A stack of purple-inked sheets, going down the rows
You are nineteen, and it is early. The radiator ticks. Somebody's coat is still on the back of a chair from yesterday. At the front of the room a tall man with a soft German accent is counting out sheets of paper, and the room is quieter than it has ever been, because everybody in it watched the same ninety seconds of sky yesterday morning.
The shuttle Challenger broke apart 73 seconds after launch on 28 January 1986, with seven people aboard, one of them a schoolteacher. Classrooms across the country were watching live. By that evening, nobody in the United States needed to be told what had happened.
The man handing out the paper is Ulric Neisser — everyone calls him Dick — who in 1967 wrote a book called Cognitive Psychology that more or less named the field, and who has since become interested in a stubborn question: does memory in a laboratory have anything to do with memory in a life? He has an idea that will take two and a half years to ripen. Working with his student Nicole Harsch, he asks 106 students, less than a day after the event, to write down exactly how they heard.
Not what happened. How you heard. Where you were. Who told you. What you were doing when they did. It takes four minutes. Then the sheets go into a drawer, and everyone forgets about them — which turns out to be the most important part of the experiment.
The questions on the sheet
- What time was it?
- Where were you?
- What were you doing?
- How did you hear about it?
- Who else was there?
- How did you feel about it?
- What did you do afterwards?
Neisser and Harsch's questionnaire asked about seven such details of the moment of hearing. Below, the same seven questions are yours — about anything you like.
The Sealed Sheet
Pick one ordinary-but-memorable thing from the last week or so — a piece of news, a phone call, a small moment you'd say you remember perfectly. Write it down the way those students did. Then seal it with today's date, and let it wait for you.
You can skip any part of this. Nothing here is scored, and nothing is sent anywhere.
The students' average confidence, two and a half years on, was 4.17 out of 5. Hold on to your number. We'll come back to it.
The same people. The same seven questions. Different answers.
Autumn 1988. Forty-four of the original 106 are still reachable, and they sit down to the same questions. Nobody has warned them what this is about. They answer easily, in detail, with the particular warmth people have for a memory they consider one of their own.
Neisser and Harsch score each new account against the sheet the same person wrote the morning after — location, activity, who told them, and so on — for a weighted attributes score out of 7.
The mean was 2.95 out of 7. Eleven of the forty-four scored zero — not one detail matching what they themselves had written. Twenty-two, half the group, scored two or below. And only about a quarter of them could even remember filling in the first questionnaire.
Now the part that makes this study famous. On a five-point scale, asked how sure they were that this memory was accurate, they averaged 4.17. Not hedging. Not "I think." Certain.
Wrong, and sure of it
44 students, autumn 1988, each compared with what they themselves wrote in January 1986. Both bars are shown as a percentage of their own scale.
Show the numbers
| Measure | Mean | Scale | % of scale |
|---|---|---|---|
| Weighted attributes score | 2.95 | 7 | 42% |
| Confidence in the memory | 4.17 | 5 | 83% |
| Scored zero of 7 | 11 | 44 | 25% |
The Scoring Desk
Here is what the drawer looked like when it was opened. Read the two accounts, then tick every detail that changed. The scoring is the easy part; the strange part is that both of these were written, in good faith, by the same person.
January 1986, the morning after
I was in my religion class when a girl came in and told us. She was upset. We were all just sitting there talking about it for the rest of the hour, and then I went to lunch with my roommate and we didn't talk about it much. I felt shocked but not really sad, more like it wasn't real.
Autumn 1988, two and a half years later
I was in my dorm room with my roommate, watching TV. It came on as a news flash and we both just sat there totally shocked. I remember the picture of the smoke trail. I called my parents afterwards. I was really upset — I remember crying about the teacher.
Honesty flag: these two accounts are ours, not the published protocols — we could not obtain the chapter's verbatim transcripts. They are written in the documented shape of the real ones, in which television frequently replaced a person as the source of the news. The numbers above are the published ones; these two paragraphs are an illustration, and are labelled as such.
"That is my writing. But I remember it the other way."
