You'll Forget Most of This Year of Your Kid's Life. Here's the Science.

By Durvesh Naik, maker of Everkid · 11 July 2026 · 7 min read

Last Tuesday my son said something at dinner that made my wife laugh so hard she had to put her fork down. I remember that she laughed. I have no idea what he said. It was gone by Thursday, and I write about memory apps for a living.

That's not a personal failing. It's how memory works, and there's a name for it: the forgetting curve. Once you understand the mechanism, the uncomfortable part isn't that you'll forget. It's how much of it is forgotten within days, not years.

The forgetting curve isn't a metaphor

In 1885, a German psychologist named Hermann Ebbinghaus ran an experiment on himself. He memorized lists of nonsense syllables, then tested his own recall at intervals afterward. What he found was a curve, not a cliff: memory doesn't fade evenly over time. It drops fast at first, then levels off. Roughly speaking, you lose the most ground in the first day, and whatever survives that first day tends to stick around much longer.

Ebbinghaus was memorizing made-up syllables, not his kid's dinner-table jokes, and it would be dishonest to hand you an exact percentage and claim it applies to parenting. But the underlying mechanism he identified, that a memory left untouched decays fastest in the first few hours and days after it's formed, has been replicated since, including a large modern re-run of his original methodology published in 2015 that found the same basic shape of the curve more than a century later. The specific numbers vary by what you're trying to remember. The shape doesn't.

And a Tuesday-night dinner joke is exactly the kind of memory this curve is unkind to. It was never important enough to write down, so it never got the one thing that would have saved it.

There's also a second finding from the same body of research that matters here: the drop-off isn't just steep, it's front-loaded. Most of what you're going to lose from a given day, you lose within that same day or the next. Wait a week to write something down and you're not writing down a slightly faded version of the memory. You're often writing down whatever thin residue survived the worst of the decay, which may be nothing more than the fact that something funny happened, with the actual words gone.

Why milestones survive and Tuesdays don't

Ask any parent to describe their kid's first steps and they'll give you the room, the shoes, who was holding the phone. Ask the same parent what their kid said at breakfast three Tuesdays ago and you'll get a shrug. Both are memories. Only one got encoded properly.

Psychologists call the difference encoding strength, and it comes down to two things: how much attention a moment gets while it's happening, and whether you do anything afterward to reinforce it. A first step gets both. It's rare, it's emotionally loaded, and you probably told three people about it that same day, which is itself a form of rehearsal. A funny sentence at dinner gets neither. It happens inside a stream of ordinary dinners that all blur together, and unless you repeat it to someone or write it down within hours, there's nothing distinguishing it from the thousand other things your brain heard that day.

This is also, separately, why researchers who study children's own memories have found that kids themselves retain almost nothing from before roughly age three or four, a well-documented pattern called childhood amnesia. That's a different phenomenon from what happens to a parent's memory, and I don't want to conflate the two. But it points at the same uncomfortable fact from the other direction: if you don't write it down, there may be nobody left, including your own kid, who remembers it happened at all.

It's worth sitting with what actually gets lost when a Tuesday goes unrecorded, because it's not the event. You'll probably remember, in outline, that your kid went through a phase of renaming household objects, or that there was a stretch where bath time involved an elaborate made-up game. What disappears first is the texture: the exact word they used, the specific rule of the game, the sentence structure that made it funny in the first place. Outlines survive. Quotes don't. And the quotes are usually the part you actually wanted.

The one thing that actually works

Here's the part memory research is genuinely useful for, rather than just depressing. The fix for fast decay isn't a better memory. It's retrieval.

A well-studied finding in cognitive psychology, often called the testing effect, is that actively recalling something and writing it down cements it far better than simply re-reading or replaying it in your head. The act of pulling a memory back out, even once, changes how durable it is. Applied to parenting, that means the two minutes you spend tonight typing "he called the moon a nightlight" doesn't just create a record. It's the single intervention most likely to keep that sentence retrievable in ten years, because you forced a retrieval and a write-down while it was still fresh, in the window before the curve does its worst damage.

The problem was never that parents don't care enough to remember. It's that remembering-on-purpose requires a prompt at the right moment, and most days don't hand you one. A milestone prompts itself. A Tuesday doesn't.

This is the whole design problem I was trying to solve when I built Everkid, and it's also the reason I wrote a full comparison of the other baby journal apps that are trying to solve it differently. Some rely on you remembering to open an app. Some ask a question over text. What the memory research above suggests is that the method matters less than the timing: something has to interrupt your evening, at the moment the day is still retrievable, and ask you to pull it back out before it's gone.

"I'll remember the important parts"

This is the objection I hear most, usually from parents who haven't tried keeping any kind of record yet, and I understand it because I believed it myself. The reasoning goes: sure, I'll lose the small stuff, but the moments that matter will stick.

The trouble is that you don't get to decide which moments matter until much later, and by then the forgetting curve has already done its work. The renamed household objects, the invented bath game, the word your kid used for the moon before they learned the real one: none of that feels significant on a Tuesday. It only becomes the thing you'd give anything to remember once your kid is twelve and talks like everyone else. Milestones announce themselves in advance. The moments you'll actually miss don't send a warning.

I also don't think this is really about willpower. Nobody sits down and decides not to remember their kid's childhood. What happens is that remembering-on-purpose competes with a full day, a hundred small decisions, and a phone that's much better at offering you something easier than a blank page and a blinking cursor. The parents who keep the best records aren't the ones with better memories or more discipline. They're the ones who found a prompt that shows up at the right moment and makes the two-minute version of remembering easier than skipping it.

Tonight is one of a small number of nights

Your kid gets 936 weekends between birth and eighteen. That's the whole math: eighteen years, fifty-two weekends a year. It sounds like a lot until you realize how many of the ordinary weeknights in between it are actually built from, and how few of those you'll be able to describe in specific detail a decade from now without help.

The forgetting curve isn't a reason to panic about the past. What's gone is gone, and there's no research that reverses that. It's a reason to treat tonight differently than you treated last Tuesday.