The thirteen-year-old still fills in part of the form

Ask someone how their memory is these days and you’ll get an answer almost immediately. It’s one of those questions people feel qualified to answer without stopping to think.

A study published this week suggests that answer may be carrying rather more than it looks like it’s carrying.

What they did

Researchers led by Gideon Neidenmark, a doctoral student at Örebro University in Sweden, went looking in an unusually long archive. The IDA-project is a Swedish longitudinal study that began in 1965, following a group of children who were around ten at the time. They were assessed again in 1968, at thirteen.

Those childhood visits gathered a lot. Standardised tests of verbal, spatial and reasoning ability. Official school grades. Information about the family’s socioeconomic circumstances. And questionnaires asking the children about their social wellbeing, and about anxiety in the school setting.

Then, in 2022, the researchers sent follow-up questionnaires to the surviving participants, now around sixty-seven. Six hundred and twenty-five people completed that phase. They filled in a standard survey about subjective memory — how often they forget where they put something, lose the thread of a conversation, can’t retrieve a name they definitely know — alongside a widely used measure of their current anxiety and depressive symptoms.

The question was simple. Of all those childhood things, which ones, if any, still show up fifty-odd years later in how a person rates their own memory?

What they found

Mostly, none of them did.

Childhood cognitive ability didn’t predict it. School grades didn’t. Social wellbeing at ten or thirteen didn’t. Family socioeconomic status didn’t. All the things you might reasonably expect to leave a mark on how sharp someone feels at sixty-seven left, in this dataset, no mark the researchers could detect.

One thing did. Participants who had reported higher school anxiety at age thirteen tended to report worse subjective memory at sixty-seven. And the link held even after the researchers accounted for how anxious or low those participants were feeling now — so it isn’t simply that anxious teenagers became anxious sixty-seven-year-olds who rate everything darkly.

The age was oddly specific, too. School anxiety at ten predicted nothing. Thirteen did.

Neidenmark’s own reading of it: subjective memory “may better reflect psychological factors than actual memory performance.”

That sentence is more or less the whole study. And it sits on top of something already known — a 2014 meta-analysis found that people’s self-rated memory problems match their actual memory test scores only weakly.

The caveats are substantial, and the researchers state them plainly. This is an association, not a cause; Neidenmark says explicitly that we should be cautious about reading lower psychological health as causing anything here. The childhood questionnaires on school anxiety and social wellbeing used only a few items each, which made them statistically imprecise — and that imprecision may be part of why some of the null results came out null. The 2022 follow-up collected no objective memory tests, so there is no way to tell whether these participants’ memories had genuinely changed or whether they were simply judging themselves more harshly. And across fifty-seven years a lot of people leave a study: those who stayed had better childhood grades and better psychological health than those who didn’t.

So it’s a real finding, and it should be held loosely.

What we keep thinking about

Here’s the part that stayed with us, and it’s slightly uncomfortable, because it points straight at what we do.

A self-rating is never a clean measurement of the thing. It’s a measurement of the thing and the instrument — and the instrument is a person, with a history, having a particular kind of week.

Most of us treat our own assessments as though they arrive from nowhere. My memory’s getting worse. I’m not coping well. I used to be better at this. These feel like observations. They have the texture of noticing. But this study is a reminder that the person doing the rating showed up with a whole life already in progress, and that some of what registers as an observation about today may be a habit of self-assessment that was set down a long time ago and never revisited.

That’s not a reason to distrust yourself. It’s a reason to be a bit gentler about the readings — and, more usefully, to be more interested in the pattern than in any single one of them.

It’s also the honest reason Nagi’s check-in is four taps rather than a test. Mood, sleep, energy — noted as the days go by, not scored, not ranked, no streak to protect, nothing to be good at. We’re a self-report tool, so this research applies to us as much as to anyone: a single tap on a Tuesday isn’t a verdict on you, and shouldn’t be dressed up as one. What it can do is sit next to the taps either side of it. The S8 wellbeing measure Nagi uses is currently going through a formal validation study, because we’d rather find out how well it works than assert it. And all of it is designed to stay with you — no name, no email, an anonymous ID rather than a name. A calm space that’s just yours.

None of which is a claim that noticing fixes anything. It doesn’t, and we won’t pretend otherwise.

The bit we’d underline

The finding people will reach for is the anxious thirteen-year-old. We’d reach for the other half — the enormous list of things that didn’t predict anything. Not how clever you were. Not your grades. Not what your family had.

Which means whatever is happening in that self-rating, it isn’t a report card that’s been quietly running since school.

And if the way you assess yourself was in some part written when you were thirteen, that’s worth knowing mostly for one reason: you weren’t the one who chose the wording. You get to look at it now, as an adult, and decide whether it’s still fair.

If you want a private place to start noticing your own patterns, Nagi is on Google Play and the App Store — or in your browser at app.nagiandme.app. The first 15 messages are free. No name, no email. Just you, checking in.

— The Nagi team

---

Source: A six-decade study links teenage school anxiety to late-life memory complaints, PsyPost, 9 September 2026, by Eric W. Dolan. Underlying study: Neidenmark, G., Tuvblad, C., & Persson, J., “Early life predictors of subjective cognitive function in older age: Emerging longitudinal associations over six decades”, Journal of Gerontology: Psychological Sciences. Design: longitudinal cohort study drawing on the Swedish IDA-project, begun in 1965 with participants aged around 10, reassessed in 1968 at age 13, and followed up by questionnaire in 2022 at approximately age 67; 625 participants completed the final phase. Childhood measures included standardised verbal, spatial and reasoning tests, official school grades, family socioeconomic status, and self-report questionnaires on social wellbeing and school anxiety; the 2022 phase used a standardised subjective memory survey plus a widely used measure of current anxiety and depressive symptoms. Findings: higher self-reported school anxiety at age 13 was associated with poorer subjective memory at age 67, and the association remained after accounting for current anxiety and depressive symptoms; school anxiety at age 10 showed no such association; childhood cognitive ability, school grades, social wellbeing and family socioeconomic status at ages 10 and 13 showed no association with subjective memory at 67. Stated limitations: the study is observational and the lead author cautions explicitly against inferring causality; the childhood school anxiety and social wellbeing questionnaires used few items and had relatively low statistical reliability, which may account for some null results; the 2022 follow-up included no objective memory testing or clinical screening, so it cannot distinguish genuine cognitive change from more pessimistic self-assessment; and attrition across 57 years left a remaining sample with better childhood grades and better childhood psychological health than those who dropped out. Additional context cited by PsyPost: a 2014 meta-analysis finding that self-rated memory problems correspond only weakly with objective memory test performance.

Nagi is a mental health companion, not a medical device, and nothing here is medical advice. If you’re struggling, please reach out to a professional or someone you trust.