Sources, worksheets and the full lesson text
Everything this lesson is built on, in one place: the research it cites, the worksheets that go with it, and the complete text if you’d rather read it in one uninterrupted piece.
Communication Across Neurotypes
A note before we start
If your child doesn’t answer to their name, won’t hold your eyes, takes what you say word for word, goes quiet exactly when you most want them to talk, or says something that lands sideways — it is tempting to read all of it as not listening, not caring, not trying. This module is about a different reading. Almost none of what looks like a communication failure is your child failing to communicate. Most of it is a translation gap — and translation gaps have two sides, which means there’s something you can actually do about your half.
The full lesson, in plain text
A. The double empathy problem
For decades, the story about autism and communication went one direction: the autistic person has a deficit, a broken part, a missing social skill, and everyone else is the fluent, correct standard they’re failing to reach. In 2012 an autistic researcher named Damian Milton turned that story inside out with an idea called the double empathy problem.1 His point was simple and, once you see it, hard to un-see: when an autistic person and a non-autistic person misunderstand each other, that misunderstanding is mutual. It’s not that one brain is broken and one is fine. It’s that two people are running different social operating systems, and neither one automatically reads the other.
Picture two people who each speak a different dialect of the same language. Your child’s dialect tends to be direct, literal, precise — it says what it means. The wider world’s dialect runs on hints, subtext, tone, and a mountain of unspoken rules where “that’s interesting” can secretly mean “I disagree” and “we should catch up sometime” can mean nothing at all. Each dialect is fully fluent. Each is a genuine language. But there’s a gap between them — and crucially, the gap runs both ways. Your child has to work to decode the world’s hidden meanings; the world has to work to hear your child’s directness as honesty rather than rudeness. A deficit lives inside one person. A translation gap lives in the space between two.
This isn’t just a nice reframe — it’s been tested. In one study, researchers passed a story down chains of people: some chains all autistic, some all non-autistic, some mixed. Information survived just as well in the all-autistic chains as in the all-non-autistic chains. Only the mixed chains lost detail and reported feeling more awkward.2 Autistic people communicate perfectly well with each other. The breakdown shows up specifically at the border between neurotypes. And the same research tradition finds that autistic people often feel most at ease, most understood, with others who share their wiring9 — which tells you the drive to connect was never missing. It just gets easier when someone meets it halfway.
Diagram: TWO DIALECTS
The double empathy problem: two people speaking different dialects of one language. Each is fluent; the gap between them runs both ways. It is a translation gap, not a one-sided deficit.
B. Eye contact, names, and processing time
Three of the most misread signals in a home deserve to be decoded one by one, because each one gets read as not caring when it’s really something else entirely.
Eye contact. We’re taught that looking someone in the eyes means respect and attention, so a child who won’t do it can feel like a child who won’t engage. But brain-imaging work tells a different story: when autistic people are made to focus on the eye region, the subcortical threat-detection system — including the amygdala — overactivates.3 Eye contact for many autistic brains isn’t neutral or warm; it’s intense, even alarming, like staring into the sun. And here’s the part that matters most: forcing it doesn’t build the skill, it floods the system, so your child understands less of what you’re saying, not more. When your child looks at the floor while you talk, that’s often them making room in their brain to actually hear you. Looking away can be how they stay in the conversation.
“They don’t respond to their name.” This one frightens parents, and it gets read as ignoring or as hearing loss. Reduced response to name is real and early — it’s one of the differences that shows up by around a child’s first birthday4 — but “doesn’t respond” is not the same as “doesn’t hear” or “doesn’t care.” Often there’s auditory processing in the way (in a noisy room, words can arrive as gibberish), or the child is locked deep in focus and pulling out costs more than you’d guess, or they’re in a verbal shutdown and simply can’t answer yet. The name landed. The reply is what’s expensive.
Processing time. You ask a question; nothing comes back; you assume the answer is no, or that they’re ignoring you, so you repeat, rephrase, or answer for them — right as their reply was loading. Think of a film with subtitles that lag a beat behind the audio. The line was spoken the instant you spoke; the subtitle — your child’s decoded, composed response — renders a moment later. Autistic thinking often leans on slower, more deliberate, analytical processing rather than a fast automatic reply,2 and that costs seconds. The silence isn’t nothing. It’s the subtitle still loading. Rush it and you overwrite the answer; hold it and the full reply arrives.
Diagram: SUBTITLE LAG
Processing time is subtitle lag: the reply is still rendering after you speak. The silence isn’t nothing — it’s loading. Give the extra beats before you repeat, rephrase, or answer for them.
