You are in a meeting, and the idea arrives whole. It feels like a small lightning strike: clean, obvious, yours. You say it out loud and watch faces brighten. Weeks later, you find it in a newsletter you skimmed last spring. Or a coworker quietly reminds you that they said almost exactly this in March.

The unsettling part is what you didn't feel. You did not feel like you were remembering. From the inside, it felt like inventing.

Psychologists have a name for this: cryptomnesia, from the Greek for "hidden memory." It is a memory that surfaces without its label. You retrieve the content, but the sense of having met it before never switches on. So the brain files it under "mine."

A memory with the tag torn off

The word dates to the early 1900s, when the Swiss psychologist Théodore Flournoy studied a medium known as Hélène Smith, who produced vivid accounts of life on Mars and in a past incarnation. Flournoy argued that much of what she delivered was not supernatural at all. It was forgotten material, absorbed earlier and returning in disguise.

The modern legal example is more famous. In 1976 a US court found that George Harrison had infringed the copyright of "He's So Fine" with his song "My Sweet Lord." The judge did not think Harrison had plotted anything. The ruling described the copying as subconscious: he had heard the earlier song, and it had resurfaced as inspiration.

Neither case proves the broader point, and you should be wary of anyone who says they do. But they show the shape of the experience, and the laboratory has since filled in the mechanism.

What the lab found

In a well-known set of studies from the late 1980s and early 1990s, Alan Brown and Dana Murphy, and later Elizabeth Marsh and Gordon Bower, ran a simple procedure. People took turns generating ideas in a group, such as examples of a category or solutions to a problem. Later, alone, they were asked to come up with new ideas that had not already been said, or to say which ideas were originally theirs.

A meaningful share of the time, people recycled ideas that others had generated and claimed them as their own. Not out of dishonesty. They believed it.

The pattern reveals what memory is actually doing. Remembering an event has two separable parts:

  • Content: the idea, the melody, the phrasing.
  • Source: where it came from, who said it, whether you made it up.

These are stored and retrieved somewhat independently, and the source is the more fragile half. If you want the related, better-known cousin of this, it is the same machinery that lets you know a fact while forgetting where you learned it. Cryptomnesia is the sharper version, because here the fact is not just source-less, it is claimed.

Why the feeling of "mine" is so easy to fake

The researchers describe this as a failure of source monitoring. When an idea comes to mind, you make a quick, mostly unconscious judgment about its origin. That judgment leans on cues. Does it feel familiar? Do I recall hearing it? Do I remember the effort of producing it?

That last cue is the trap. Ideas you generate yourself tend to come with a trace of effort and a sense of fluency. An idea retrieved from memory can also arrive fluently, especially if the original encounter was shallow or happened while your attention was elsewhere. Fluent plus no sense of familiarity looks, to your brain, just like invention.

A few conditions make it more likely:

  • Divided attention when you first encountered the idea. You absorbed it without registering that you did.
  • Time. The source detail fades faster than the content.
  • Taking turns or brainstorming, where you are busy preparing your own contribution and only half-listening.
  • Wanting to be original. The motivation to have a good idea quietly lowers your standard of proof.

Why this is more than a curiosity

If you write, design, code, teach, or make music, you will have this experience, probably repeatedly, and mostly without finding out. The costs are real in a few places. There is accidental plagiarism, with professional and legal consequences. There is relational damage: someone's idea absorbed into your own, and a colleague who notices before you do. And there is a quieter cost, which is epistemic. If your sense of originality is unreliable, so is your confidence in your own conclusions.

There is also a kinder reading. Cryptomnesia is evidence that you learn. Ideas you have taken in genuinely become part of how you think. The problem is only the label, and labels can be rebuilt.

Your next moves

You cannot make source memory perfect. You can add scaffolding where it is weakest.

  • Keep a one-line source log today. Open a note and, whenever something strikes you as useful (an article, a talk, a comment), write the idea plus one phrase of where it came from. Do it at the moment of encounter, since that is when the source is still available.
  • Before you present a "new" idea, run a thirty-second origin check. Ask: have I read, heard, or seen anything like this? Search two or three key phrases. If it feels too smooth, treat that smoothness as a prompt, not a proof.
  • Credit generously when unsure. "This builds on something I may have picked up from..." costs almost nothing, and protects you if your memory turns out to be bluffing.
  • Draft first, check later, and keep the date. Write down your idea with a timestamp before you research it. Then compare. Seeing how often your "original" thought has precedent will calibrate you better than any warning.
  • In group brainstorms, write down who said what as it happens. Even a rough note-to-self column of names makes later attribution far more accurate than trying to reconstruct it.

Retrieval, with the source attached

The deeper lesson is that remembering a thing and remembering the story of the thing are different skills. Most study habits train only the first. You review the fact and never practise the provenance.

This is one place where a flashcard habit can quietly help. A good card does not only ask the question. It can carry a field for where the answer came from: the book, the lecture, the person who taught it to you. Each time you review, you retrieve the content and its origin together, which keeps the two glued more tightly than passive exposure ever would.

That is how we think about Recall, our flashcard app. It uses modern FSRS spaced repetition to schedule each card just before you would forget it, works fully offline, and imports your existing Anki and Quizlet decks. Add a source line to the cards that matter and you are training the exact thing cryptomnesia exploits. If that sounds useful, you can look at it at https://recall.lumenlabs.works. Either way, start the source log tonight. The next time an idea feels like lightning, you will have somewhere to check where it actually came from.