Show the full module text
Module 15 — Where You Misread Each Other
The session where the two of you find the places you get each other wrong, in both directions, and write down what was true.
Every couple misreads each other. A neurodiverse couple misreads each other in specific, repeatable places, and in both directions. This module walks through the session where the two of you find those places and write them down, with what was actually true beside each one.
What this is about
A misread is a wrong guess about what was going on inside your partner.
What was read, and what was true. “He is angry with me.” He was overloaded.
“She thinks I am failing at this.” She was deciding whether to worry.
It is not a wrong opinion about the dishwasher, and it is not a difference in what the two of you want. It is a reading that missed, and one true sentence could have corrected it.
Both of you do it. And both of you have some bridging to do, some work to reach the other. Most couples arrive with only half of that accepted. The autistic partner, they tell us, misses things. Almost nobody arrives with a list of the times the other partner missed things. In the story most of us grew up inside, the non-autistic partner is simply assumed to be good at reading people.
You have already collected most of the material for this session without meaning to. The noticing log had a line for the story you told yourself about your partner. Read those entries back a week later, and a surprising number were wrong. The Trait Wheel numbers your partner disagreed with last session belong here too.
At the restaurant she read his silence as anger and he read her question as a verdict, and only her misread is the kind a couple arrives expecting to find.
What happens in the session
The session runs on the shared screen. There are two columns, her reads of him and his reads of her, and six ordinary moments from the last month. Not the worst six. Small ones, easy to look at without the room heating up. Most come from the noticing log, and the disputed wheel numbers supply the rest.
This is the page the session produces: three misreads in each direction, with what was actually true written under every line.
The Misread Session
1 Three each, both directions. Each of you brings three moments when you now think you misread the other, and three when you felt misread. Both columns get filled before anything is discussed.
2 The reversal. The reader says what they read. The one who was read says what was true, using the lens words from the last two sessions. The read is stated, not defended.
3 The sort. Was the guess about the inside of the other person wrong? Then it is a misread, and it ends with “now I know.” Was the guess right, and you want different things? Then it is a disagreement, and it goes on a separate list for later.
4 The tell. For each misread, the visible signal that would have told the reader the truth. “When he stops using my name at a restaurant, it is the noise.” It goes beside the line, and it is what the reader looks for next time.
The person who was read is the only authority on what was true. The reader’s job is to hear it, not to grade it.
The reversal. Say what you read. Just the read. Now you: what was actually true, in the words we have been using. Sensory, energy, direct, mask, time, memory, attention, focus.
When the reader defends the read. I am not asking whether it was reasonable. It was reasonable. I am asking whether it was right, and she has just told you it was not.
Offering the screener. This one is for both of you, and it is about naming feelings, not about autism. Take it if you want to, and we will read the items you stopped on, not the score.
Almost every misread on the list was reasonable. Silence does mean anger in most of the rooms we learned to read in, and a person who leaves a party at nine usually did want to leave. That is exactly the problem.
A reasonable read never gets checked, because it never feels like a guess.
The reversal makes it feel like one, once, with the person it was about supplying the correction.
The way this goes wrong. The couple arrives with one column complete, dated and specific: every time the autistic partner missed a cue this month. The other column is empty, because “I read people fine.” The session cannot run on one column. So the partner with the empty column opens the noticing log, reads out the last three stories they told themselves about their partner, and checks each against what the partner says was true. Most couples find two of the three were wrong, and the second column starts.
“He said he was fine.” That line belongs on almost every list. The mask exists to produce that exact read. Here it gets a tell: the slightly too polite voice, the “it is fine” with the eyes on the floor, whatever the signal is in your house. A partner who knows the tell is a partner the mask can come off in front of.
The sort is the step couples skip. Some of the six were never misreads. The guess was right, and the two of you simply want different things. Nothing in this session fixes that, and later modules take it up one subject at a time.
A misread is fixed here, by information; a disagreement is two accurate reads and two different wants, and it is negotiated later.
Why we do it this way
The two columns are there because the misreading runs both ways.
That is what thousands of neurodiverse couples have told us, once someone asked about the second direction. Research supports it. In one study, non-autistic adults watched autistic and non-autistic people react to the same events. They read the autistic people wrong more often, while rating them as just as expressive.1
In a second study, non-autistic adults were worse at reading short animations that autistic adults had made. The autistic adults read both kinds equally well.2 Autistic people react differently, not less. The other direction is a misread too, and it usually goes unchecked.
The lenses are a checklist of misreads. Each of the eight is a place where a visible behavior reliably gets the wrong inner state attached to it. They run mostly in one direction, because they were written to explain the neurodivergent partner. The other direction has its own regulars: the non-autistic partner’s feelings travel by hint, sigh and change of plan, and the autistic partner reads a verdict where there was only a mood.
