Double Empathy Problem: What It Means at Work

The double empathy problem says misunderstandings between autistic and non-autistic people go both ways. What it means, the research, and how to use it at work.

The double empathy problem is the idea that when two people experience the world very differently, misunderstandings between them run both ways. It was described by autistic researcher Damian Milton in 2012. Instead of seeing autistic people as lacking social understanding, it says autistic and non-autistic people can find each other equally hard to read, so the gap is shared.

What is the double empathy problem?

For a long time, the common view of autism was that autistic people struggle to understand other people's thoughts and feelings. Communication problems were seen as coming from one side only.

The double empathy problem turns that view around. It says that when two people have very different ways of seeing and experiencing the world, both of them can find the other hard to understand. A non-autistic person may misread an autistic colleague just as much as the other way around. The difficulty sits in the gap between them, not inside one person.

The word "double" is the key. Empathy has to travel in two directions, and it can break down in both.

Who came up with the double empathy problem?

Damian Milton, an autistic sociologist, set out the idea in a short 2012 paper in the journal Disability & Society called "On the ontological status of autism: the double empathy problem" (University of Kent repository).

Milton argued that explanations of autism had focused on what autistic people supposedly lack, and had overlooked the fact that social interaction always involves at least two people. When those people have very different outlooks, mutual understanding is harder for both of them.

The paper was a theoretical argument, not a study. The evidence came later. The National Autistic Society has a clear overview of the idea and how it is used in practice.

What does the research show?

The best-known test of the idea is a 2020 study by Catherine Crompton and colleagues, published in the journal Autism (full paper on PubMed Central).

The researchers used a "diffusion chain," which works like the children's game of telephone. One person hears a story and retells it to the next, who retells it to the next, and so on down a chain of eight people. There were three kinds of chain:

chains of only autistic adults
chains of only non-autistic adults
mixed chains of autistic and non-autistic adults

The results:

Autistic chains passed on information as well as non-autistic chains. There was no significant difference between them.
Mixed chains lost detail more quickly than either of the others.
People in mixed chains rated their rapport with each other lower.

In other words, autistic people communicated well with other autistic people. The difficulty showed up when the two groups mixed, which is what the double empathy problem predicts.

What the research does not show

It is worth being honest about the limits. This was one laboratory task (retelling a story) with 72 adults in the UK. Real conversations at work are longer, messier and full of context. The study supports the idea, but it does not prove everything that is sometimes claimed in its name.

It is also important that the research so far is mostly about autistic and non-autistic people. Many people find the idea useful for thinking about other neurotypes too, such as ADHD, and about differences in communication style generally. That can be a helpful way to think, but it goes beyond what the studies have tested, so it is best described as an idea people find useful rather than a proven finding.

Why does the double empathy problem matter?

The double empathy problem changes the question we ask when communication goes wrong.

The old question was: what is wrong with this person's communication, and how do we fix it? That puts all the work on one person, often the autistic or neurodivergent one, who is asked to mask, translate and adapt all day.

The new question is: where is the gap between these two people, and what can each of them do to close it? That shares the work.

This matters because a lot of advice about communicating with neurodivergent people still treats them as the problem to solve. The double empathy problem is a reminder that neurotypical communication has its own habits that can be just as confusing to someone else, such as hints, implied deadlines, small talk with hidden meaning and "read the room" expectations.

How does the double empathy problem show up at work?

Here are a few everyday situations where a two-way gap can open up. None of these describe every autistic or neurodivergent person. Neurodivergent people are not a monolith, and two people with the same diagnosis can communicate very differently.

