Science1 distinct publisher2 min readPublished
Steve Ramirez's 2013 mouse result turned on an injection that missed its target, and the book he has written since names PTSD and dementia as goals, which puts the question of who authorizes an edit ahead of the evidence.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
The four mice that did nothing were a control group, and nobody designed them that way. Ramirez and Liu put their tagging protein where their hypothesis said the fear memory should live, hit that target in four animals, and got no response [3][4]. In the fifth, the protein went slightly wide, and that animal froze [5]. The variable carrying the result was surgical variance. The effect at the moment of discovery was one animal in five, 20 percent [14], and the thing separating the responder from the non-responders was anatomy the team had not chosen.
Ramirez does not oversell the luck. Asked what would have happened without the error, he said the correctly aimed surgeries would all have hit the wrong spot, the result would have read as negative, and they would have gone hunting through other parts of the hippocampus; he puts the delay at "maybe six months or less" [11]. The accident bought months rather than the insight. His more portable claim from the episode is about how labs handle failure: unintended outcomes are common, a negative result arrives with a million available excuses, and few people go back to look [12]. He credits the fifth mouse with teaching him to read head-scratching data more slowly [13].
The thing the mouse work does not tell you is what the mouse was remembering. The measured output was freezing [16]. That establishes a defensive state evoked in a safe cage by switching a tagged cell population back on, which is a real behavioral fact and not a description of a memory's content. Content is the part a patient would be asked to consent to having altered. The interview credits the 2013 paper with propelling the field [6] without laying out what has been replicated since [16], and it is that later record, not the original five animals, on which any clinical claim would have to rest.
Twelve years separate that observation from the book in which Ramirez sketches easing PTSD and guarding against dementia [15][10]. My view, with its condition attached: the consent rules for memory editing are worth drafting now, while the evidence is entirely rodent, because the alternative is that they get written by whichever group first holds a fundable human protocol. The condition is the readout. If the only available measure of a successful edit stays behavioral, informed consent is being asked to cover a change nobody can yet describe to the person agreeing to it.
Ranked by verification strength, evidence, and original report placement.
In 2013, MIT neuroscientists Steve Ramirez and Xu Liu published a paper detailing how they had created false memories in mice.
The team used optogenetics, in which cells' activity is manipulated using light, to plant a fearful memory in the brain and then reactivate that memory with pulses of light.
The initial experiments used a protein to tag brain cells in the hippocampus that were active during formation of a fearful memory, and those fear memory-bearing cells were then reactivated while the mice were perfectly safe.
The first four mice showed no response; the fifth froze in fear.
Analysis of the fifth mouse's brain showed the protein had been placed by accident in a slightly different location of the hippocampus than in the others, inadvertently finding the spot where the memory was stored and showing it could be manipulated and controlled.
Ramirez is now associate professor of psychological and brain sciences at Boston University and works at its Center for Memory and Brain.
Distinct publishers with included, body-backed reporting in this cluster.
Follow any of these and your For You feed starts watching them — no settings page required.
science
Feeds are triaging patients by engagement, and the proposed defence is literacy, not moderation1 distinct publisher
science
A drug-insensitive enzyme in the mouse gut strips 80% of metformin's glucose-lowering effect1 distinct publisher
product
MIT's magnet trick makes correlated microwave signals without the cryostat1 distinct publisher
product
MIT's bacterial transistors do one calculation every eight hours. That is the useful part.1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
A twelve-year-old result, retold by its author
The pivotal fact — that the injection missed and the miss is what found the memory cells — comes to us as Ramirez remembering his own experiment for Live Science, twelve years on and while promoting the book about it. The underlying 2013 paper is referenced but never quoted, its methods are not reproduced, and no one else in our coverage looks at it. The claims are plausible and internally consistent; they are also entirely unchecked here.
Nothing to count yet
We cannot score uptake because the reporting never offers any. PTSD relief and dementia protection appear as things a book conceptualizes; the freezing mouse is the only demonstration. No trial, no protocol in use, no other lab named, no count of anything.
Therapeutic framing outruns five mice
"Holy grail," a quest to alter the past, PTSD and dementia in the same breath as one hippocampal injection that landed in the wrong place — the packaging travels considerably further than the result. Ramirez himself pulls the other way, calling the engram a theoretical construct and conceding there is no zoomed-in map of one, which is why this reads as inflation in the framing rather than overclaiming by the scientist.
A book tour, plainly labelled
This is a publicity interview and does not pretend otherwise: Princeton University Press has a 2025 title to sell, the book is on a 2026 prize shortlist, and Live Science runs the conversation next to an excerpt. The serendipity story — botched surgery, career and then some — is exactly the anecdote a book tour is built around. The interests are visible rather than hidden, which is the mitigating half of the score.
Confident about what was said, not about what was verified
We are on firm ground reporting Ramirez's words, dates and titles — those are unambiguous in Live Science's piece. We are on soft ground everywhere the story characterizes the world beyond his lab, including the claim that the field was propelled and the counterfactual six-month delay. With one outlet and one voice, there is no second reading to raise or lower that.