We often think of memories like the contents of a museum: static exhibits that we view to understand the present and prepare for the future.
The latest research, however, suggests they are more like well-thumbed library books that wear and change a little bit every time they’re pulled off the shelf.
Think of one of your happiest memories. For real. Sit with the recollection. Let your mind’s eye wander around the scene. See if you can feel a spark of the joy or hope you felt at the time. Let a minute pass. Maybe two.
If you played along with this experiment, you are physically different now than you were a few minutes ago.
When you began to reminisce, brain cells dormant just seconds before began firing chemicals at one another. That action triggered regions of your brain involved in processing emotions, which is why you may have re-experienced some feelings you did at the time of the event.
Chemical and electrical signals shot out to the rest of your body. If you were stressed before you began this exercise, your heart rate probably slowed and stabilized as levels of cortisol and other stress hormones decreased in your blood. If you were already calm, your heart rate may have quickened with excitement.
In either case, regions of the brain that light up when you get a reward jittered with dopamine.
The memory changed you. But by pulling this memory to mind, neuroscientist Steve Ramirez says, you also changed the memory.
Some elements of the memory heightened in importance. Others receded. Your brain snipped out and inserted details without your conscious knowledge. The mood you were in at the time of reminiscence left emotional fingerprints on the memory, as neurons activated by your mental environment synced up with those activated by the recollection.
Every time you revisit this heartwarming scene you change it a little bit, both as a subjective experience and a physical network of cells.
Humans have engaged in this two-way operation of memory revision for as long as we’ve been conscious. But over the last two decades, neuroscientists have found mind-bending ways to control this process (in mice, at least): implanting false memories, deleting real ones, resurrecting memories thought lost to brain damage, detaching the memory of an emotional reaction to one event and attaching it to the memory of another.
“It is all part of a larger revolution brewing in science to make memory manipulation a commonplace practice in the lab,” Ramirez writes in his recent book, “How to Change a Memory: One Neuroscientist’s Quest to Alter the Past” (Princeton University Press). “A memory may transform me entirely, but I have the power to transform it as well — both with my mind and with my science.”
In movies about stuff like this, there’s often a sinister air around the memory-tweaking scientist character. Ramirez, a Boston University professor, is friendly, earnest and keeps a giant inflatable T-rex named Henry in his office.
He sees this research not as the next frontier of coercive mind control but as another way to alleviate mental suffering, alongside medications and cognitive therapies.
“It’s amazing that we can do these things in contemporary neuroscience,” Ramirez said recently from his lab in Boston. “But the real-life, overarching goal of all of this is to restore health and well-being to an organism. … Memory manipulation is another antidote [that] can be part of our toolkit in the clinic.”
Memory is the reason Ramirez exists at all.
His father was once kidnapped at gunpoint by soldiers in his native El Salvador and falsely accused of being a left-wing guerrilla. (Their “evidence”: He had a beard.) He was spared execution when one of his captors took a second look at his face and recognized him as the generous schoolmate who used to share his lunch.
Both of Ramirez’s parents emigrated to the U.S. before his birth, and raised him and his older siblings in Boston. Ramirez got a bachelor’s in neuroscience from Boston University in 2010 and his doctorate from MIT in 2017. As a graduate student he joined the lab of Nobel laureate Susumu Tonegawa, where he was paired with a postdoc fellow named Xu Liu.
Both Ramirez and Liu were drawn to the study of memory as a possible therapeutic tool, and instantly hit it off as friends and lab partners.
Their first major breakthrough together came in 2012.
Three years earlier, a University of Toronto team identified the neurons that lit up when a mouse was exposed to a scary stimulus — in this case, a sound that earlier accompanied a shock. The Toronto researchers then injected the mice with a toxin that killed only those brain cells that lit up when they heard the sound.
The result: The treated mice no longer demonstrated a fear response when the sound was played. Essentially, the scientists had erased a specific memory.
If a memory could be deleted in the lab, Ramirez and Liu reasoned, one could be implanted.
For their experiment, the pair identified brain cells in a mouse hippocampus that activated when the animal received a startling shock. Then they took the mouse out of the enclosure where the shock occurred and placed it in a new box with no sights or other sensory cues associated with the memory of its old environment. Next, using millisecond-long pulses of light, they activated those same brain cells — without the physical shock of the earlier stimulus.
The mouse acted exactly as it had when the shock happened, even though no shock occurred.
You can’t interview a mouse about its memories. Researchers base their conclusions on the animal’s behavior. And in this case, it appeared that they’d turned a memory on.
