The double empathy problem

For a long time, autism research described social difficulty as something located inside the autistic person: a deficit in reading minds or showing empathy. The double empathy problem asks a different question. If an autistic person and a non-autistic person misunderstand each other, why do we only count it as the autistic person's failure?

What is the double empathy problem? Meaning and origin

The term comes from a 2012 paper by Damian Milton, a sociologist who is himself autistic. His argument was that empathy depends on shared experience. When two people experience the world very differently, through different senses, priorities and styles of talking, both will find it harder to understand the other. The non-autistic person is just as likely to misread the autistic person as the other way round. Because non-autistic people are the majority, their misreadings are rarely noticed or named.

The research behind it

Several studies have tested predictions that follow from the idea. Two are cited especially often.

Researchers have also linked the idea to mental health. One account proposes that repeated misperception by the majority can push autistic people toward isolation, which in turn harms wellbeing. It explicitly rejects the older idea that autistic people simply lack social motivation.

Double empathy problem examples from adult life

• At work: an autistic employee answers "Do you have a minute?" with "No, I am in the middle of something." They meant it literally and helpfully. The manager hears rudeness. The manager's indirect request was just as unclear to the employee.
• In a relationship: one partner shows care by fixing a problem or sharing a related fact; the other expected a hug and sympathetic words. Both care, and both feel unseen.
• Eye contact: a non-autistic interviewer reads low eye contact as evasive, while the autistic candidate is looking away precisely so they can concentrate on the question.
• With other autistic people: many autistic adults describe conversations with autistic friends as easy, with no need to decode hints or perform interest, and long silences that feel comfortable rather than awkward.

• Autism and Empathy: What Research ShowsDo autistic people lack empathy? No. What research says about cognitive vs emotional empathy, double empathy, alexithymia, and empathy for objects.
• Autism and Eye ContactWhy many autistic adults avoid or struggle with eye contact, what research says it feels like, and practical ways to handle meetings, dates and interviews.

Criticisms of the double empathy problem

The idea is popular and influential, but it has critics, and a fair account should include them.

Other points raised in the wider debate include that many studies are small, that "empathy" is used loosely to mean several different things, and that the idea does not erase real difficulties some autistic people have even with each other. Supporters respond that the core claim, that mutual understanding depends on both people, is modest and well supported. Both sides agree more and larger studies are needed.

How it differs from older theories

Earlier accounts, especially theory of mind research from the 1980s and 1990s, framed autism as difficulty imagining what other people think. Those studies usually tested autistic people with non-autistic researchers and non-autistic norms, so a mismatch could only ever be scored as an autistic error. The double empathy idea does not claim autistic people read everyone perfectly. It claims the test setup was one-sided, and that a fair measure should look at how well each group reads the other.

This shift also changes what counts as a social skill. If autistic people communicate well among themselves, then teaching them to copy non-autistic habits is only half the picture. Some researchers argue for teaching non-autistic people too, for example managers, teachers and clinicians, about direct communication and different ways of showing interest.

Why it matters in practice

Whatever its final scientific status, the idea has practical value. It shifts some responsibility for good communication onto both people, which supports simple fixes: written agendas, direct questions, explicit feedback, and less weight on eye contact or small talk in interviews. For autistic adults, it can also ease self-blame after years of being told they were the problem.