Show the full module text
Module 4 — Reading Your Own History
A structured re-examination of a life you have already lived — and the document an assessor will actually ask you for.
Self-Identification · Part Two — The evidence
The big idea
Short on capacity today? The big idea: Module 3 said that both criterion sets require a history: symptoms in early childhood, in more than one setting, costing you something. Almost nobody arrives with that written down. What people arrive with is a feeling of recognition and a handful of scenes.
This module turns the scenes into a document — and the first thing it has to tell you is that your memory of your own childhood is not a reliable witness. That is true of everyone, including the people who will disagree with you.
Step 1 — The lesson
A. Why memory is a poor witness
Start with the study that should be in front of anyone doing this work.
Ninety-four children who met the full DSM-IV criteria for ADHD were assessed between the ages of 7 and 11, and then reassessed between 16 and 22. Because the childhood assessment had actually happened, the researchers could compare what the young people and their parents now recalled against what had been recorded at the time.
Both groups recalled badly. And the error had a direction: current symptoms improved recall accuracy. Participants who endorsed ADHD symptoms now were more likely to recall clinically significant childhood symptoms. The past was being read through the present.
That is not a quirk of one sample. In a sixteen-year prospective follow-up, 176 adults who had been diagnosed with ADHD as children — 85 per cent of the original cohort — were interviewed at a mean age of 25 by clinicians who did not know their childhood status, alongside 168 comparison adults. Seventy-eight per cent of the childhood-ADHD group and 11 per cent of the comparison group were classified as having had childhood ADHD on the basis of what they recalled. Adjusted for how common ADHD actually is in the population, the positive predictive value of a retrospective self-report was 0.27 — of all adults given the diagnosis on recall alone, roughly a quarter would be correctly identified.
Diagram — A · The past, through today's glass. In a sixteen-year prospective follow-up, 176 adults who had been diagnosed with ADHD as children — 85 per cent of the original cohort — were interviewed at a mean age of 25 by clinicians who did not know their childhood status, alongside 168 comparison adults. Seventy-eight per cent of the childhood-ADHD group and 11 per cent of the comparison group were classified as having had childhood ADHD on the basis of what they recalled. Adjusted for how common ADHD actually is in the population, the positive predictive value of a retrospective self-report was 0.27 — of all adults given the diagnosis on recall alone, roughly a quarter would be correctly identified, and roughly three quarters would not. An earlier study makes the mechanism plain, and it is the mechanism the drawing shows: ninety-four children who met the full DSM-IV criteria for ADHD were assessed between the ages of 7 and 11 and reassessed between 16 and 22, and current symptoms improved recall accuracy — the past was being read through the present.
The same problem exists on the autism side, and it is not confined to the person themselves. When parents were interviewed about their child’s early development at age 2 to 3 and again about six years later, agreement between the two accounts on onset pattern was low (kappa = 0.38), and about a quarter of the sample changed onset category between interviews. Same child, same life, different account.
Two things follow, pointing in opposite directions. The first is a caution: if you are reading your childhood while feeling strongly that you are autistic or ADHD, you are in exactly the condition that produced the biased recall in the first study. The second is a defence. When somebody says I don’t remember you being like that, that sentence carries the same evidentiary weight as yours. Poor recall does not run in only one direction.
The answer to bad memory is not better memory. It is anything that is not memory.
B. The documents that outrank memory
Anything written at the time, by someone who was not trying to settle this question, is worth more than an hour of remembering. Most people have more of it than they think.
School reports and old report cards. The richest source, because the comments box was written by an adult watching you against thirty comparators, with no stake in this. The grades matter less than the free text. “Bright but careless.” “Does not seem to hear instructions.” “Would rather work alone.” Look for the same phrase recurring across different teachers and different years — that is more than one setting, and persistence over time.
Work reviews, appraisals, disciplinary letters. The adult equivalent, often more specific. A review that praises your output and flags your emails, your timekeeping and your handling of one particular colleague is describing a pattern you have probably explained to yourself as bad luck four times.
Medical records. Good for what you were taken to someone about — sleep, eating, stomach aches before school, speech, coordination, referrals that went nowhere. Also good for the misfires: the anxiety diagnosis at nineteen, the antidepressant that did nothing.
Letters, diaries, emails, old messages. Best for internal state and for what your life cost. A diary entry from a bad month is not a memory; it is a record.
Photographs and home video. Narrow but sharp. Where you were standing in the group. Whether you are looking at the camera. What you are wearing, repeatedly.
Your own logistics. Bank statements, calendars, tenancy records. Six addresses in eight years, late fees every month for a decade — records of how a life actually ran, written by nobody trying to prove anything.
