In simple terms, the ‘double empathy problem’ refers to a breakdown in mutual understanding (that can happen between any two people) and hence a problem for both parties to contend with, yet more likely to occur when people of very differing dispositions attempt to interact. Within the context of exchanges between autistic and non-autistic people however, the locus of the problem has traditionally been seen to reside in the brain of the autistic person. This results in autism being primarily framed in terms of a social communication disorder, rather than interaction between autistic and non-autistic people as a primarily mutual and interpersonal issue.

It has been 10 years since the ‘double empathy problem’ as a term was first described within the pages of an academic journal (Milton, 2012). Although, importantly, the conceptualisation of the issue has since its inception been influenced by and framed within a broader history of academic theorising (particularly from the disciplines of Sociology and Philosophy). Yet, this coining of the term helped express an issue that had long been discussed within autistic community spaces. The initial conceptualising of the double empathy problem was critical of theory of mind accounts of autism and suggested that the success of an interaction partly depended on two people sharing similar experiences of ways of being in the world. This is not to say that autistic people will automatically be able to connect and feel empathy with other autistic people they meet any more than two random non-autistic people would; however, there is greater potential for such, at least in how being autistic (or not) shapes experiences of the social world. An obvious example would be how differing sensory perceptions would impact communicating with others and shared understanding.

While there is much work to be done to explore these issues across multiple disciplines, the concept of the double empathy problem has the potential to aid a reframing of autism itself from a social communication disorder to a description of a broad range of developmental differences and embodied experiences and how they play out in specific social and cultural contexts. If this were so, it would lead to a radical change to current diagnostic criteria. This is most important however when considering best practice models for supporting autistic people in a variety of settings. We already know that interpretations about autistic sociality from observations alone may not be accurate (Doherty et al., 2022; Mitchell et al., 2021). Instead of focusing on perceived social deficits and normative remediation, the concept suggests a position of humility in the face of difference, the need to build rapport and understanding and not assume a lack of capacity for understanding. Ultimately, the concept reminds us of the social situatedness of the lives of autistic people and those who support them.

The ‘double empathy problem’: Ten years on – Damian Milton, Emine Gurbuz, Betriz Lopez, 2022

The original published definition of the double empathy problem is as follows: 

A disjuncture in reciprocity between two differently disposed social actors which becomes more marked the wider the disjuncture in dispo- sitional perceptions of the lifeworld – perceived as a breach in the ‘natu- ral attitude’ of what constitutes ‘social reality’ for ‘neuro-typical’ people and yet an everyday and often traumatic experience for ‘autistic people’. 

(Milton 2012a, p. 884) 

Due to differing qualia of experience, social lifeworlds, dispositional viewpoints and discursive repertoires, interactions between autistic and non-autistic people are vulnerable to breaches in mutual understanding, framed as a ‘double problem’ as both parties in the interaction will experience a sense of disjuncture, not simply a deficit in the autistic person’s mind. Whilst this experience may be novel for many non-autistic people, it is common- place for autistic people. Such a framing would also suggest a greater likelihood of feelings of empathy between autistic people with one another and with those they have close relationships with, yet perhaps over differing elements of their lives.

(PDF) Autism and the ‘double empathy problem’

These studies suggest that stereotyped views of autistic people are likely to contribute to the double empathy problem, and that there may also be differences between people’s perceptions of being helpful and actu- ally being so to others. 

In recent research by Crompton et al. (2020), the transfer of informa- tion between people were studied across a diffusion chain of eight people in total, similar to a game of ‘telephone’. When there were only autistic par- ticipants or only non-autistic participants, there was equally good transfer of information. However, when there was a mixed diffusion chain of autistic and non-autistic people, there was a much greater reduction in information successfully passed on. 

Further research reflects the ‘double empathy problem’ resulting in social breakdowns within a given group. The dominant form of sociality could be suggested to be based on social group identification and dominated by non-autistic people. The basis of autistic socialisation is interest-based (Bertilsdotter-Rosqvist 2019). The mismatch of social form and enacting the necessary mode (interest-led versus social alignment) may hinder the flow of the group and ultimately result in social exclusion. The analysis of bloggers’ posts indicate a ‘double empathy problem’ through the dispar- ity of metaperception and the consequential impact (Welch et al. 2022). There are real-life applications of the double empathy problem across set- tings and dimensions, such as in the criminal justice system (Holloway et al. 2020), education (Hummerstone and Parsons 2021), employment and job interviews (Maras et al. 2021; Remington and Pellicano 2019), and even the daily dissonance of the autistic lived experience (e.g., impression man- agement: Cage and Troxell-Whitman 2019; Cook et al. 2021; Schneid and Raz 2020; understanding the use of gaming: Pavlopoulou et al. 2022) that may include ‘thwarted belonging’ and lead to suicidality (Cassidy et al. 2018; Pelton et al. 2020), and breakdowns in feelings of social inclusion and belonging between autistic and non-autistic individuals (Waldock et al. 2021). In a study by Chen et al. (2021), natural peer interactions among six autistic and six non-autistic young people were observed over a five-month period to examine peer preferences and real-world social interactions. The findings showed that the young people preferred within neurotype interac- tions and that such interactions were more reciprocal and relational (rather than instrumental), such as sharing thoughts and experiences. 