There was a third session. Neisser and Harsch brought some of the students back and put the January 1986 sheets in front of them — their own paper, their own pen, their own handwriting, in their own words.
You would expect the old sheet to win. It doesn't, really. People recognised the handwriting as theirs and could not explain the contradiction — and the 1988 memory went on feeling like the true one, because feeling true is not something a memory does by being true. It is something a memory does by being vivid, and by fitting.
That is the finding, and it is worth sitting with for a second before we do anything clever with it. The false memory was not a fog. It was sharper than the real one. It had weather in it, and furniture, and a roommate. It had a whole afternoon attached.
You will often see a participant quoted as saying "Yes, that's my handwriting, but I still remember it the other way." That line is widely repeated; we could not check it against the original chapter, so treat it as a paraphrase of the documented result rather than a verified quotation.
A baseball game in December
Neisser did not exempt himself. For most of his life he carried a clear, warm memory of the day Pearl Harbor was attacked — 7 December 1941, when he was thirteen. In his own words:
I recall sitting in the living room of our house … listening to a baseball game on the radio. The game was interrupted by an announcement of the attack, and I rushed upstairs to tell my mother.Ulric Neisser, on his own flashbulb memory
He believed it for decades. Then one day he noticed something.
The Impossible Detail
Here is his memory, drawn out. Four details, one of which cannot possibly be true. Tap the one that breaks.
Nobody broadcasts baseball in December. The season has been over for two months. What he had almost certainly been listening to was football — and somewhere between 1941 and the retelling, his mind swapped in the more national, more resonant sport, the quintessentially American thing to be doing on the day America was attacked. The memory did not preserve the day. It composed one.
And this is the hinge of the whole edition. If memory were a recording, the errors would look like noise: blurs, gaps, static. They don't. They look like edits — small improvements in the direction of a better story about a particular person. Which raises a question you can actually investigate, right now, in your own head.
Can you see yourself in it?
Three years before Challenger, Neisser and his colleague Georgia Nigro published a small, beautiful study with a question so simple it sounds like a party trick. When you remember something, where is the camera?
Sometimes you are behind your own eyes: you see the room, the other person's face, your own hands if you look down. Nigro and Neisser called that a field memory. And sometimes — often, for older memories — you see yourself, from a few feet away, like a character. That is an observer memory.
Stop there a moment. You were never standing a few feet away. No such picture was ever taken. If you can see yourself in a memory, then that image is not a recording of the event — it is a reconstruction, built afterwards, by you. And you can catch yourself doing it in about ninety seconds.
The Point-of-View Bench
Eight everyday situations. For each one, call up one particular occasion — not the general idea, one specific time. Then say where the camera was, and roughly how long ago it was. Skip any that don't bring anything up.
You can skip any part of this. Nothing here is scored, and nothing is sent anywhere.
Your camera, by how old the memory is
Nigro and Neisser predicted that older memories drift toward the observer view, because the further a memory gets from the event, the more of it has been rebuilt. Each dot below is one of your memories.
In the original study, only 9% of memories came back as "neither," which is the first useful result: the distinction is not an invention of the researchers, it is something people recognise immediately in their own heads. Remote memories were less vivid and more often seen from the outside. And situations loaded with self-awareness — giving a talk, being watched, being in trouble — were the ones most likely to be remembered from outside the body, as though the mind had quietly stepped back to get a look at you.
The Flip
Nigro and Neisser then did something you can do to yourself. They changed one word in the instructions. When people were asked to recall a scene focusing on their feelings, 64% of the memories came back in the field view and 35% in the observer view. When they were asked instead to focus on the objective circumstances, the pattern reversed. The camera, it turns out, is on a rail — and the rail is attention.
Pick one of your observer memories from the bench above. Close your eyes for fifteen seconds and go back in — but this time go for the feeling in your chest, your hands, your stomach, not the look of the room. Then say what happened to the camera.
The instruction that moves the camera
Nigro & Neisser (1983), feelings-focus condition, beside your own eight. Their objective-circumstances condition reversed the pattern; we have not reproduced those percentages here because we could not verify them.