C. Literal thinking, and when speech goes quiet
Literal and precise thinking. “I’ll be there in a second.” “Go wash up” — and your child washes only their hands, because that’s what the words said. Literal interpretation gets read as difficult or pedantic, but there’s a gentler explanation: for many autistic minds, the brain trusts the actual words more than it trusts a guess about your hidden meaning.8 When language leans on prediction — sarcasm, hints, figures of speech, “you know what I meant” — the safest, most certain reading is the literal one. This isn’t your child being obtuse. It’s your child taking you at your word, which is a reasonable thing to do with words. The fix isn’t to demand they read between the lines; it’s for you to say the line.
Sometimes-speaking and situational mutism. One of the most confusing things for parents is a child who talks fluently on Tuesday and can’t get a word out on Wednesday. It looks like choice, or manipulation, or regression. It is almost always none of those. For many autistic people, access to speech isn’t fixed — it depends on energy, regulation, sensory load, and stress, and it’s often the first thing to drop when the nervous system is overloaded. And situational mutism — going silent in specific settings — overlaps heavily with autism; in one study of children with a selective-mutism diagnosis, well over half turned out to be autistic.5 The community increasingly prefers “situational” over “selective,” precisely because “selective” implies a choice that usually isn’t there. Your child isn’t withholding words to be difficult. The words are temporarily out of reach.
Diagram: THE TAUT LINE
Connection is a tin-can telephone: the line carries only when the string is taut from both ends. Forcing eye contact or demanding speech yanks it slack — and the signal drops. Looking away can be how they keep the line open.
D. Speech is not the same as communication
Here is the reframe underneath all of it: speech is just one form of communication, and not always the most accessible one. We tend to treat spoken words as the real thing and everything else as a lesser substitute — so a child who types, signs, uses a letter board, gestures, or speaks through a device can feel, to a worried parent, like a child who hasn’t “really” communicated. That hierarchy is the mistake. All of these are the same message in a different envelope. “I need help,” sent by voice, by note, or by AAC, is the identical letter inside; only the format on the outside changed.
The fear that hands a child a device or a letter board might “keep them from talking” is one of the most understandable worries in this whole module — and one of the most thoroughly answered. Across the research, AAC does not slow or prevent speech; in review after review, spoken language holds steady or actually increases when a child is given another reliable way to communicate.6 And aided AAC produces large, real gains in communication, with knock-on benefits for social connection and fewer meltdowns born of not being understood.7 Giving your child more envelopes doesn’t shrink their voice. It gives the message more ways out.
There’s a quiet grief that can sit under this for parents — a picture of easy back-and-forth conversation that doesn’t arrive the way you imagined. That grief is real and it’s allowed. But it helps to notice what you might be trading it for: a child who is understood in the language they actually have, instead of one who’s constantly failing a test written in a dialect that costs them everything to speak.
Diagram: A DIFFERENT ENVELOPE
Speech, writing, and AAC are different envelopes for the same message — the format changes, the letter inside does not. AAC doesn’t shrink a child’s voice; it gives the message more ways out.
E. What helps: be the interpreter, not the corrector
If the whole problem is a two-way translation gap, then the single most powerful thing you can do is step into the interpreter’s booth — carry your child’s meaning across to a world that doesn’t speak their dialect, and carry the world’s hidden meanings back to them in plain terms — instead of sitting in the corrector’s chair, red pen out, marking your child wrong for how they said it. A corrector rewrites the speaker. An interpreter carries the meaning intact. Every tool below is a version of that.
Diagram: THE INTERPRETER'S BOOTH
The parent’s job is interpreter, not corrector: carry your child’s meaning across without rewriting who they are. A corrector marks them wrong; an interpreter keeps the message whole.
1. Drop the eye-contact demand.
Stop “look at me when I’m talking to you.” Let your child listen the way their brain can — looking at the floor, a wall, a fidget. If you want a connection cue, offer a low-stakes one (“you can look at my shoulder”) rather than forcing the eyes, which floods the system and costs comprehension.3 You’re trading a performance of listening for the real thing.
2. Give processing time — then give more.
After you ask or ask for something, count silently to ten before you repeat, rephrase, or fill the silence. Let the subtitle load. If you must add something, add time (“no rush”), not more words — extra words reset the render. Silence is often the answer being built, not the absence of one.
3. Say the line. Don’t make them read it.
Swap hints for direct, literal language: not “it’s getting late” but “shoes on, we leave in five minutes.” Not “are you going to leave that there?” but “please put your plate in the sink.” Being clear isn’t being cold — for a brain that takes you at your word, an explicit request is a kindness, and it prevents the misfire you’d otherwise blame on them.
4. Honor sometimes-speaking — never force words.
When speech drops, don’t demand it back (“use your words” is the worst thing to say to a nervous system already at its limit). Offer another channel in the moment: a nod, a thumbs-up, a written note, a point to a picture, a yes/no card. Make it normal that some moments are low-word or no-word, and that this is allowed rather than a failure to fix.
