My four-year-old grandson has a near photographic memory for outings we took a year ago, for people he met six months ago, and for books he has not seen in weeks. Soon that will all disappear. By the time he is eight, he will remember almost nothing of his first four years. And by the time he is eighteen, he will remember details going back to age four and he will retain language and habits acquired before age four, but in terms of autobiographical or episodic memories, the earliest years will be almost a complete blank. This normal loss of early memories is called infantile amnesia.
Freud, a source not cited often in this space, was one of the first to write about infantile amnesia. He attributed the loss of early memories to repression, an active forgetting of early experiences because of their heavily charged psychosexual content. Others have explained infantile amnesia as due to the absence of language, since words seem important for encoding certain kinds of memory. Still others have cited the need for a sense of self to provide a reference for early memory.
The problem with these explanations is that other mammals, lacking language and presumably less burdened by psychosexual conflicts, show the same loss of early memories while having perfectly good long-term memory for later experiences. If mice and rats and other mammals also have infantile amnesia, what is the mechanism? It’s not simply an inability to learn early in life. Very young mammals, like my grandson, are sponges for information. The problem is that this information is either not accessible or not retained in later development.
Recent research on this problem by Paul Frankland, Sheena Josselyn, and their colleagues at the Hospital for Sick Children in Toronto suggest a surprising explanation.1,2 They suggest that infantile amnesia is due to the rapid birth of cells in the hippocampus during infancy. The hippocampus has long been known to be associated with episodic memory, including autobiographical memory. And many scientists over the past two decades have shown that brain cells in the hippocampus continue to be generated through adulthood. This process, called neurogenesis, occurs in humans as well as mice. Many studies have shown that exercise increases neurogenesis, while stress decreases it. And several investigators have suggested that neurogenesis is important for learning.
Frankland, Josselyn, and colleagues are suggesting paradoxically that neurogenesis is important for forgetting. Specifically, they hypothesize that the increased rate of neurogenesis during infancy leads to loss of the memories that would otherwise be stored long-term. What’s the evidence? When they experimentally increase neurogenesis in adult mice, they induce a change in memory that looks like infantile amnesia. And when they experimentally decrease neurogenesis in infant mice, they reduce the amnesia. They have also studied mammals that are born relatively mature, like guinea pigs and degus. These species do not have the rapid neurogenesis of infancy and do not show infantile amnesia.
Why is this important? Beyond the mystery of why we lose our earliest memories, understanding a natural process like infantile amnesia can help us discern the fundamental connections between brain and behavior. Do traumatic events (which presumably influence neurogenesis) get encoded and stored differently during infancy? What about attachments in the first three years? Do individual differences in neurogenesis translate into differences in recall? Would changes in neurogenesis lead to changes in memory in humans? All of these issues remain to be explored. At this point we can say that a mystery that was fundamental to Freud’s theory of sexuality may now be solved with the tools of modern neuroscience.
2 Josselyn SA, Frankland PW. Infantile amnesia: a neurogenic hypothesis . Learn. Mem. 2012 19:423-433. Doi:10.1101/lm.021311.110.