The Double Empathy Problem: Autism Isn't a Deficit

The Short Version

You're not bad at communicating. You're fluent in a style that doesn't match the style most people around you were raised on, and the mismatch reads as your failure when it's actually a two-way gap.

That's the double empathy problem, proposed by autistic researcher Dr. Damian Milton in 2012. He argued that the communication breakdown between autistic and non-autistic people runs in both directions, not just from the autistic side (Milton, 2012).

Where the 'Autism Equals No Empathy' Idea Came From

For decades, autism was described using theory of mind research: the idea that autistic people struggle to model what other people are thinking or feeling.

That research measured autistic people against neurotypical social norms and called the gap a deficit. It never asked whether neurotypical people could read autistic communication just as poorly, which turns out to be the missing half of the story.

The damage from that framing runs deep. It's the reason 'autistic people lack empathy' became a public assumption instead of a myth.

What the Actual Study Found

In 2020, researchers ran a real test of this. They passed a story down a chain of eight people, like a game of telephone, and compared three kinds of chains: all autistic, all non-autistic, and mixed.

Autistic-only chains held onto detail just as well as non-autistic-only chains. Mixed chains lost detail significantly faster, and people in mixed chains rated their rapport with each other lower too (Crompton et al., 2020).

Read that again: autistic people talking to autistic people communicated fine. The breakdown only showed up when neurotypes mixed. That's not what a one-sided deficit would predict.

Double Empathy in Numbers

So What Actually Happens in a Mixed Conversation

Both people are reading a different rulebook and assuming the other person is using theirs. You pause to process fully before responding. They read the pause as disinterest. They soften a request with three qualifiers to be polite. You take the qualifiers literally and miss that it was actually urgent.

Neither person is wrong about their own rules. The friction comes from nobody naming that two rulebooks are in play at all.

Why 'Autistic People Lack Empathy' Is Backwards

Plenty of autistic people report feeling other people's emotions intensely, sometimes overwhelmingly so, to the point of needing to leave a room because someone else's distress became unbearable to sit with.

What's harder isn't the feeling. It's the specific, unwritten social encoding neurotypical culture uses to display it: a particular tone shift, a facial micro-expression, a pause of exactly the right length. Missing that encoding gets read as not caring. It's closer to being fluent in a different empathy dialect.

What to Say When Someone Assumes You Don't Care

'I care, I just don't show it the way you're expecting. Tell me directly what you need from me and I'll do it.' That sentence redirects the conversation from a character judgment to a solvable logistics problem.

In a relationship, agreeing on explicit signals ahead of time, 'when I go quiet during a hard conversation, I'm processing, not shutting you out', prevents the same misread from repeating every time.

What This Means for How You Explain Yourself

Stop apologizing for a communication style that works fine with the right audience. Start being specific instead: 'I process better in writing than out loud, can you send that as an email' gets you accommodated. 'Sorry, I'm bad at this' just reinforces the deficit story.

The same goes for friendships and work. Autistic-to-autistic communication in online communities and workplaces built around neurodivergent norms tends to run smoothly, which lines up exactly with what Crompton's study found.

Where This Theory Came From

Milton is autistic himself, and he built the theory directly from noticing that the standard deficit framing didn't match his own social experience or that of people he studied.

That matters because most of the research autism was built on for decades came from outside observers studying autistic people as a puzzle to explain, not from autistic researchers explaining their own experience. Double empathy is one of the clearest cases of that shift actually changing the conclusion.

The Autism Diagnosis Rested on This for Decades

Theory of mind testing, most famously the Sally-Anne task, became the standard proof that autistic people struggle to understand other minds. Kids who failed it got read as lacking a basic human capacity.

What that test actually measures is whether a child can predict a neurotypical adult's false belief using neurotypical social logic. An autistic child who reasons differently but consistently can still fail it, not because they can't model other minds, but because they're modeling them by a different method.

Double Empathy at Work

A manager reads your lack of small talk as disengagement. You read their small talk as a delay before the actual point. Both readings are wrong about intent and right about style.

Naming this directly changes outcomes: 'I skip small talk because I find it easier to focus straight on the task. It's not a signal I'm unhappy.' Said once, clearly, it prevents months of quiet misjudgment.

Double Empathy in Healthcare

A doctor reads flat affect during a pain description as the pain being mild. An autistic patient's flat affect during severe pain is a communication style mismatch, not a symptom of how much it hurts.

If this applies to you, say the number, not just the face: 'This is a 7 out of 10, even though my expression isn't changing much.' Give the clinician the data point their training relies on, since your face isn't going to give it to them reliably.

Why This Doesn't Mean 'Just Find Autistic Friends'

Autistic-to-autistic communication research is encouraging, but most of your life still runs through mixed interactions: family, work, healthcare, school.

The point isn't to withdraw into an all-autistic bubble. It's to stop blaming yourself entirely for a gap that belongs to both sides, and to start using direct, explicit communication as a bridge instead of trying to intuit a rulebook that was never fully explained to you.

A Script for Explaining This to Family

'You've probably noticed I communicate differently than you expect. Research on autism now shows that's a style difference, not a deficiency on my end. I'd rather we both adjust a little than have you read my style as rudeness.'

That one paragraph, said once at a calm moment rather than mid-argument, does more long-term work than years of unexplained friction.

