Skip to main content

Neuroethics Journal Club Report: “Creating a false memory in the hippocampus” Ramirez et al. Science 2013

Our memory can be unreliable, that comes as no surprise. But beyond forgetting where the car is parked or misremembering a date, a perhaps more interesting phenomenon is that of false memories of events that have never happened, or at least not to us directly. In most cases, the fallibility of memory is benign or occasionally embarrassing, but in the courtroom it can have serious consequences. In the final Neuroethics Journal Club of the semester, Emory University graduate student and AJOB Neuroscience editorial intern, Katie Strong, led a thought-provoking discussion of Ramirez’s 2013 Science paper1 entitled “Creating a false memory in the hippocampus” with a focus on the potential neuroethical implications of this research on the justice system.

The discussion paper comes from 1987 Nobel laureate Susumu Tonegawa’s lab and is in some ways a sequel to their 2012 paper published in Nature2. In both studies this group utilized an elegantly-designed mouse model with the aim of targeting the cells in the hippocampus constituting the memory engram. The search for the engram, or memory trace in the brain, is not a recent pursuit. Karl Lashley’s seminal mid-20th century work suggested that memories are dispersed throughout the cortex. Lashley’s lesion studies surprisingly indicated that the amount of cortex damaged mattered far more than the location of the lesions3.

More recently the search has moved toward molecular changes in individual cells and at particular synapses4.  It is now thought that Lashley’s findings may have been the result of the complexity of the tasks that his animals performed, which involved multiple brain regions, since emerging evidence suggests that in some circumstances the same particular cells are activated during recall of certain memories5. Conversely, fear conditioning is a relatively simple paradigm that has been widely used because animals rapidly learn to pair an innocuous cue such as a light, tone, or an environment with an aversive stimulus such as a foot shock. Moreover, the neural circuitry controlling these behaviors has been extensively studied6.

In this paper, Ramirez and colleagues used a very similar approach to that in their 2012 paper in which they employed optogenetics (a technique to genetically target a group of cells that can later be activated by pulses to light) to label and reactivate hippocampal engram neurons2. Mice were unilaterally injected into the dentate gyrus of the hippocampus with a virus to drive the expression of a fluorescently-tagged, light-activated protein (channelrhodopsin-2) under the control of a drug-responsive promoter. These mice had been genetically engineered to activate this expression system when the immediate early gene c-fos, a marker of neuronal activity, is induced. The result was that expression of the light-activated channelrhodopsin-2 only occurs when (1) the animals are taken off of the expression-suppressing drug doxycycline and, (2) when neurons are sufficiently activated to induce expression of c-fos. This allowed the experimenters to essentially label the memory engram of a neutral context (context A) by removing doxycycline from the animals’ diet. Then, when doxycycline was replaced and neuronal labeling was halted, they were able to selectively re-activate the neutral context engram-bearing cells with rapid pulses of light.

In order to induce a false memory, the animals were removed from context A and placed in context B where the engram cells that encoded for context A were selectively stimulated with light while a foot shock was delivered. Later, when the animals were placed back in context A they demonstrated a typical fear response for rodents – freezing, even though they had actually experienced the foot shock in context B. The experimental group of mice exhibited a higher level of freezing than those animals that lacked the engram-labeling genetic engineering and were thus nonresponsive to the subsequent artificial light stimulation. As others7 have pointed out, this is not the first time that fear conditioning has been achieved with artificial stimulation in place of the conditioned stimulus, but it is the first time that this is been done by directly activating individual neurons in the brain.

These experiments may seem to be far-removed from typical human experience, but they may provide the basis to understand how false memories can be formed. As the authors point out, recall is known to make memories more labile and external information can occasionally be incorporated into existing memories over time. Moreover, they argue that these results may in fact be relevant to humans. In their words,

“we speculate that the formation of at least some false memories in humans may occur in natural settings through the internally driven retrieval of a previously formed memory and its association with concurrent external stimuli of high valence.”

For example, experimental psychologists have often been able to induce false memories in study participants through the use of leading questions and suggestion at a success rate of nearly 1 in 38. Often these paradigms use realistic and traumatic circumstances such as being lost as a small child in a mall. Perhaps the simultaneous recollection of a fear of being lost as a small child and memories of going to the mall at that age is enough to implant a false memory in some individuals.

Beyond the laboratory, false identification is an obvious and persistent problem for the justice system. The Journal Club discussion centered on a recent New Jersey case, State v. Henderson, which led to major reforms being enacted to change how eyewitness testimony is evaluated based on social science and psychology research9. This particular case involved the reliability of an eyewitness who encountered the suspect at gunpoint in a dark hallway when he had been drinking alcohol and smoking crack cocaine, and who continued to use crack daily until the police first contacted him more than a week later. Moreover, the witness reportedly struggled with a photo identification procedure and was pressured by the police to make a decision.

This case provides a hopeful example for how scientific research can spur progress improving accuracy and judicial outcomes, but what relevance does the Ramirez article really hold for understanding false memory? In these studies, the experimental group of animals – that which later displayed a false memory – had the memory of a neutral context linked to a foot shock by reactivating specific neurons using artificial means (optogenetics). This ability to activate a memory only by stimulating those (relatively few) neurons that were active during its encoding is strong, direct evidence for the existence and identification of the engram that can then be linked to a situation of high valence, such as a foot shock, to create a false fear memory. In reality, however, this is not how false memories are encoded in humans. It is an important step forward that elegantly demonstrates that the neurons involved in encoding a contextual fear memory are also sufficient for recall, but this technology is certainly not close to being used in humans. However, researchers have already found other ways to experimentally manipulate false memories in humans.

Transcranial magnetic stimulation (Source: TIME Magazine)

No one will be having viral injections and fiber optic cables implanted into their hippocampi anytime soon, but transcranial magnetic stimulation (TMS) is a safe, non-invasive technology that has already been used to affect false memory acquisition in humans10. Gallete and colleagues reasoned that since patients with left anterior temporal lobe (LATL) dementia often become very literal and less vulnerable to false memories, perhaps temporary inactivation of this brain area with TMS would reduce the rate of false memory acquisition and indeed it did. These results suggest that localized TMS to the LATL during learning could aid in factual recall so should it be more widely available to students?

Understanding how memories are encoded, consolidated, retrieved – and how this process can go awry – has long been a fundamental aim of neuroscience. New technologies such as optogenetics and TMS are allowing investigators to ask questions that were never possible until now. However, there is still much to learn in terms of how false memories are formed and how they can be minimized. While this aim will require continued work from social scientists, psychologists, and neuroscientists, neuroethical discussions will also be important in framing how false memories are understood and addressed from the laboratory to the courtroom.


1. Ramirez, S. et al. Science 2013.

2. Liu, X. and Ramirez, S. et al. Nature 2012.

3. Lashley, K.S. Physiological mechanisms in animal behavior 1950.

4. Govindarajan, A. et al. Nature Reviews Neuroscience 2006.

5. Josselyn, S.A. Journal of Psychiatry & Neuroscience 2010.

6. Kim, J.J. and Jung, S.W. Neuroscience and Biobehavioral Reviews 2006.

7. Saksida, L.M. Trends in Cognitive Science 2013.

8. Loftus, E.F. American Psychologist 2003.

9. Harvard Law Review 125:1514, 2012.

10. Gallate, J. et al. Neuroscience Letters 2009.

Want to cite this post?

Purcell, R. (2014). Neuroethics Journal Club Report: “Creating a false memory in the hippocampus” Ramirez et al. Science 2013. The Neuroethics Blog. Retrieved on

, from


Emory Neuroethics on Facebook