My husband was recently describing something that happened on a past holiday. I, however, had no recollection of what he was telling me. He couldn't quite believe it. We know that "recollections may differ", but why do I not have this memory? I'm busy at work – have I simply run out of space?
It's a tempting explanation. We talk about "information overload" as though the brain were a container that eventually reaches capacity. But the brain does not fill up. Instead, it filters. At any given moment, far more information is available to us than we could ever realistically store. The conversations of even a single day would overwhelm any system that attempted to record them in full.
Instead, the brain relies on selection. Attention determines what is noticed. Emotion helps determine what matters. Then, structures decide what is worth committing to longer-term memory. If your attention is elsewhere, the process falters at the first step. On that holiday, my husband may have paused long enough to register the moment. I may have been thinking about where we were going next, moving through the day without stopping to take it in. Without focused attention, experiences are only weakly encoded. In that sense, the memory was not lost. It was never fully formed.
Even when memories are successfully encoded, they are not stored as fixed records. Each time we recall an event, we reconstruct it, drawing on sensory detail and expectation. With repetition – through conversation or retelling – those reconstructions become stronger. This helps explain why shared experiences can diverge dramatically. We assume that living through the same moment should produce the same memory, but the brain does not work that way. It does not passively record experience. It actively selects, prioritizes, and discards.
The feeling that our brains are "full" arises not because we have run out of storage, but because we have reached the limits of what we can process at once. Attention is finite. When these systems are saturated, new information struggles to gain a foothold. This is the mental equivalent of too many tabs open: nothing has been permanently lost, but everything becomes harder to manage.
Computing analogies are useful up to a point. If working memory resembles RAM, then long-term memory is compared to a hard drive. But this is where the parallel breaks down. A hard drive stores files in fixed locations. The brain does not work this way. Memories are not stored as discrete files; they are distributed across networks of neurons, reshaped and reassembled each time they are recalled. New experiences do not simply add to what is already there – they interact with it.
One widely cited figure puts brain capacity at around a petabyte. It is an impressive number, but also a somewhat misleading one. It implies a storage system that fills up over time, when in reality, the brain is constantly reorganizing itself. Capacity is not fixed. Information is integrated, modified, and, when no longer useful, allowed to fade.
Some memories fade because they are not continually reinforced. Memory is not preserved simply because it matters to us. It is preserved when it is retold or reconnected to other experiences. What is lost, in most cases, is not the memory itself but our ability to retrieve it. The trace remains, but it has slipped out of reach. The absence of a memory is rarely evidence of a system at capacity – more often, it is the trace of a moment that was never fully stored.