Hints and implied requests. A manager says, "It might be worth looking at the numbers again." They mean, "Please fix the numbers." A colleague who reads words literally may take it as an optional idea. Neither person is wrong about what the words say. They are reading them differently.
Directness. One person gives a short, direct answer: "That won't work, the deadline is too tight." Another hears it as blunt or dismissive. The first person meant it as helpful honesty.
Small talk and "real" talk. One person sees small talk as the way trust is built. Another sees it as a delay before the actual topic. Each may think the other is being odd.
Body language and tone. Some people rely heavily on facial expressions and tone of voice to understand meaning. Others do not express or read these cues in the same way. In meetings, this can lead to both sides misjudging how the other feels.
Unwritten rules. "Everyone knows" you reply to the boss quickly, or that "let's discuss later" means "no." Not everyone knows. Rules that are never written down are often the first place a gap appears.

Try it: one message, two readings

The message as written:
"It might be worth having another look at the client summary before Friday."

How it could be misread:
A reader who takes words literally may think this is an optional suggestion. Someone who reads between the lines may hear "this is not good enough" and feel criticized. Both readings are reasonable, and neither is what the writer meant: "Please fix two numbers by Friday."

A version that works for both people:
"Could you update the client summary by Friday? The totals on page 2 don't match the latest figures. Everything else looks good."

The second version is not colder. It says what is needed, why, and by when, and it says plainly that the rest is fine.

How can teams close the gap?

Because the gap belongs to both people, both people can do something about it. These habits help every kind of communicator.

If you are writing the message

Say the actual ask. Replace hints with a clear request.
Give real times. "By Thursday at 3pm" instead of "soon" or "when you get a chance."
Say how you mean it. "This is a quick question, not a problem" removes guesswork about tone.
Put important things in writing. A short written summary after a meeting helps people who process information differently, and anyone who missed part of it.

If you are reading the message

Ask instead of guessing. "Do you need this fixed, or was that a suggestion?" is a perfectly professional question.
Assume good intent first. A short reply may simply be efficient, not rude.
Say what helps you. "I work best with clear deadlines" or "I prefer a quick call for complicated topics" gives others something to work with.

As a team

Agree how you communicate. A simple team agreement can cover response times, when to use chat or email, and how to flag something urgent.
Ask everyone about preferences. Make "How do you like to get feedback?" a normal question for every new colleague, not a special question for some.
Write down the unwritten rules. If something "everyone knows" matters, put it somewhere people can read it.

What the double empathy problem is not

It is not blaming neurotypical people. It says the gap is shared. The aim is for both sides to adjust a little, not for one side to carry the guilt.
It is not saying autistic people never find communication hard. Many do, and that experience is real. The point is that difficulty in mixed communication is not one-sided.
It is not a reason to stop trying. Understanding that the gap goes both ways is the start of closing it.

Sources

Milton, D. (2012). On the ontological status of autism: the double empathy problem. Disability & Society, 27(6), 883 to 887.
Crompton, C. J., Ropar, D., Evans-Williams, C. V. M., Flynn, E. G. and Fletcher-Watson, S. (2020). Autistic peer-to-peer information transfer is highly effective. Autism, 24(7), 1704 to 1712.
National Autistic Society, The double empathy problem
Reframing Autism, Milton's double empathy problem: a summary for non-academics

Frequently Asked Questions

What is the double empathy problem in simple terms?

It is the idea that when two people experience the world very differently, they can find each other hard to understand, in both directions. The misunderstanding is shared, not one person's fault.

Who came up with the double empathy problem?

Damian Milton, an autistic sociologist, described it in a 2012 paper in the journal Disability & Society.

Is there evidence for the double empathy problem?

Yes, though it is still growing. A 2020 study by Crompton and colleagues found that autistic-only and non-autistic-only groups passed on information equally well, while mixed groups lost more detail and reported lower rapport. It was a single laboratory task, so it supports the idea rather than proving every claim made about it.

Does the double empathy problem apply to ADHD?

The research so far is mainly about autistic and non-autistic people. Many people find the idea helpful for thinking about ADHD and other differences in communication style, but that has not been tested in the same way.

How can teams use the double empathy problem at work?

Treat communication as a shared responsibility. Be specific in what you write, ask instead of guessing when you read, and agree as a team how you communicate, so no one has to do all the translating.

BRYGE AI helps with communication. It does not diagnose or treat any condition.