The Double Empathy Problem: Rethinking Autistic Communication

For decades, the standard account of autism described a deficit in social communication located in the autistic person. Autistic people, the theory went, find it hard to read other people’s minds, so communication breaks down. The double empathy problem offers a different explanation, and a growing body of research supports it.

Milton’s argument

Damian Milton is an autistic academic at the University of Kent, with a background spanning sociology, psychology, philosophy and education. In a short paper in Disability & Society in 2012, “On the ontological status of autism: the ‘double empathy problem'”, he argued that the communication difficulties seen between autistic and non-autistic people are a mismatch between two groups who experience and express the world differently.

Non-autistic people, Milton pointed out, are often just as poor at understanding autistic people as the reverse. The difference is that non-autistic people are the majority, so their way of communicating is treated as the norm, and the breakdown is recorded as a problem in the autistic person.

The paper was a theoretical argument, drawing on sociology and philosophy. It did not present new data. Empirical tests came later.

The evidence

Thin-slice judgements

Noah Sasson and colleagues (2017), in Scientific Reports, showed non-autistic observers very short video clips of autistic and non-autistic adults. Observers rated the autistic adults less favourably on almost every trait, and said they were less willing to interact with them. These impressions formed within seconds.

When observers read transcripts of what the same people said, with no video or audio, the difference disappeared. The content of what autistic participants said was judged just as favourably. The negative judgements were driven by voice, face and movement, not by what was said.

The diffusion-chain study

The best-known direct test of the double empathy idea is a study by Catherine Crompton, Danielle Ropar, Claire Evans-Williams, Emma Flynn and Sue Fletcher-Watson (2020), published in Autism.

The researchers borrowed a method from studies of cultural transmission: the diffusion chain, a structured version of the children’s game in which a message is whispered along a line. Seventy-two adults were placed in nine chains of eight people. Three chains were all autistic, three were all non-autistic, and three alternated autistic and non-autistic participants. The first person heard a story and retold it to the next, who retold it to the next, and so on.

The results:

• Autistic chains passed on as much detail as non-autistic chains. There was no significant difference.
• Mixed chains lost detail significantly faster than either.
• Participants in mixed chains also reported lower rapport with their partners.

If autism simply involved a communication deficit, the all-autistic chains should have done worst. They did as well as the non-autistic ones. The difficulty appeared where neurotypes met.

Rapport

In a companion study in Frontiers in Psychology (2020), Crompton and colleagues found that observers watching videos of pairs rated rapport lower in mixed autistic and non-autistic pairs than in matched pairs. Autistic and non-autistic observers showed similar patterns.

Limits of the evidence

These studies are recent and relatively small, and mostly involve UK and US adults. The diffusion-chain study used one storytelling task. Peter Mitchell, Elizabeth Sheppard and Sarah Cassidy (2021), reviewing the field in the British Journal of Developmental Psychology, argue that the double empathy idea has implications for development and mental health: if autistic people are routinely misread, they have fewer chances to take part in social life, and both groups have fewer chances to learn each other’s ways of communicating. More research with children, and in everyday settings such as classrooms, is needed.

The double empathy problem does not deny that many autistic people find social situations hard, or that some need significant support with communication. It changes where we look for the problem, and who is expected to adapt.

Neurodiversity-affirming communication

The neurodiversity approach treats neurological differences, including autism, as natural variation in human minds, while recognising that disability and support needs are real. Communication that follows from it puts effort on both sides of the conversation.

Language

Lorcan Kenny and colleagues (2016) surveyed 3,470 people in the UK autism community about terminology. Many autistic adults preferred identity-first language (“autistic person”), while many professionals preferred person-first language (“person with autism”). There was no consensus, and the authors recommended asking individuals what they prefer. All groups disliked “high-functioning” and “low-functioning”.

In the classroom

• Teach about communication differences explicitly, to all pupils. A lesson on how people show interest, or that eye contact is not the only sign of listening, benefits the whole class.
• Do not require eye contact as proof of attention. Many autistic pupils listen better when they are not forced to look at a face.
• Say what you mean. Literal, specific instructions (“Put your book in the red tray”) are clearer for many pupils than hints (“Is that where your book goes?”).
• Check understanding in more than one way: written, spoken, drawn or demonstrated.
• Create chances for autistic pupils to work and socialise with each other, such as interest-based clubs, as well as with non-autistic peers.
• Treat social skills interventions with care. If a programme teaches only the autistic pupil to adapt, ask what the rest of the school community is learning.

At home

• Value your child’s way of communicating, including scripts, special interests, info-dumping and stimming, as forms of connection.
• Help family members and friends understand your child’s communication, not only the other way round.
• Look for autistic-led communities and resources. Many autistic adults write about their experience in ways parents find helpful.

Getting support

If you think your child may be autistic, talk to the school’s SENCO and your GP. In England, the GP or school can refer for an NHS assessment. A diagnosis is made by specialist professionals, and support in school does not have to wait for one.

Further reading

• Crompton, C. J., Ropar, D., Evans-Williams, C. V., Flynn, E. G. and Fletcher-Watson, S. (2020). Autistic peer-to-peer information transfer is highly effective. Autism, 24(7), 1704-1712.
• Crompton, C. J., Sharp, M., Axbey, H., Fletcher-Watson, S., Flynn, E. G. and Ropar, D. (2020). Neurotype-matching, but not being autistic, influences self and observer ratings of interpersonal rapport. Frontiers in Psychology, 11, 586171.
• Kenny, L., Hattersley, C., Molins, B., Buckley, C., Povey, C. and Pellicano, E. (2016). Which terms should be used to describe autism? Perspectives from the UK autism community. Autism, 20(4), 442-462.
• Milton, D. E. M. (2012). On the ontological status of autism: the “double empathy problem”. Disability & Society, 27(6), 883-887.
• Mitchell, P., Sheppard, E. and Cassidy, S. (2021). Autism and the double empathy problem: implications for development and mental health. British Journal of Developmental Psychology, 39(1), 1-18.
• Sasson, N. J., Faso, D. J., Nugent, J., Lovell, S., Kennedy, D. P. and Grossman, R. B. (2017). Neurotypical peers are less willing to interact with those with autism based on thin slice judgments. Scientific Reports, 7, 40700.