Order the paper trail before you sit down and think. Once you have thought hard about a question, your memory of it is contaminated by the thinking.
Diagram — B · The box is already behind you. The answer to bad memory is not better memory; it is anything that is not memory. Weakest of all is a memory formed after you have thought hard about the question, because thinking about a question contaminates your memory of it. Weak is your own memory of it now. Stronger is someone else's memory of you: in a sample of 200 adults diagnosed with ADHD in childhood and 121 comparison adults, parent report was more diagnostically sensitive than self-report, and the young adults tended to under-report their own current symptoms. Strongest is anything written at the time by someone who was not trying to settle this question: school reports and old report cards, where the free text matters more than the grades and the same phrase recurring across different teachers and different years is both more than one setting and persistence over time; work reviews, appraisals and disciplinary letters; medical records, including the misfires; letters, diaries, emails and old messages; photographs and home video; and your own logistics, such as bank statements, calendars and tenancy records. Order the paper trail before you sit down and think.
C. The people who were there
Informants — a parent, an older sibling, a friend of twenty years, an ex-partner — are the other route to the past, and they need handling with the same care.
What they are genuinely good for: specific external behaviour. What you did, how often, and what they had to do about it. “You wouldn’t wear anything with a seam.” “You had one friend and you saw her every day.” “I stopped inviting you to things because you never came.”
What they are not good for: your inner life, and whether you were struggling. From the outside, a child who is quietly coping and a child who is quietly drowning look identical.
They also have the recall problem — the parent-interview study above is a study of parents. But the informant literature complicates that usefully: in a sample of 200 adults diagnosed with ADHD in childhood and 121 comparison adults, parent report was more diagnostically sensitive than self-report, and the young adults tended to under-report their own current symptoms while comparison adults over-reported. Informants are not a lesser source. On some questions they are the better one.
Two practical points. Ask open questions before specific ones — “what was I like at seven?” before “was I always losing things?” — because a leading question gets you the answer you implied. And ask more than one person; where two accounts diverge, the gap is information rather than a problem to be resolved.
If there is nobody left to ask, or nobody safe to ask, that is common and not disqualifying. The paper carries the load instead.
Diagram — C · Both heads above the water. A parent, an older sibling, a friend of twenty years or an ex-partner is genuinely good for specific external behaviour: what you did, how often, and what they had to do about it. You wouldn't wear anything with a seam. You had one friend and you saw her every day. I stopped inviting you to things because you never came. In a sample of 200 adults diagnosed with ADHD in childhood and 121 comparison adults, parent report was more diagnostically sensitive than self-report. What informants are not good for is your inner life, and whether you were struggling: from the outside, a child who is quietly coping and a child who is quietly drowning look identical, which is why the two heads in the drawing are the same and everything that differs is under the line. They also have the recall problem — when parents were interviewed about their child's early development at age 2 to 3 and again about six years later, agreement between the two accounts on onset pattern was low, at kappa 0.38, and about a quarter of the sample changed onset category between interviews. Ask open questions before specific ones, because a leading question gets you the answer you implied. If there is nobody left to ask, or nobody safe to ask, that is common and not disqualifying.
D. A pass through the years
Work forwards, one stage at a time, and prefer questions whose answers you cannot predict.
Before school. What would you not eat? How did you sleep? What did you do when the room got loud? What did adults say about you when they thought it was a compliment?
Primary school. Where were you at breaktime, physically? Who came to your birthday? What did the comments box say? Were you ever moved seats, and why?
Secondary school. What happened to your marks between eleven and sixteen? What did you stop doing? Which subjects did you drop, and what was the stated reason?
Late teens and first jobs. How many jobs, and how did each one end? Which parts were hard in a way you never said out loud? What did you drink or take, and what did it fix?
Relationships and friendships. How do friendships end for you — a rupture, or a fade? What did partners say in the same words that other partners had used?
Parenthood, if it applies. What in your child’s development did you recognise instantly?
At each stage, write what you did — not what you were like. “Quiet” is a judgement. “Sat at the end of the row” is a fact.
E. A memory is not evidence, and the trap
This is the one thing in the module worth learning properly.
“I was shy” is a memory. “I ate lunch in the library every day for three years” is evidence.
The difference is not detail for its own sake. It is that the second sentence tells an assessor — and tells you — the four things a criterion set actually asks about: what you did, how often, for how long, and what it cost you.
Run your own sentences through it:
Memory. I was always disorganised. Evidence. I have lost three passports and my mother kept a spare key with the neighbour from when I was nine.
Memory. I was sensitive to noise. Evidence. I left my own leaving party after twenty minutes and sat in the car.
