matter of if we’re going to get there, it’s a matter of when we’re going to get there and how we’re going to get there. And I think that that proactive approach to science is like, yeah, we’ve got to get our hands dirty, right? We’ve got to continue to have our bulletproof vest on and just plow forward fearlessly with the kinds of experiments that we’re doing.
IRA FLATOW: Well, we got there. I mean the end of the program, that’s where we’re at. So I want to thank you, Steven Ramirez, professor of neuroscience at Boston University. Thanks for enlightening us today about the movies.
STEVEN RAMIREZ: Oh it was a two-way street, my pleasure. Thank you so much.