Memories Are Made Of This
We often assume that our memories are set in stone. That they are play-by-play recollections of the past so that no matter how many times they are recalled or how long ago they occurred they remain in our brains as if they happened yesterday. But memories are not set in stone. Everyday we witness their loss, alteration and even creation. More often than not the malleability of a memory triggers mundane second-guessing - those ‘did I really lock the front door when I left the house this morning?’ moments. But this property can give rise to more serious repercussions. So we’re here to give you the low-down on all things memory – what they are, how they can be altered and what consequences this might have.
Memory, in its simplest terms, is the capacity of an organism to acquire, store and recover information based on experience . It is of utmost importance in allowing us to carry out our day-to-day lives. Despite this significance, the neural mechanisms subserving this phenomenon have remained ambiguous until recently.
In the early twentieth century a German zoologist named Richard Semon coined the term engram, described as ‘the enduring though primarily latent modifications in the irritable substance produced by a stimulus’ – in other words the physical changes in the brain that constitute a memory . But what are these physical changes - and how do they come about?
Canadian psychologist Donald Hebb was a pioneer in answering these questions. He proposed that during learning, synaptic connections between a subset of neurons are strengthened, resulting in co-ordinated firing. These neurons that fire together at the time of learning do so again when this memory is later retrieved. According to his postulate ‘cells that fire together, wire together’ .
Hebb’s work acted as a predecessor to a much broader term, synaptic plasticity, which is a fancy name that refers to the ability of neurons to bi-directionally modify the strength of their connections. Synapses can either increase in strength, a process known as long-term potentiation, or decrease in strength, long-term depression. Following long-term potentiation (LTP) two principle changes occur; an increase in the efficiency of AMPA receptors at the post-synaptic membrane, and the insertion of new ones . These changes improve communication between the pre-synaptic and post-synaptic neuron, in turn enhancing the strength of the synapse – this is what is responsible for the increased connectivity between a subset of neurons when we learn.
So neuronal connectivity has been widely accepted to underlie memory formation, but how can we prove this? After all it does sound like quite a simple mechanism. Well, to demonstrate that a cell population is the basis of a specific engram it is necessary to conduct a mimicry experiment to show that direct stimulation of such a population is sufficient to re-activate the memory with which it is associated . The development of recent tools and technologies has opened up doors to do this. However, by exploring these new methods by which memory can be manipulated much has been revealed about its labile nature upon re-activation.
Typically, a memory is elicited when the quiescent engram is awakened by an external retrieval cue. Studies in rodents have revealed that engrams can also be awakened artificially by stimulating their neuronal components . In order to do this, these components must first be identified and labelled. When neurons are active during memory formation they express activity-dependent genes called immediate early genes (IEGs). By coupling the expression of these IEGs to a neuronal marker these neurons can be identified. One way to do this is by using a light-sensitive channel such as channelrhodopsin-2 (ChR2). These neurons can then be optically re-activated by light directed at these channels. The idea is, since this subset of neurons represent a particular memory, their stimulation should lead to its retrieval – clever right? .
This crafty tool has been utilised in a number of recent rodent studies looking at the hippocampus, which is a structure critical for the formation of contextual memories. In one such study, mice were exposed to a particular context A, and active neurons were labelled with ChR2. Mice were then placed into a different context B, and a foot-shock was administered whilst these labelled neurons were optically reactivated with light. When placed back into context A, mice subsequently exhibited freezing behaviour and hence were freezing to a context in which they were never actually shocked . In this way, optical reactivation of cells that were naturally activated during the formation of a contextual memory induced the retrieval of that memory. More importantly, the retrieved memory became associated with a foot-shock to form a new memory that had never had its component experiences naturally linked – a false memory had been created!
If memory distortion can occur so easily in rodents, it begs us to question the reliability of memories in general. Is it probable that false memories in humans can occur as a result of this interference? If so, how much of what we remember actually happened and how much of it is false? This is a scary thought for such a precious phenomenon.
In most circumstances a false memory is probably a minor modification of the true details of an event rather than a complete distortion, but this isn’t always the case; sometimes false memories can have more serious consequences. In 1986, Ronald Cotton was sentenced to life imprisonment in North Carolina for the rape of Jennifer Thompson. 11 years later, the emergence of DNA testing proved Cotton’s innocence . Thompson, having made every effort to study the perpetrator’s face while he was assaulting her, had chosen Cotton as her attacker in a photo line-up. Somehow the memory of her attacker, upon retrieval, had become distorted to the extent that she identified an innocent man as responsible for the crime. This is a prime example of how dangerous memory fragility can be – quite unsettling that an innocent individual like ourselves has ended up behind bars!