Each of the eight lenses pairs what a moment usually gets read as with what is usually going on underneath, so a couple that knows them can write much of its own list before the session.
The screener goes to both of you. As the module on your childhood wound said, about half of autistic adults have real difficulty identifying and describing what they feel, and some non-autistic adults do too. In a marriage that shows up as two misreads at once.
“I do not know,” given honestly, gets read as a refusal to say.
A partner who cannot find the word for a feeling gets read as not having it.
Research separates trouble naming feelings from autism itself. How strongly people respond to another person’s pain tracks how hard they find it to name feelings, not whether they are autistic.3 And naming a feeling is a different skill from working out what somebody intends.4 The couple that knows which skill missed argues about the right thing.
The screener is the TAS-20. It is twenty questions about naming feelings, in three parts, with decades of independent checking behind it. Each of you takes it only if you want to, and your therapist reads back the items that made you stop, not the score.
It decides nothing. It is a mirror, not a diagnosis.
When outside observers watched autistic and non-autistic people in conversation, they judged the autistic people more harshly than the people actually talking with them did.5 The person across the table is in a better position to read you than anyone watching from outside, provided they are reading you and not the story about you.
After the session
Both columns go into the shared folder the two of you are building, beside the wheel and the drawing from earlier sessions. The next module, The Working Map, Not the Label, builds the map out of the three. The disagreements you sorted out go with you into the later modules on the hard subjects, where they become agreements small enough to keep.
Expect the tells to start working before anybody has practiced anything.
Knowing that the flat voice at the restaurant is the noise, and not you, changes what you do next. The workbook below is the list in writing: the three you misread, the three where you were misread, the sort, the tells, and whether either of you took the screener.
The one thing, if that is all you have. If six is too many this week, bring one: one moment when you now think you read your partner wrong, with what was true.
Free reading, elsewhere. The screener, written up in full and free:
TAS-20: Alexithymia, When You Can't Name What You Feel — its three parts, and which of them to trust
New Path Family is a separate nonprofit. Its courses are free, carry no advertising, and it does not take a fee from us or from you.
Your workbook
Your answers save to this device only - we cannot see a word of what you write. This one is the misread list, in writing, and it has a column for each of you.
The misread list
From the last month. Small moments, not the worst ones.
Three moments when you now think you read your partner wrong: what you read, and what your partner says was true
Three moments when you felt read wrong: what your partner read, and what was actually going on in you
The sort and the tells
A misread ends with now I know. A disagreement is two accurate reads and different wants. A tell is something visible.
Of your six, how many turned out to be disagreements rather than misreads? — None, One, Two or three, Most of them
Which of the eight lenses (sensory, energy, direct, mask, time, memory, attention, focus) came up most, in each direction?
For each misread: the tell, and who looks for it next time
The screener
Offered to both of you. Read for the items, not the score.
The TAS-20: — I took it, My partner took it, We both did, Not this week
Where this comes from
Every source below was checked against the published record. They are grouped by the kind of evidence they are, so the numbers may not run straight down the page — a number points back to where the source is used in the lesson.
Research discussion
The misread list, the reversal, the sort into misread and disagreement, and the tell are practice moves. They come from our own sessions and from one line in the guide our therapists work from: identify the specific places each partner misreads the other. They were developed in use and have not been validated in a trial. In particular, nobody has tested whether changing the way you read your partner (reattribution) changes what you do next.
The double empathy idea the session rests on is still argued over in the literature. The session does not need the theory settled; it needs the couple’s own list. Five studies and one questionnaire paper sit behind the module, and none of the five is a study of couples. Two show the non-autistic direction of the misread on its own.
The study
Reading an autistic person’s reaction, and getting it wrong1
The study
Who is mind-blind toward whom2
Both are small laboratory studies with no partners in them. They establish that the misread has a non-autistic direction, not how often it happens at home.
The scanner study3 is thirty-six men, half autistic, matched for alexithymia on the TAS-20;6 the empathic signal in the brain tracked alexithymia in both groups and did not differ by group once alexithymia was accounted for. The eyes-test study4 matched autistic and non-autistic adults for alexithymia and gave both a face-based emotion task and a film-based intention task; alexithymia predicted the first and autism the second. Both samples are small and drawn from research volunteers.
The observer study5 is 102 non-autistic students rating videos of 42 autistic and 44 non-autistic men in conversation. The people in the videos are all men, and the conversations were first meetings, not marriages. It is cited for one comparison: the participants’ own conversation partners rated the autistic participants more favorably than the outside observers did.