The Word 'Rapport' Means Different Things

Neurotypical rapport often runs on shared small talk, matched tone, and reciprocal vulnerability shared gradually. Autistic rapport often runs on shared special interests, direct honesty, and comfortable silence without needing to fill it.

Both are real forms of connection. The mismatch happens when one person expects the other's style and reads its absence as coldness. If a friendship or relationship feels one-sided in effort, check whether you're actually measuring different currencies rather than different amounts of care.

When to Push Back on the Deficit Framing

If a school, employer, or clinician describes your communication purely as a deficit to fix, you can push back with the research itself: cite Milton's 2012 paper and Crompton's 2020 study by name if you need the authority of a citation in the room.

You're not obligated to accept a framing that treats a two-way mismatch as entirely your problem to solve alone.

What Changes Once You Stop Blaming Yourself Entirely

A lot of autistic adults carry years of quiet shame about being 'bad at people.' Double empathy research doesn't erase every real social difficulty autism brings, but it does remove the idea that the entire burden of the mismatch sits on you.

That shift alone, understanding the gap as shared rather than a personal failing, is often the first real relief people report after learning about this theory.

The Version of This You'll Actually Use in a Fight

Mid-argument, nobody has patience for a research citation. What works instead is a shorthand you agree on beforehand: a word or phrase that means 'we're hitting a communication style gap, not a values gap.'

Some couples use 'translation error' as the phrase. Whatever you pick, agreeing on it in a calm moment means you both have an exit ramp available the next time a normal disagreement threatens to spiral into a character attack neither of you intended.

Why Autistic Advocacy Groups Push This Theory Hard

Double empathy shows up constantly in autistic-led advocacy because it directly counters decades of research written entirely from a non-autistic observer's viewpoint.

It's also one of the few theories in autism research built by an autistic researcher studying his own community from the inside, which is part of why it landed differently than earlier deficit models did.

One More Thing Worth Saying Plainly

None of this means every social difficulty you have is someone else's fault. Sensory overload, executive dysfunction, and genuine exhaustion from masking are real and separate from the double empathy gap.

What double empathy does is take one specific accusation off the table: that you're inherently worse at connecting with people. You're not. You're running a different protocol, and the people around you can learn to read it if you're willing to hand them the manual.

What Doesn't Help, and Why People Keep Suggesting It

Traditional social skills training teaches autistic people to imitate neurotypical norms: more eye contact, smaller pauses, matched small talk. It gets recommended constantly because it's the oldest, most established intervention, and because it treats the visible half of the mismatch, yours, while leaving the other half untouched.

The problem isn't that eye contact or small talk are bad skills to have in your toolkit. It's that training only one side of a two-way gap doesn't close the gap, it just shifts more of the adjustment cost onto you permanently. A double-empathy-informed approach asks both sides to adjust, which is a genuinely different intervention, not a stricter version of the same one.

How This Lands Differently If You Were Diagnosed Late

If you spent decades believing the deficit story before encountering double empathy, the shift can be jarring in a specific way: it means years of self-blame were aimed at the wrong target.

That's worth sitting with rather than rushing past. Grieving the years spent thinking you were simply bad at people, while also feeling real relief at the reframe, can happen at the same time. Neither cancels the other out. Some people find it helps to say the grief out loud once, specifically: 'I spent years thinking this was entirely my fault. It wasn't, and I'm allowed to be angry about that before I move on.'

What This Looks Like With Your Own Kids

If you're an autistic parent of an autistic child, double empathy explains something specific: why the two of you often understand each other instantly on things that confuse everyone else in the family.

It also explains friction with a non-autistic partner or grandparent who reads the same interaction completely differently. Naming it out loud in front of the whole family, 'we process this differently, and that's fine, we just need to translate sometimes,' prevents the child from absorbing the idea that one parent is right and the other is broken, or worse, that the way they naturally connect with you is somehow the wrong way to be close to a parent.

When a Relationship Is Worth the Translation Work

Not every relationship is worth the ongoing work of naming the gap and adjusting for it. Some people, once they understand there's a real, two-way mismatch rather than a one-sided flaw, still aren't willing to meet you halfway.

That's useful information, not a reason to blame yourself further. A relationship where only you do the translating, indefinitely, isn't one where double empathy is actually being applied. It's the old deficit model wearing new language, and it's fair to name that plainly to the other person before deciding how much further to invest.

Why Online Autistic Communities Feel Different

A lot of autistic adults describe feeling instantly understood in autistic-led online spaces, a group chat, a forum, a specific subreddit, in a way years of in-person effort never produced. Double empathy offers a straightforward explanation: those spaces run on a shared communication style, so the usual translation tax simply isn't there.

That's not proof you're incapable of connecting outside those spaces. It's evidence of how much smoother connection can be once you're not constantly bridging a two-way gap by yourself.

References

Milton, D. (2012). On the ontological status of autism: The 'double empathy problem.' Disability & Society, 27(6), 883-887. doi.org/10.1080/09687599.2012.710008

Crompton, C. J., Ropar, D., Evans-Williams, C. V. M., Flynn, E. G., & Fletcher-Watson, S. (2020). Autistic peer-to-peer information transfer is highly effective. Autism, 24(7), 1704-1712. pubmed.ncbi.nlm.nih.gov/32431157