5. Welcome every envelope, including AAC.
Treat typing, signing, gesturing, a letter board, or a speech device as full, valid communication — not a last resort and not a threat to speech (it isn’t; it tends to help).6 Respond to the message, not the medium. A child who is understood in any format melts down less than a child straining to produce the one format that costs them the most.
6. Translate both directions.
Be the booth. To your child, unpack the world’s hidden meanings in plain words (“when Grandma said ‘maybe later’ she meant no”). To the world — teachers, relatives, siblings — carry your child’s directness and silences across as what they are (“he’s not being rude, he answers honestly”; “she’s listening, she just isn’t looking”). You’re not fixing your child. You’re staffing the translation booth they deserve to have.
If your child also has ADHD, the “not listening” reading gets even more unfair.
ADHD stacks a second layer on top of the autistic translation gap. The blurted interruption isn’t rudeness — it’s a thought that will vanish if it isn’t said right now, because working memory won’t hold it. The child who talks over you, or drifts mid-sentence, or answers a question you asked two minutes ago, is running a brain where attention and impulse move faster than the social clock. And the interruptions cut both ways: an AuDHD child can be simultaneously the one who won’t answer (autistic processing lag) and the one who can’t wait (ADHD impulsivity), which looks like contradiction and is really just two systems with different tempos.
What helps is the same interpreter stance, plus a little mercy on timing. Let the blurt happen and circle back (“hold that — I want to hear it”). Give the processing pause AND forgive the impulsive jump. Externalize the thought before it’s lost — a quick note, a “parking lot” for ideas. You’re not teaching them to communicate “normally.” You’re building a shared rhythm that fits how their particular brain sends and receives.
Related modules in our other free courses
Videos in this module
Short, real clips from our own team — Brooke Tidwell of the Parenting Autism Therapy Center (lived experience) and Dr. Michelle Karth (research) — on sometimes-speaking, eye contact, name-response, literal thinking, and the double empathy problem.
Sometimes-Speaking Autism (Brooke Tidwell · Parenting Autism Therapy Center)
Sometimes-Speaking Autism: Not a Choice, but It Should Be (Brooke Tidwell · Parenting Autism Therapy Center)
Why Speech Is Variable in Autism (Brooke Tidwell · Parenting Autism Therapy Center)
Why I Don't Always Respond to My Name (Brooke Tidwell · Parenting Autism Therapy Center)
Autistic Speech Differences (Brooke Tidwell · Parenting Autism Therapy Center)
Let's Stop Asking “Why Don't Autistic People Communicate Normally?” (Brooke Tidwell · Parenting Autism Therapy Center)
The Double Empathy Problem (Dr. Michelle Karth)
Literal Thinking Is Prediction Uncertainty (Dr. Michelle Karth)
Why Autistic Brains Pause Before Answering (Dr. Michelle Karth)
We're Capable of Great Eye Contact — It Just Matters Who We're With (Dr. Michelle Karth)
The workbook, as text
Your answers save to this device only — we can't see a word of what you write. This module is about learning your child's dialect: name a signal you've been misreading, catch one hint you can make direct, and pick one non-spoken envelope to welcome instead of steering back to speech.
1. The signal you keep misreading
Which of your child's communication signals have you most often read as 'not caring' or 'not trying' — not answering to their name, no eye contact, going quiet, taking you literally? Name the one that stings most, and try on the other reading.
2. Your child's dialect
Describe how your child actually communicates when they're regulated and comfortable — direct? few words? through a special interest? by showing rather than telling? This is their fluent dialect, not a broken version of yours.
3. Where speech drops
When does your child's access to speech tend to disappear — which places, times of day, or states (tired, overwhelmed, transitioning)? What could you offer instead of words in exactly those moments?
4. A hint to make direct
Catch one thing you routinely say as a hint or a question that's really a request ('it's getting late', 'are you going to leave that there?'). Rewrite it as a clear, literal line.
5. The processing pause
Honestly — how long do you usually wait after asking before you repeat, rephrase, or answer for them? What would it take for you to hold a silent count of ten and let the subtitle load?
6. Every envelope counts
List the ways your child already communicates that aren't spoken words — gestures, typing, signs, showing you things, a device, a meme, a look. Which do you want to welcome more instead of steering back toward speech?
7. Translating to the world
Optional. Who in your child's life most needs your interpreting — a teacher, a grandparent, a sibling? What one sentence would help them read your child's directness or silence correctly?
Want to talk it through with someone who gets it?
Feeling like you and your child speak different languages? You're not wrong — and you're not stuck.
