Engram
Memory, the ability to store and recall information, is a fascinating phenomenon that has intrigued scientists for decades. In this blog, we’ll delve into the concept of an “engram,” a fundamental unit of physical memory substrate. We’ll explore the biological and computational principles governing memory formation and linkage. Join us on this captivating journey into the depths of memory research!
Tracing the Origins of Memory Study
The study of memory dates back to the early 20th century when German biologist Richard Semon coined the term “engram.” He described engrams as lasting physical changes in the brain that occur after learning or experiencing something. These engrams, once formed, lie dormant and can be awakened by similar experiences or cues, giving rise to what we know as memory recall. However, Semons’ initial definition left many questions about the biological nature of these changes unanswered.
Cracking the Memory Code
Over the years, advancements in neuroscience have shed light on the mechanisms underlying memory. The brain consists of nerve cells that communicate through electrical impulses, and memories are believed to be stored as changes in patterns of how neurons interact with each other.
Unraveling Memory Formation
Studying memory formation requires two crucial elements: a suitable behavioral task and the means to monitor brain changes experimentally. While human subjects can explicitly describe their memories, studying individual neurons’ activities is only feasible in model organisms such as rodents. To overcome this, scientists employ clever experimental designs, like the fear conditioning paradigm.
Fear Conditioning: A Gateway to Memory Research
Fear conditioning is a popular experimental setup used to study memory formation. In this paradigm, an association is formed between a neutral stimulus (like a sound tone) and an aversive stimulus (like a mild foot shock). The animal learns to associate the previously neutral stimulus with pain, resulting in the formation of an associative memory.
Shining a Light on Memory Encoding
To visualize memory encoding in the brain, scientists use immediate early genes, such as Fos and Arc, which get activated in neurons undergoing plastic changes during learning. By coupling the activation of these genes with the production of fluorescent proteins, researchers can visually identify neurons involved in memory encoding.
Challenges in Memory Tagging
One challenge in memory research is the timing of tagging memory-related neurons. Traditional methods may tag neurons that encode irrelevant memories before the desired memory is formed. To overcome this, researchers have engineered a tagging system that activates only during the presence or absence of a specific chemical compound, offering better control over the timing of tagging.
The Engram Revealed
With the improved tagging system, scientists can now isolate and study the engram responsible for storing a specific memory. The engram consists of a subset of neurons that underwent modifications, indicating their involvement in memory encoding.
Cracking the Memory Recall Code
During memory recall, the same pool of engram cells becomes activated. But does their activation cause memory recall, or is it just a byproduct? Studies using selective suppression of engram neuron activity show that engram neurons are indeed necessary for memory recall. Activating these neurons in the absence of the conditional stimulus also induces memory recall, proving that engram neurons are both necessary and sufficient for retrieving specific memories.
Beyond Fearful Memories
The principles of engram reactivation extend beyond fearful memories. Similar results have been observed in memory paradigms where mice associate conditional stimuli with rewards, not just pain. This suggests that the activation of engram neurons is a fundamental mechanism governing memory recall.
The Engram Sparsity Mystery
In the brain, experiences activate a large number of neurons, but only a small proportion of them become part of the engram. This sparsity varies across brain regions, with the amygdala showing 10–20% engram allocation and the dentate gyrus only 2–6%. Surprisingly, within one brain area, the sparsity remains constant across different memories, irrespective of stimulus strength or memory content. The engram’s constant sparsity suggests the presence of internal mechanisms that control memory allocation.
A Sparse Distributed System
The brain seems to employ a sparsely distributed system for information coding and computations. Representations with non-overlapping codes provide higher storage capacity and robustness to noise, making sparsity a critical characteristic to maintain.
Competition through Neuronal Excitability
Neurons are electrically excitable cells, that generate action potentials or spikes when stimulated. Intrinsic excitability refers to a neuron’s readiness to fire and transmit information. Neurons with higher excitability have a higher probability of being recruited into a memory trace. Experimental interventions, like altering the excitability of neurons using light-sensitive ion channels, offer opportunities to control memory allocation.
Inhibitory Neurons and Memory Gating
The competition among neurons for memory allocation is carried out through local microcircuitry, involving inhibitory neurons. Principal neurons, the memory competitors, connect to interneurons that inhibit other nearby principal neurons. The most excitable neurons indirectly suppress their neighbors, affecting the final engram size.
A Distributed Memory Network
Research has shown that memories are not confined to one specific brain region but are distributed across sparse ensembles of neurons scattered throughout the brain. A single fear memory, for example, can elicit an engram complex spanning various brain regions, including the hippocampus, amygdala, thalamus, hypothalamus, and brain stem.
Links between Memories
To utilize stored information, memories must be linked together to form abstract concepts or principles. The brain encodes the connections between memories as the degree of overlap between the populations of engram neurons. This physical manifestation of memory connections allows for the integration of related memories.
Time Windows of Excitability
Neuronal excitability changes throughout a neuron’s lifetime, with elevated levels lasting several hours before gradually decreasing. If excitability were fixed, the same small pool of neurons would win memory competitions repeatedly, leading to overlapping memories and reduced storage capacity.
The Overlapping Engram Puzzle
Experiments with closely-timed fear conditioning sessions reveal overlapping engrams for two different memories. The linked memories become functionally associated, leading to simultaneous extinction. However, if memory sessions are separated by a longer duration, the engrams remain non-overlapping, allowing for the extinction of one memory without affecting the other.
Co-retrieval and Memory Linkage
Memories can also be linked through co-retrieval, where initially non-overlapping engrams become repeatedly reactivated together. Co-retrieval reorganizes memory traces, generating an assembly of neurons shared by both engrams. This shared pool of neurons represents the linkage between memories, allowing for subsequent recall of one memory to trigger the recall of the other.
The Emergence of Memory Abstraction
The shared pool of neurons resulting from co-retrieval is not essential for recalling individual memories on their own. However, it holds critical information about the link between the memories, contributing to memory abstraction and the general principles of learning.
The Enigmatic Engrams
In summary, engrams encode information about experiences as synaptic changes in sparse populations of neurons. This sparsity is maintained through competitive selection, where highly excitable cells are preferred for memory allocation. Activating these chosen neurons is both necessary and sufficient for memory recall, enabling the manipulation of memories experimentally.
The Puzzle of Memory Linkage
Multiple engrams can become linked through shared neurons, enhancing our ability to recall related memories. This linkage may underlie the brain’s capacity for abstraction, forming a complex web of connected memories, like pieces of a puzzle.
Unveiling the Potential of Cognition
Understanding the intricate workings of engrams not only unravels the mysteries of memory but also holds the key to unlocking the vast potential of our cognitive abilities. As we continue to explore the fascinating world of memory research, we unveil the secrets of our own minds and gain deeper insights into the wonders of human cognition.
