Neuroscience

Why You Forget What Your Matches Told You

July 18, 2026 · by Brendon Kennedy

In short: You forget what matches told you less because the memory fell out and more because it was never fully written down: ADHD memory difficulty concentrates at encoding, and a divided, app-fragmented attention span is most damaging at exactly that input stage.

Why do you forget what your matches told you?

Short answer: the detail probably never got fully written down. For people with ADHD, the long-term-memory difficulty sits largely at the encoding stage — the moment where an experience is turned into a durable trace — while storage and retrieval of episodic memory stay comparatively intact.² So it’s less “the memory fell out” and more “less got recorded.”

That distinction changes what you’d do about it. If memories were leaking out of storage, the fix would be review and reminders. If the problem is at input, the fix is protecting the moment of input — which is exactly the moment a dating app is engineered to interrupt.

I know this one from the inside. A match — Jane, on a recent date — tells me something that actually matters to her, and by the time I go to reply there’s a smooth blank where the detail should be, plus a small flush of social panic that I’m about to get caught not having held onto a person I genuinely wanted to hold onto. The wanting-to was never the problem. The holding-on was.

This isn’t a niche footnote, either. About 6.0% of U.S. adults — roughly 15.5 million people — had a current ADHD diagnosis in 2023, in the CDC’s first national adult estimate in nearly two decades.¹ A lot of people are running this exact hardware through a swipe interface.

Is ADHD forgetting a storage problem or an encoding problem?

It’s mostly an encoding problem. The meta-analytic picture is that ADHD is associated with reduced spontaneous, deep, and organized encoding — the brain does less of the elaborative work that makes a memory findable later — while the storage and retrieval machinery is comparatively intact.²

Think of it as the difference between a note you wrote badly and a note you lost. If you never wrote your date’s sister’s name in full — just half-heard it while glancing at a notification — no amount of later effort recovers it, because there’s nothing coherent to recover.

Does splitting your attention actually hurt memory?

Yes, and the timing is the whole story. In classic divided-attention experiments, splitting attention during encoding sharply reduces later recall and recognition. Splitting the same attention during retrieval barely touches accuracy — the cost there shows up as slower reaction time, not more errors.³

That’s a strong, specific claim: attention is most protective at input. A second task running while you’re trying to remember something is annoying but survivable. A second task running while you’re taking something in is where the damage lands.

A buzzing phone during a date is that second task, arriving at the worst possible stage. It’s not stealing the memory later — it’s preventing the memory from forming now.

What’s actually happening when a moment gets “written down”?

Two systems work as a loop. Prefrontal control biases what gets selectively encoded — what you attend to, essentially — and the hippocampus binds that moment’s context into a single retrievable trace.⁴ Attention decides what makes the cut; binding stitches the surviving details into something you can pull back later.

Retrieval then rides on that stitching. A cue works only insofar as it recreates features that were present at encoding — the principle of encoding specificity.⁵ If a detail was never bound into the trace in the first place, there is no cue in the world that will surface it. The reminder can only re-present what was recorded.

This is why “I know we talked about it” can be simultaneously true and useless. You were present for the conversation. You just weren’t fully encoding during it.

Why is meeting in person a richer place to remember someone?

Because in-person encoding is multichannel — face, voice, context, timing all arriving together — and semantically congruent multisensory encoding can beat single-channel encoding, with the benefit surviving even when the later test uses only one modality.⁶ More congruent channels at input means more features bound into the trace, and more possible cues later.

But hold the honesty line here. The multisensory boost is not automatic: in one study, adding faces and voices while learning biographical facts did not reliably improve recall.⁷ So the accurate framing is that in-person is a richer place to encode — more raw material for binding — not that in-person always wins. Presence gives your attention more to work with; it doesn’t guarantee your attention shows up.

So what does this mean for using dating apps?

Put the two facts together. Memory difficulty concentrates at encoding, and divided attention is most damaging at encoding.²³ A dating app hits both pressure points at once.

A text thread is close to the floor of the multichannel effect — one channel, stripped of face, voice, and shared context, so there’s very little for the hippocampus to bind.⁶ And the interface itself is a divided-attention generator: notifications, other matches, the pull to keep swiping. If you want the deeper reason the app fragments your attention in the first place, that’s the slot-machine design of the feed; even the platform’s own eight-conversation turn cap reads, through this lens, as a de facto attention cap on how many encoding streams you can actually keep straight.

None of this is a moral failing, and it’s not a memory that “fell out.” It’s a moment that was never given a quiet enough input channel to be recorded. For the lived, personal version of this — what it’s like to run an ADHD brain through these apps — there’s the longer essay on the dopamine loop. The mechanism, though, is the plain part: protect the encoding moment, and there’s actually something there to remember.

← All writing