This course was built by the therapist-parents at the Parenting Autism Therapy Center — clinicians who help families learn each other's dialect, several of them raising neurodivergent kids of their own. If the daily misreads are wearing everyone down, reach out. A conversation costs nothing, there's no pressure, and saving this for later counts too.
Next in this course
Module 11 — Masking, and Letting Your Child Unmask at Home
The research behind this module
Every factual claim above traces to a source. Here they are, in full.
Milton (2012). .
Disability & Society 27(6):883-887. Damian Milton (an autistic researcher) reframed autistic social difficulty as a two-way breakdown: autistic and non-autistic people each struggle to read the other, because they run on different social operating systems. It relocates the 'deficit' from inside the autistic person to the mismatch between two neurotypes. A theoretical framework that has since gathered empirical support (see Crompton), not a single experiment — so we treat it as a clinical/theoretical lens.
Crompton, Ropar, Evans-Williams, Flynn & Fletcher-Watson (2020). .
Autism 24(7):1704-1712. A 'diffusion chain' study (n=72): a story was passed person to person down chains that were all-autistic, all-non-autistic, or mixed. Information survived just as well in autistic chains as in non-autistic chains; only the MIXED chains lost detail and rated rapport lower. Direct evidence that the breakdown lives in the cross-neurotype gap, not in an autistic skills deficit.
Hadjikhani, Åsberg Johnels, Zürcher, Lassalle, Guillon, Hippolyte, Billstedt, Gillberg & Gillberg (2017). .
Scientific Reports 7:3163. fMRI study: when autistic participants were made to focus on the eye region, the subcortical face-processing system (including the amygdala, the brain's threat detector) OVERactivated, most strongly for fearful faces but present across expressions. The authors warn that forcing eye contact in therapy 'may create a lot of anxiety.' Looking away can be regulation, not disrespect.
Nadig, Ozonoff, Young, Rozga, Sigman & Rogers (2007). .
Archives of Pediatrics & Adolescent Medicine 161(4):378-383. Prospective study of infants at high and low likelihood of autism: diminished response to name became clearest by 12 months in the high-likelihood group. It establishes reduced name-response as a real, early autistic difference in auditory/social orienting — evidence that it reflects processing, not indifference or defiance.
Steffenburg, Steffenburg, Gillberg & Billstedt (2018). .
Neuropsychiatric Disease and Treatment 14:1163-1169 (with a published corrigendum). Among 97 children carrying a selective-mutism diagnosis, 63% met criteria for autism — yet none had been identified as autistic beforehand. Supports understanding situational speechlessness as an anxiety-and-capacity phenomenon closely tied to autism, and the community's preference for 'situational' over 'selective' (it isn't a choice).
Millar, Light & Schlosser (2006). .
Journal of Speech, Language, and Hearing Research 49(2):248-264. Research review of high-quality studies (67 individuals, ages 2-60): augmentative and alternative communication did not slow or prevent spoken language; speech production actually increased in about 89% of participants, with no case of decrease. Directly refutes the fear that giving a child a device or letter board will 'keep them from talking.'
Ganz, Earles-Vollrath, Heath, Parker, Rispoli & Duran (2012). .
Journal of Autism and Developmental Disorders 42(1):60-74. Meta-analysis of 24 single-case studies: aided AAC (picture systems, speech-generating devices) had large positive effects on communication, with benefits also seen for social skills and reductions in challenging behavior. Evidence that a 'different envelope' for the message is a genuine, effective form of communication, not a consolation prize.
Vicente, Michel & Petrolini (2023). .
Review of Philosophy and Psychology (published online 2023). Proposes that autistic literalism arises from a different balance in predictive processing: when the brain trusts incoming words more than its own top-down guesses about hidden meaning, the safest reading is the literal one. A theoretical account with growing traction rather than a settled finding, so we flag it as emerging — useful for understanding why hints, sarcasm, and 'you know what I meant' can genuinely miss.
Crompton, Hallett, Ropar, Flynn & Fletcher-Watson (2020). .
Autism 24(6):1438-1448. Thematic analysis of autistic adults' relationships: many described social time with other autistic people as markedly easier and more comfortable than time with non-autistic people. Counters the 'autistic people don't want connection' myth — the drive to connect is there; the ease depends on whether communication is being met halfway.
Numbered (1, 2, 3…) = peer-reviewed studies, checked by independent experts before publication. Lettered (a, b, c…) = clinical models — established professional frameworks, not single studies.
How this guide was made. Written from peer-reviewed research, clinical frameworks used in practice, and lived neurodivergent experience. Each module is also reviewed for neuroaffirming language. Where the evidence is still emerging or contested, we say so. Some screeners are validated research instruments; others are in-house reflection tools we built to help you notice patterns.