Imagine learning to ride a bike for the first time. Initially, it feels awkward. With each attempt to balance, pedal, and steer, a memory trace begins to form in your brain. Areas responsible for motor control and coordination light up as your motor cortex learns how to direct your muscles efficiently. The feeling of gripping the handlebars and the wind blowing simultaneously activate sensory regions of the brain, further enhancing the memory trace. Practice strengthens the memory trace, consolidating it into procedural memory. Years later, even if you have not ridden a bike in a while, you are able to ride with little to no difficulty. This “muscle memory” is only possible because the memory trace remains strong in your brain.
Every memory stored in our brains is physically represented by memory traces. German biologist and psychologist, Richard Semon, was the first to introduce this concept in 1904, coining the term engram. Engrams describe the physical and biochemical changes that occur in the brain when memories are formed. Semon suggested that lasting traces are created from our experiences that can be reactivated when recalling the memory. However, when studies failed to locate engrams in the brain, the concept was largely abandoned for decades. Engrams only resurfaced when Canadian psychologist Donald O. Hebb began investigating the connections between neurons in the 1940’s.
The hippocampus serves as the primary learning and memory center of the brain, but this is not where our memories are held. Rather, memory traces or engrams are represented by the connections between neurons throughout the brain. Depending on which neurons are activated during learning, the subsequent connections formed when multiple brain regions fire simultaneously form the physical representation of the memory. Hubb’s theory summarizes “neurons that fire together, wire together,” meaning that neurons that fire at the same time during an experience from connections.
Each time these connections are activated, through practice or recall of that memory, the network becomes stronger. Eventually, the engram is so strong that activation of one aspect of the memory activates the entire network. For example, imagine walking into a bakery and being hit with the smell of fresh bread. The sensory stimulation may transport you back to your grandmother’s kitchen, where you awoke to freshly baked bread every summer. Now you do not just smell the bread, but you can also hear the sound of her humming, feel the warmth of the kitchen, and envision the smile on her face as you walk in. Each component of the memory is part of a single engram, activating the entire network of related memories.
This phenomenon was first confirmed in a series of studies involving mice. Engrams were able to be localized and measured in the brain during experiments involving a type of learning process called classical conditioning. Classical conditioning, in its simplest form, associates normally neural stimuli with a conditioned response. For example, in a fear conditioning context used in many early engram studies, a mouse may experience a shock in a particular region within their environment that was previously safe. The experience of getting shocked activates an engram in the brain, which stores the unpleasant memory. As shown in the figure below, the activated neurons are then labeled with fluorescent molecules. When the mouse encounters the shock again in that region, the same neurons appear to reactivate. Even when a shock is not given, being placed in the same location is enough to reactivate the engram and cause the mouse to freeze in fear.
In the more than one hundred years since engrams were first unveiled, the mechanics of how these connections form, strengthen, and weaken over time have largely been a mystery. Now, emerging research seems to be piecing more of the puzzle together. The implications of this research may one day pave the way for novel therapeutic approaches that protect against Alzheimer’s disease and other forms of dementia.