Memory. I struggled socially. Evidence. I have never once made a friend at work without them approaching me first.
Memory. School was hard. Evidence. I was moved out of top set in two subjects at thirteen, and my reports say “does not listen” in four different years.
Memory. I’m bad at jobs. Evidence. Seven jobs in twelve years; three ended after a conflict with one specific person.
Notice that the evidence column contains no diagnostic language at all. That is deliberate. The history’s job is to describe a life so exactly that someone else could draw the conclusion. If you write the conclusion in, you have supplied an opinion where a fact was wanted.
Diagram — D · A word stamped over the room. I was shy is a memory. I ate lunch in the library every day for three years is evidence. The difference is not detail for its own sake; it is that the second sentence tells an assessor, and tells you, the four things a criterion set actually asks about: what you did, how often, for how long, and what it cost you. I was always disorganised is a memory; I have lost three passports and my mother kept a spare key with the neighbour from when I was nine is evidence. I was sensitive to noise is a memory; I left my own leaving party after twenty minutes and sat in the car is evidence. School was hard is a memory; I was moved out of top set in two subjects at thirteen, and my reports say does not listen in four different years is evidence. Quiet is a judgement; sat at the end of the row is a fact. Notice that the evidence column contains no diagnostic language at all. If you write the conclusion in, you have supplied an opinion where a fact was wanted.
Now the trap.
If you go looking through forty years for evidence that you are autistic or ADHD, you will find it. Everybody’s childhood contains lonely lunchtimes and lost coats. The recall studies say the search is running with a thumb on the scale — you are recalling the past through a present that already has an answer in it.
The defence is not to doubt yourself into silence. It is to make the search symmetrical. Spend a deliberate half-hour hunting the counter-example, and write what you find under its own heading:
- The years it did not apply. What was going on between fourteen and seventeen, when things were fine?
- The settings where you were fluent. If you ran a club, or acted, or were the one everybody rang, that belongs in the document.
- The other explanations. A bereavement, a move, an illness, a chaotic house, undiagnosed hearing loss, anxiety. Module 3 called this the “not better explained by” requirement; this is where you meet it.
- The people who would say no. What would your sister say if she read this? Write her version down too.
A history containing its own counter-case is a stronger document than one that does not — to an assessor, and to you at three in the morning. A file with nothing in it that does not fit is not a record of a life. It is a case for the prosecution.
And the point of this work is not to pass. Self-identification and an in-depth assessment are two different things a person can do, and this document serves both. If you never go through an assessor’s door, you will still have the most accurate account of your own life you have ever had.
Diagram — E · What the magnet leaves. If you go looking through forty years for evidence that you are autistic or ADHD, you will find it: everybody's childhood contains lonely lunchtimes and lost coats, and the recall studies say the search is running with a thumb on the scale. The defence is not to doubt yourself into silence; it is to make the search symmetrical. Spend a deliberate half-hour hunting the counter-example and write what you find under its own heading, with four sub-headings: the years it did not apply; the settings where you were fluent; the other explanations, such as a bereavement, a move, an illness, a chaotic house, undiagnosed hearing loss or anxiety; and the people who would say no. A history containing its own counter-case is a stronger document than one that does not — to an assessor, and to you at three in the morning. A file with nothing in it that does not fit is not a record of a life; it is a case for the prosecution. Keep the finished version to two pages, dated, specific, with no diagnostic vocabulary and the counter-case included, and keep the long version underneath. And the point of this work is not to pass: if you never go through an assessor's door, you will still have the most accurate account of your own life you have ever had.
F. What helps
1. Build it in two columns.
Left column: the claim, in ordinary words. Right column: what makes you say that — with a date and where it came from. Anything with an empty right-hand column is a hypothesis, and should be marked as one rather than deleted.
2. Get the paper first, then think.
Request school records, order medical notes, open the box in the loft. Do this before the reflective work, because thinking about a question changes how you remember it. Some records take weeks to arrive; start today and write while you wait.
3. Ask your informants blind.
Open question first, specific question second, and do not tell them what you are looking for until afterwards. If they ask, “I’m putting together a history of my childhood” is true and does not lead them.
4. Write the counter-case at the bottom, in its own section.
Headed What does not fit. Fill it honestly. If you cannot fill it at all, you have not looked — and that is the finding, not the absence of one.
5. Keep it to two pages, and keep the long version underneath.
The two-page version is what gets read: dated, specific, no diagnostic vocabulary, counter-case included. The forty-page version is for you. Both are worth having, and only one of them is a document anyone else can use.
Up next
Module 5 — Where Screeners Fit