However, sometimes the ability to modify a memory may actually prove to be a useful tool. For instance, imagine living with a memory so distressing it leaves you unable to carry out your day-to-day life – just like the millions of war veterans coping with traumatic stress. Now imagine being able to delete this memory or to dampen the emotional connotations associated with it. Such modifications could prove invaluable in situations like these.
Fortunately, such techniques have demonstrated utility in rodent studies. In a similar experiment to the above, mice were placed into a novel context and exposed to a foot-shock, with fear memory engram-bearing cells labelled using ChR2. Mice were then placed into a different context and exposed to both light-on and light-off epochs. During light-on conditions, when these cells were optically re-activated, the mice demonstrated freezing behaviour indicative of retrieval of the foot-shock memory. This behaviour was not witnessed during the light-off epochs, so essentially the authors were able to turn the fear memory on and off with the use of light . Being able to control an engram at the flick of a switch is an astonishing thought! With development of this technology, sufferers of devastating conditions such as post-traumatic stress disorder (PTSD) could be relieved of their constant anguish. Obviously there are major ethical considerations that should be considered in translating the removal of a memory for use in human maladies, but it is an interesting idea to consider.
Another, perhaps less severe approach, would be to edit the emotional connotation of a memory. In a different experiment, mice were exposed to a foot-shock and once again the engram-bearing neurons were labelled with good old ChR2. These mice were then exposed to a positive experience consisting of an interaction with a female mouse whilst the original fear engram was re-activated. As a result, when this memory was now stimulated the mice displayed a response indicating a switch of the memory valence from negative to positive . Translating this to human maladies, it might be possible to take a sufferer of PTSD, label the engram of a traumatic event and re-associate it with a nice memory – that holiday to Spain, or your last birthday perhaps? This could be a potentially life-changing tool.
To sum up, memories are precious. They constitute what we know about ourselves, our relationships and our surroundings. We often assume that memories are fixed, error-free recollections of our past, but they are much more prone to error than many people realise. It goes without saying that the fragility of an engram can lead to devastating consequences, but being able to utilise this property to engineer traumatic memories may offer a bright future for sufferers of PTSD.
By Harriet Gallegos
Harriet is a research masters student at the University of Bristol. During her undergraduate degree in Neuroscience, she acquired a particular interest in molecular cell biology, looking at the protein molecules involved in receptor trafficking during synaptic plasticity. She was specifically curious about studying these processes in the context of neurological disease; subsequently, her project is currently looking at AMPA receptor trafficking in Alzheimer's Disease.
Harriet is interested in using her research experience in the future to work in the pharmaceutical drug discovery process and contribute to finding life-saving medicines.
1. Josselyn, S., Köhler, S., & Frankland, P. (2015). Finding the engram. Nature Reviews Neuroscience, 16(9), 521-534.
2. Semon, R. W. (1921). The mneme. London, New York: G. Allen and Unwin Ltd.; The Macmillan Company.
3. Hebb, D. O. (1949). The Organization of Behavior: A Neuropsychological Theory. New York: Wiley.
4. Bear, M.F., Connors, B.W. and Paradiso, M.A. (2015) Neuroscience: Exploring the brain. 4th edition. Philadelphia: Lippincott Williams and Wilkins.
5. Liu, X., Ramirez, S., Pang, P., Puryear, C., Govindarajan, A., Deisseroth, K., & Tonegawa, S. (2012). Optogenetic stimulation of a hippocampal engram activates fear memory recall. Nature.
6. Liu, X., Ramirez, S., & Tonegawa, S. (2013). Inception of a false memory by optogenetic manipulation of a hippocampal memory engram. Philosophical Transactions Of The Royal Society B: Biological Sciences, 369(1633), 20130142-20130142.
7. Loftus, E. (2003). Our changeable memories: legal and practical implications. Nature Reviews Neuroscience, 4(3), 231-234.
8. Redondo, R., Kim, J., Arons, A., Ramirez, S., Liu, X., & Tonegawa, S. (2014). Bidirectional switch of the valence associated with a hippocampal contextual memory engram. Nature, 513(7518).