Childhood amnesia
Also known as: Infantile amnesia
The near-absence of memories from our first few years of life.
What it means
Childhood amnesia is the typical inability of adults to recall episodic events from roughly before age three to four, together with a sparseness of memories up to about age seven. Proposed causes include the immaturity of the hippocampal and prefrontal systems that bind episodic memories, the absence of a developed self-concept and language to scaffold autobiographical narrative, and rapid neurogenesis that may overwrite early traces. The offset age varies with culture and with how much parents reminisce in elaborative ways with children. It is not a failure to learn — infants learn prodigiously — but a failure of those early episodes to remain accessible in autobiographical form. It marks the lower boundary of the autobiographical memory distribution.
Two families of explanation
For a century the debate has split between two accounts, once treated as rivals. On the first, early memories are never durably built: the hippocampus and prefrontal cortex are still maturing, and the child lacks the stable self-concept and language that later knit an event into an autobiographical narrative. On the second, the memories form but become irretrievable, either decaying or being overwritten as the brain reorganizes. The current view is that both operate, with the weight of recent evidence tilting toward the second. A concrete candidate mechanism is the rush of new neuron growth in the infant hippocampus, which may destabilize the circuits holding an existing trace, clearing capacity for new learning at the cost of old episodes.
What the evidence shows
Animal work made the mechanism testable. Suppressing hippocampal neurogenesis in infant mice slows the rapid forgetting of learned fear; boosting it in adults speeds their forgetting, implicating neuron growth directly. More strikingly, Guskjolen and colleagues optogenetically reactivated the original memory trace and recovered a 'lost' infant fear memory in adult mice, showing the trace had persisted while being inaccessible. In humans, functional imaging of awake infants found the hippocampus already encoding individual events from roughly one year of age, with encoding activity predicting later memory-driven looking. That the machinery to form episodic memories is present, yet the record is later unavailable, argues that post-encoding inaccessibility, not a failure to encode, does much of the work.
The offset is not a fixed age
The 'three-to-four' boundary is soft and partly social. Cross-cultural work reports the earliest memories sit later for adults raised in some East Asian settings than for New Zealand European or Maori adults, tracking how much a culture elaborates on the personal past. Within families, mothers who reminisce in an elaborative, detail-seeking style have children whose autobiographical memory emerges earlier. Measurement matters too: people systematically misdate old memories as more recent than they were, and revise their reported 'first memory' forward with age, a telescoping effect. The offset one records therefore depends on the interview method as much as on the brain, which is why point estimates for the earliest memory range across roughly two-and-a-half to four years.
Why it matters
Childhood amnesia is a developmental boundary, not ordinary decay, and treating it as such sharpens judgment about early recollections. A confident, vivid 'memory' from before age three is more likely a reconstruction assembled from photographs, home video, and repeated family retellings than a preserved episode, however sincerely it is felt. That has real stakes in clinical and legal settings, where narratives of pre-verbal experience are sometimes offered as recovered fact. The phenomenon also reframes a common worry: forgetting infancy is not a defect but the normal lower edge of the autobiographical record, the floor beneath the reminiscence bump. What survives from those years is largely semantic knowledge and skills, not the felt, first-person episodes that define autobiographical memory.
Examples
Few adults can recall genuine episodic events from their second birthday, however momentous at the time.
You 'remember' the beach holiday you took at three, but what you have is the photograph and your mother's retelling. The episode itself is gone, though you were plainly there.
A four-year-old will describe her brother's birth in detail. Ask her at twenty and there is nothing — she learned it fine; it simply never survived into adult autobiographical form.
Brands selling 'the cereal of your childhood' trade on warm feeling, not recall. Adults retain essentially no episodic memory of eating it as toddlers, only the family lore and packaging around it.
A therapy client 'recovers' a vivid, detailed memory of an event from age two. Because durable episodic recall from that age is near-absent, clinicians treat such early narratives as reconstructions, not verbatim records.
First described in Noted by Freud; modern study by Patricia Bauer, Carole Peterson, and others.
Key references
- Yates, T. S., Fel, J., Choi, D., Trach, J. E., Behm, L., Ellis, C. T., & Turk-Browne, N. B. (2025). Hippocampal encoding of memories in human infants. Science, 387(6740), 1316-1320. doi.org/10.1126/science.adt7570
- Peterson, C. (2021). What is your earliest memory? It depends. Memory, 29(6), 811-822. doi.org/10.1080/09658211.2021.1918174
- Guskjolen, A., Kenney, J. W., de la Parra, J., Yeung, B. A., Josselyn, S. A., & Frankland, P. W. (2018). Recovery of "Lost" Infant Memories in Mice. Current Biology, 28(14), 2283-2290. doi.org/10.1016/j.cub.2018.05.059
- Bauer, P. J., & Larkina, M. (2014). The onset of childhood amnesia in childhood: A prospective investigation of the course and determinants of forgetting of early-life events. Memory, 22(8), 907-924. doi.org/10.1080/09658211.2013.854806
- Josselyn, S. A., & Frankland, P. W. (2012). Infantile amnesia: A neurogenic hypothesis. Learning & Memory, 19(9), 423-433. doi.org/10.1101/lm.021311.110
- MacDonald, S., Uesiliana, K., & Hayne, H. (2000). Cross-cultural and gender differences in childhood amnesia. Memory, 8(6), 365-376. doi.org/10.1080/09658210050156822