“It just blew everyone away,” said Sheena Josselyn, a University of Toronto neuroscientist who led the 2009 work on erasing fear memories. “When you can do those sorts of things to memories, you know you have found the neural basis of a memory.”
In 2013, Ramirez and Liu set a mouse loose in a box — let’s call it, as Ramirez does in his book, Box A — and took note of the brain cells that activated as it explored the environment.
They then scooped it up and placed it in a second box, Box B. With minuscule pulses of light, they reactivated the cells that lit up in Box A, triggering a memory of that earlier environment as it explored the new one. At the same time, they gave the mouse a shock.
When they put the mouse back in Box A, a place where it had never been harmed, it froze in fear.
The mouse’s negative memory of being shocked in Box B had, essentially, been remapped to what was previously a neutral memory of Box A. The scientists had created a false memory, another seminal feat.
For their final project together, they put a mouse in an enclosure with other mice and took note of the neurons that fired as it responded positively to the social interaction.
Then they moved that mouse to a smaller cage than usual, where it was alone.
At first, this rodent equivalent of downsizing dimmed the mouse’s mood.
Given the choice between plain and sugary water, healthy mice prefer the latter. But when stressed or depressed, mice show no preference. That’s how Ramirez and Liu’s lonely mouse acted initially.
But when the scientists activated neurons associated with the memory of hanging out with other mice, the mouse’s behavior suddenly changed. It enthusiastically slurped the sweet water. Remembering better times had changed its behavior to resemble that of a healthy mouse.
The paper was published in 2015 in the prestigious journal Nature. But unlike their past shared achievements, this one couldn’t be celebrated together. As it was going through the review process, Liu died suddenly at the age of 37.
Grief, Ramirez writes, is not so different from memory: “Both endure across the entire lifespan, forever changing us, helping us to decide what matters most.”
Ramirez, now 37, opened his own lab at Boston University in 2017. In the years since, memory researchers have made impressive strides: restoring memories lost to amnesia, activating a memory while suppressing the emotions attached to it, detaching the emotional reaction to one memory and attaching it to another. The tools now exist to erase whole events and corresponding emotions from mouse brains, or to artificially jump-start memories and all the feelings that go with them.
But there is no expectation in the research community that laser-wielding doctors will one day artificially reshape human patients’ memories.
For one, these experiments are possible only with mice that have been genetically modified to have brain cells that light up when exposed to lasers. Genetically altering a human in this manner, researchers interviewed for this story said, is neither ethical nor practical.
It’s also not necessary.
“We don’t need to generate technophobic fears of a digital future where our memories will be distorted — our memories can already be distorted very effectively by nondigital means,” memory scientists Ciara Greene and Gillian Murphy wrote in “Memory Lane: The Perfectly Imperfect Ways We Remember,” published earlier this year.
Humans are suggestible creatures with extremely pliable memories. Armed with little more than a few leading questions, researchers have found that most humans can be easily manipulated into believing that they did or saw something they didn’t. We don’t need lasers to activate our memories, which can be summoned at will or triggered by any number of sensory cues, or to edit their contents, which our brains do constantly without any conscious input from us.
The real goal of research like his, Ramirez said, is to establish the biological mechanisms of memory and apply that knowledge to noninvasive therapies.
If researchers understand exactly how to retrieve a memory from a mouse hippocampus that brain damage has rendered inaccessible, for example, that information could be the basis for a drug that helps preserve or strengthen certain types of memory in people suffering from dementia or other cognitive disorders.
Understanding how an animal brain encodes memories and the emotional responses they evoke could lead to better cognitive therapies for post-traumatic stress disorder.
The obvious dark side of this line of research is that someone who understands how to boost well-being through memory manipulation could just as easily use the same knowledge for pernicious ends.
“The idea of artificially changing our own memories might elicit uneasy feelings of a dystopic future where relationships are erased, identities are replaced, and governmental powers implant thoughts in our heads to mind-control society,” Ramirez writes in his book. But, he said, any tool in existence can be used to harm or help, and he’d rather make well-intended progress than none at all.
“The idea of memory manipulation, to me, makes sense if we have an ethically bounded goal, and that ethically bounded goal is to restore health and nourish human well being,” he said. “Exercise is an antidote for the brain, and social enrichment is an antidote [and] a good night’s sleep is an antidote. What if toggling with memories in a therapeutic manner can also be an antidote? Then we’re in business.”