Recall your earliest memory: birthday candles, the coat of a childhood dog, the scent of your mother’s perfume. Regardless of the specific memory, it is likely you were at least a few years old when it occurred. Toddlers under three are highly conscious of their surroundings—observe any toddler happily chewing on a toy or shrieking at fireworks—and they appear to recall recent events. Then, suddenly, they do not.
A mysterious divide separates us from our earliest life experiences. This phenomenon is not exclusive to humans. Studies indicate that mice exhibit a comparable pattern of memory loss. Mice that master maze escape as mere pups forget the skill by adulthood. Similarly, young mice conditioned to fear a shock chamber fail to recognize it once matured.
Recently, researchers investigating this occurrence—often referred to as childhood or infantile amnesia—have uncovered surprising insights that shed light on the mechanics of this nearly universal type of forgetting.
Are the memories erased, or simply inaccessible?
It may appear that early life memories are simply deleted. However, certain studies imply that in mice, these memories persist and can be retrieved.
In the Toronto lab of Paul Frankland, a senior scientist at the Hospital for Sick Children, scientists tagged brain cells activated when young mice learned to fear a specific chamber. Three months later, after the adult mice had lost their fear, the researchers re-stimulated those cells—and the fear suddenly returned.
This indicates that—at least in mice—early memories are not obliterated; they are merely beyond the scope of standard recollection. It is possible that aspects of brain development render these memories inaccessible, akin to locked rooms with lost keys.
Indeed, animals that typically generate fewer new neurons after birth—such as guinea pigs—do not exhibit signs of this amnesia, Frankland and his team found.
The laboratory of Tomás Ryan, a neuroscientist at Trinity College Dublin, has also discovered that certain mice are less susceptible to forgetting. Specifically, male mice born to mothers with activated immune systems during pregnancy showed reduced forgetting.
Furthermore, Ryan’s team has emphasized the role of immune cells known as microglia in infantile amnesia: suppressing microglia activity during a critical developmental phase prevents mice from forgetting maze solutions as they age. This points to the immune system’s potential involvement in infantile amnesia.
But what about humans?
Determining what occurs inside the brains of infants and toddlers is challenging. (One major hurdle: sitting still for brain scans is not their forte.) Nevertheless, Nick Turk-Browne at Yale University and his associates have successfully scanned the brains of numerous young children. They found that children as young as one year old form episodic memories, creating recollections of past events just like adults do. This implies that humans may also be generating memories that later become inaccessible.
Is taking a two-year-old on vacation worthwhile if the memories will vanish? ‘I am frequently asked: What can be done to stop this?’ says Turk-Browne. ‘You can discuss the trip often or look at photos. But the authentic, pure memory—the one he hasn’t recalled in a long time, the one you haven’t discussed with him—will soon disappear, for better or worse.’
To gain a clearer understanding of exactly when memories are created and lost, Sarah Power at the Max Planck Institute for Human Development and her team constructed a media room where children undergo experiences found nowhere else. ‘A crucial aspect of this task is that everything exists solely within the lab. We wanted to ensure complete uniqueness so that the contextual environments do not exist in the real world; this way, if they remember these associations, we know it is only because they were in the lab,’ she explains. They have observed 400 toddlers aged 18 to 24 months, tracking their memory formation of the lab space, and plan to monitor them over time. The project is in its nascent stages, but ‘based on preliminary data, we have been pleasantly surprised by their capacity to encode and retain these episodic-like memories,’ she notes.
In a smaller experiment led by Turk-Browne, parents filmed events from their child’s perspective. Subsequently, the researchers showed the children these personal videos alongside footage from strangers during brain scans conducted over several sessions spanning two years. The objective is to determine whether videos depicting a child’s own viewpoint trigger detectable memory activity in brain scans, and if so, precisely when that effect fades.
Why do we forget?
It remains a mystery why the brains of humans and other mammals discard our early life experiences. ‘I question what this reveals about human memory in general, education, early life, and learning. Is this a biological switch, or merely a consequence of intense learning?’ Ryan asks. ‘Essentially, is the brain actively deciding to suppress these memories, or is it an unintended side effect of rapid learning during that period?’
Could holding onto our earliest memories somehow threaten our survival? Alternatively, is the value of those memories tied to functions that do not require conscious retrieval, meaning their forgetting is inconsequential?
Turk-Browne speculates that the purpose of our earliest memories is to construct a mental database of how the world functions. The specific details—the elements preserved in episodic memory—may not be the most valuable aspect.
‘Most memory researchers view the adaptive value of memory as the ability to act appropriately in new situations based on past experience,’ he states. ‘There is substantial behavioral evidence that even newborns excel at aggregating statistics’—constructing a consistent model of the world that helps us make decisions and control our environments. Regardless of why our memories disappear, they may persist in ways we fail to recognize.