Who this research was done with. The studies behind this module, none of them of couples, drew heavily on white, comparatively well-off, English-speaking participants. If your household carries pressures those samples did not — money, immigration, racism, disability, unsafe housing — the practice still applies, but the room you are practicing in is harder. That is the room, not you.
Peer-reviewed research
1. Sheppard E, Pillai D, Wong GTL, Ropar D, Mitchell P (2016) How easy is it to read the minds of people with autism spectrum disorder? Journal of Autism and Developmental Disorders, 46(4), 1247-1254. https://doi.org/10.1007/s10803-015-2662-8 Autistic and neurotypical targets were video-recorded reacting to four events; neurotypical participants then had to identify which event each target had experienced. Participants were more successful for neurotypical than autistic targets, rated autistic targets as equally expressive for three of the four events, and gave different verbal descriptions of the two groups' reactions. The authors conclude that autistic people react differently but not less expressively, and that neurotypical people are ineffective at interpreting the difference. Limitation: a small laboratory study of staged events and strangers, not partners.
2. Edey R, Cook J, Brewer R, Johnson MH, Bird G, Press C (2016) Interaction takes two: Typical adults exhibit mind-blindness towards those with autism spectrum disorder. Journal of Abnormal Psychology, 125(7), 879-885. https://doi.org/10.1037/abn0000199 Autistic and typical adults each produced triangle animations depicting mental-state interactions; the autistic participants' animations showed increased jerk on kinematic analysis. Typical adults identified the mental state more accurately in animations made by typical than by autistic participants, while autistic participants performed comparably on both. Limitation: a laboratory task with animated shapes; group sizes are not stated in the abstract.
3. Bird G, Silani G, Brindley R, White S, Frith U, Singer T (2010) Empathic brain responses in insula are modulated by levels of alexithymia but not autism. Brain, 133(5), 1515-1525. https://doi.org/10.1093/brain/awq060 Functional MRI study of 18 men with autism spectrum conditions and 18 controls matched for alexithymia (TAS-20), age and IQ, using an empathy-for-pain paradigm in a real social setting that did not rely on facial cues. Empathic responses in left anterior insula predicted the degree of alexithymia in both groups and did not vary by group; there was no difference in empathy between groups after accounting for alexithymia. Limitation: thirty-six male volunteers; a single task.
4. Oakley BFM, Brewer R, Bird G, Catmur C (2016) Theory of mind is not theory of emotion: A cautionary note on the Reading the Mind in the Eyes Test. Journal of Abnormal Psychology, 125(6), 818-823. https://doi.org/10.1037/abn0000182 Autistic participants and alexithymia-matched controls did not differ on the Reading the Mind in the Eyes Test, while the autistic group was impaired on an alternative theory-of-mind measure, the Movie for the Assessment of Social Cognition. Alexithymia, but not autism diagnosis, influenced Eyes Test performance and did not affect the film task. The authors conclude the Eyes Test measures emotion recognition rather than theory of mind. Limitation: group sizes are not stated in the abstract; the claim in this module is restricted to the Eyes Test.
5. Jones DR, Botha M, Ackerman RA, King K, Sasson NJ (2024) Non-autistic observers both detect and demonstrate the double empathy problem when evaluating interactions between autistic and non-autistic adults. Autism, 28(8). https://doi.org/10.1177/13623613231219743 102 non-autistic adults rated the interaction quality and traits of 42 autistic and 44 non-autistic men interacting in same- or mixed-neurotype pairs. Observers rated mixed interactions as the least successful, evaluated autistic participants more negatively than their conversation partners had, underestimated the trust and intelligence ratings the autistic participants made, and reported less social interest in the participants than the participants reported in each other. Limitation: student observers, male participants only, first meetings between strangers.
6. Bagby RM, Parker JDA, Taylor GJ (1994) The twenty-item Toronto Alexithymia Scale - I. Item selection and cross-validation of the factor structure. Journal of Psychosomatic Research, 38(1), 23-32. https://doi.org/10.1016/0022-3999(94)90005-1 The paper that introduced the twenty-item TAS-20 and its three-factor structure: difficulty identifying feelings, difficulty describing feelings, and externally oriented thinking. Cited as the source of the screener this module offers; the New Path Family module linked in the lesson discusses the reliability of the three parts. Limitation: a scale-development paper in non-autistic samples; cited for the origin of the screener only.
Both columns, filled in the room
The Neurodiverse Couples Counseling Center works with couples where one or both partners are autistic, ADHD or AuDHD. The misread session puts six ordinary moments on the screen and asks each of you what you read and what was true, in both directions, with a therapist who does not let one column stay empty. Therapy for clients in California, coaching worldwide, all by telehealth. A first conversation costs nothing.
Up next
Module 16 — The Working Map, Not the Label