Learning to Remember: Untangling the formation of memories

Image courtesy of Wikimedia Commons.

Can you remember when you started to remember? At what point in time, and by what mechanism, does our brain gain the ability to form memories? A recently published study conducted by Usman Farooq and George Dragoi, assistant professor of psychiatry and neuroscience at the Yale School of Medicine, has probed the developmental timeline of memory formation and extracted key information about this process. This study aimed to uncover the mechanism through which animals like us develop the ability to form memories that link our past experiences with our present and future selves.

Conventional views of memory development

The scientific community has long known that memories are established, stored, and maintained in the brain. Broadly speaking, there are three types of memory: sensory, which lasts only a few seconds; working, which lasts from a few seconds to a minute; and long-term, which can last for years. Different parts of the brain, such as the hippocampus or amygdala, serve as mediators for these different types of memory. These brain regions house the networks and ‘bundles’ of interconnected neurons whose interplay forms the physiological foundation of memory.

Moreover, it is known that newborns are not able to form long-lasting memories immediately post-birth; for their first weeks to months, they are consciously in the present, after which they gain the ability to form memories. Although much is known about the anatomical and neurological support of new memory formation in adulthood, scientists know far less about the exact mechanisms that prompt the genesis of memory in early animal development. The present study fills this gap in knowledge, looking specifically at the neurological changes during the development of episodic memory–a type of long-term memory dealing with first-hand experiences–in rats.

Stages of memory formation

Farooq and Dragoi wanted to study the developmental timeline of memory formation in real time, so they monitored the electrophysiological activity of different “neuronal ensembles”–bundles of functionally connected neurons–in the hippocampi of rats over the third and fourth weeks after birth. They chose to monitor the hippocampi because they are known to be critical for the formation of new memories in adulthood. These rats were placed onto a linear track and allowed to move freely while their neuronal activity was recorded and decoded to display spatial information. “Particular neurons in the hippocampus ‘code’ or ‘fire’ for discrete locations; together, a population of these so-called ‘place cells’ can represent a whole arena composed of many locations,” Farooq said.  In other words, each instantaneous location along the track was represented by a specific pattern of neuronal activity. Furthermore, the neural activity of the rats was recorded during periods of sleep and “awake rest,” where the rats were awake but resting on the track.

The goal of the study was to see how the activity of these neuronal ensembles changes over time, as the rats gain the ability to remember their previous experiences on the track. This experimental design allowed for direct investigation of the development of episodic memory.

The researchers observed that episodic memory development occurs in three separate stages. In the first stage, around two weeks after birth, rats displayed neural activity characterized only by discrete, instantaneous locations represented by single neurons, lacking any pattern of activity characteristic of memory. “These young animals are completely ‘stuck’ in the present and cannot form memories,” Farooq said.

In the second stage, around the middle of week three, the rats began to display signals that functionally connected different neurons to one another. Although these signals were not yet characteristic of full-fledged memories, the neural patterns of the rats at this stage gradually started to organize into sequential motifs that “preplay” the patterns expressed during memory encoding, a kind of memory storage practice intrinsic to the adult animal brain. Interestingly, these preconfigured patterns were expressed during sleep, independently of any external stimuli.

By the onset of stage three–about three and a half weeks after birth–the rats had developed an ability to modify their preconfigured neural network of activity in response to a new experience. With this in place, the rats were now able to develop episodic memories. These memories were characterized by the strengthening and manipulation of the preexisting sequential patterns. Crucially, the rats in this final stage had developed the ability to use “replay”, wherein the preconfigured neural connections are used to relate different neurons–and thus, different locations on the track–in a sequential, long-lasting temporal manner. In other words, these rats had developed the ability to form memories.

Learning about memory

“Prior to our study, there was essentially a gap in our understanding of how and when these sequential patterns develop,” Dragoi said. Having observed these three stages of memory formation, the researchers were able to identify two interdependent prerequisites for memory development. First, the preconfigured patterns of sequential neural activity must form; next, the external world must impose some changes to these neural patterns of the brain. These steps must occur in that order, otherwise the brain has an insufficient foundation upon which to build a memory.

These distinct stages of memory development might have evolutionary roots. In the same time it takes for a newborn experimental rat to progress through the developmental steps in the confines of the linear track, a wild newborn would leave its nest to explore the surrounding environment. The emergence of an episodic memory is almost certainly essential for these animals to navigate an unforgiving external environment and stay alive.

The results of this study provide interesting insights into the animal mind. They bring us closer to elucidating the system by which the brain makes and stores memories. “This is an important question which has attracted the attention of philosophers for centuries. Now we’re able to directly test some of their hypotheses,” Farooq said.

Experimental challenges and future plans

Aside from valuable scientific contributions made by this study, it is also an archetypal developmental study. A tremendous amount of time and preparation was necessary to lay the groundwork and design the right approach to the problem. “For this developmental study, it took close to four years from the start of the experiments until its publication,” Dragoi explained.

As for the future, Dragoi and Farooq are looking to pursue a few different routes. One possible effort might focus on dissecting the factors, mechanisms, and neuronal circuits underlying the developmental processes investigated in this study. Another potential direction involves studying the development of neural patterns in animal models of neurodevelopmental disorders, potentially allowing comparisons to the current findings, to elucidate any deficits in the developmental system associated with the disorders.

In any case, this study has managed to provide some answers to the fundamental question of memory formation. This understanding will prove crucial to support further exploration of the miracle that is the animal brain.