Show the numbers
If it is rebuilt every time, who is doing the building?
This is the point where a certain kind of person gets gloomy — if my memories are edits, is anything mine? But look at it the other way round. A recording has an owner and no author. A reconstruction has an author, and the author is you, working with material you did not choose, in a room you are still standing in.
Nine hundred miles north of Emory, at Northwestern University, Dan McAdams spent those same decades sitting people down with a tape recorder for two hours and asking them to do something that sounds almost childish. It is the opening move of what he calls the Life Story Interview, and it goes like this:
Please begin by thinking about your life as if it were a book or novel. … Please describe very briefly what the main chapters in the book might be. … There are no right or wrong answers to my questions.Dan P. McAdams, The Life Story Interview — II, Foley Center for the Study of Lives, Northwestern University
People can always do it. That is the quietly astonishing part. Nobody has ever divided their life into chapters before, there is no correct number, and yet within a minute or two almost everyone has four or five, with titles, and knows where the joins go. McAdams' claim is that this is not a party game: that the story is a real layer of a person, sitting on top of traits and goals, and that it is where identity actually lives.
The Chapter Spine
Here is the spine of a book with your name on it. Drag the joins to make the chapters longer or shorter, name them, and set how each one felt. A chapter can be thin and enormous. A chapter can be sixteen years of nothing much.
You can skip any part of this. Nothing here is scored, and nothing is sent anywhere.
The line your chapters make
Your own tone ratings, in order, across the length of the book. This is your data and nobody else's; there is no good shape for it.
Three Scenes
After the chapters, McAdams asks for scenes — particular moments, with a time and a place and people in them. Three of his prompts are below, in close to his own words. Answer any, all or none; a sentence is plenty. If a prompt goes somewhere heavy, you are allowed to put it down and take an easier one.
You can skip any part of this. Nothing here is scored, and nothing is sent anywhere.
Check the early one: can you see yourself in it? Most people can. That scene has been rebuilt more times than any of the others, which is exactly why it is so smooth.
McAdams' team then read thousands of these scenes looking for shapes. Two kept coming up. A redemption sequence runs from bad to good: the thing was awful, and out of it came something. A contamination sequence runs the other way: it was wonderful, and then it was spoiled. In a 2001 study of midlife adults and students, people who told more redemption sequences tended to report higher well-being, and contamination sequences ran with lower well-being — and the highly generative adults, the ones pouring themselves into the next generation, told redemptive stories markedly more often.
Be careful with that, though, and we will be careful with it in a minute. It is a correlation, in two directions, in a particular country, in a particular decade. What it is not is an instruction to turn your worst year into a lesson by Friday.
The Next Sentence
Take one of your scenes and write the sentence that comes after it — the "and so" or the "and then". Not the moral. Just what came next, as far as you can honestly see from here.
All three are ordinary. McAdams' own data has plenty of level sentences and plenty of downward ones, told by people who are perfectly well. A story that goes down is not a broken story; sometimes it is simply an honest one, and sometimes the upward sentence just hasn't been written yet.
And if writing this one stirs something up: that is a normal thing for memory to do. Talking it through with somebody you trust is part of how people work things out — it is not a last resort, it is the usual method.
What you just told us, handed back
This panel does exactly one thing: it counts what you entered and reads it back in your own words. It cannot see you, it does not know you, and it is not allowed to tell you what kind of person you are. It will name one pattern. You get the last word on whether that pattern is real.
Six things that wobble
This subject attracts more confident nonsense than any other. Here is where our own story is shakier than it sounds.
How this idea arrived
Eight questions — and these ones are graded
Everything above this line was yours and unscored. Everything below is about the people, dates and findings in the story, and gets marked out of eight. That line is deliberate: what you remember about Neisser can be right or wrong; what you wrote about your own life cannot.
Take your page with you
Nothing you typed was saved to this device or sent over the network. If you want to keep it — and the Sealed Sheet is worth keeping until you've forgotten writing it — copy it out or print it now.