The evidence is thus building to suggest that the theory of mind deficit theory of autism is indeed ‘partial at best’ with growing support for the double empathy problem. 

(PDF) Autism and the ‘double empathy problem’

Being autistic affects how people make sense of the world around them, and some autistic people can find it hard to communicate. For a long time, research has shown that autistic people can have trouble figuring out what non-autistic people are thinking and feeling, and this can make it difficult for them to make friends or to fit in. But recently, studies have shown that the problem goes both ways: people who are not autistic also have trouble figuring out what autistic people are thinking and feeling! It is not just autistic people who struggle.

A theory that helps to describe what happens when autistic and non-autistic people struggle to understand each other is called the double empathy problem. Empathy is defined as the ability to understand or be aware of the feelings, thoughts, and experiences of others. According to the double empathy problem, empathy is a two-way process that depends a lot on our ways of doing things and our expectations from previous social experiences, which can be very different for autistic and non-autistic people. These differences can lead to a breakdown in communication that can be distressing for both autistic and non-autistic people. It might sometimes be difficult for non-autistic parents to understand what their autistic child is feeling, or autistic people might feel frustrated when they cannot effectively communicate their thoughts and feelings to others. In this way, communication barriers between autistic and non-autistic people can make it more difficult for them to connect, share experiences, and empathize with one another.

Double Empathy: Why Autistic People Are Often Misunderstood · Frontiers for Young Minds

• Cameron (2012) uses the term ‘dyspathy’ to highlight how empathy is often blocked or resisted by people.
• Cameron (2012) cites a number of recent studies using fMRI scanning claim to demonstrate a bias towards in-group members in ‘automatic’ empathy.
• Such findings support the earlier social psychological theories of Tajfel (1981), which found that people felt increasing emotional connection to those deemed within their social ‘in-group’, whilst stereotyping ‘outsiders’.

Source: From finding a voice to being understood: exploring the double empathy problem

To be defined as abnormal in society is often conflated with being perceived as ‘pathological’ in some way and to be socially stigmatised, shunned and sanctioned. Then, if there is a breakdown in interaction, or indeed a failed attempt to align toward expressions of meaning, a person who sees their interactions as ‘normal’ and ‘correct’ can denigrate those who act or are perceived as ‘different’ (Tajfeel & Turner, 1979). If one can apply a label on the ‘other’ locating the problem in them, it also resolves the applier of the label’s ‘natural attitude’ of responsibility in their own perceptions and the breach is healed perceptually, but not for the person who has been ‘othered’ (Said, 1978).

A Mismatch of Salience | Pavilion Publishing and Media

For autistic people, we don’t feel this aligned from a very early age, so it’s, other people are not mirroring us so much or there’s this disjuncture often. So we don’t build up an expectation of alignment.

Autism’s Double Empathy Problem Conference

Firstly, we’ve had a huge amount of first-person accounts and anecdotal evidence that autistic people can find spending time with other autistic people more comfortable and easier and less stressful, and just easier than interacting with non-autistic people. We’ve heard a lot from people who have said, “once I found more autistic people I thought I had found my community” and this kind of stuff. And we didn’t have any empirical evidence to back that at all.

We’ve got a theoretical framework within the double empathy problem which kind of says a similar thing, in that the problems of interacting and interactions between autistic and neurotypical people isn’t necessarily all down to a deficit on the part of the autistic person. It’s more to do with a mismatch in communication style, and mismatch in background.

There is now a growing body of evidence that’s looking at double empathy problem matters, but when we started this project we were really keen to try to address these two areas in an empirical and data-driven way, to see whether this is something that we could explore scientifically in a controlled fashion. We were really interested to see if our theories would stand up to empirical tests.

The Problem With Autistic Communication Is Non-Autistic People: A Conversation With Dr. Catherine Crompton — THINKING PERSON’S GUIDE TO AUTISM

Whilst it is true that autistic people can struggle to process and understand the intentions of others within social interactions, when one listens to the accounts of autistic people, one could say such problems are in both directions. Theory of autistic minds often seem to leave a lot to be desired, and we would not need organisations like the National Autistic Society trying to spread awareness and understanding of autism if it were so easy to empathise with autistic ways of perceiving and being in the world. From the earliest written accounts of autistic people one can see numerous mentions of this lack of understanding from others. It is this issue of empathy problems between autistic and non-autistic people being mutual in character that led to the development of the ‘double empathy problem’ as a theory.

Simply put, the theory of the double empathy problem suggests that when people with very different experiences of the world interact with one another, they will struggle to empathise with each other. This is likely to be exacerbated through differences in language use and comprehension. I first started to publish theoretical accounts of this issue in the early 2010s, yet similar ideas can be found in the work of Luke Beardon regarding ‘cross-neurological theory of mind’ and in that of the philosopher Ian Hacking.

More recently research by Elizabeth Sheppard and team at the University of Nottingham, Brett Heasman at the London School of Economics, and Noah Sasson at the University of Texas at Dallas, have shown that in experimental conditions, non-autistic people struggled to read the emotions of autistic participants, or form negative first impressions of autistic people. Such evidence would suggest that the dominant psychological theories of autism are partial explanations at best.

According to the theory of the ‘double empathy problem’, these issues are not due to autistic cognition alone, but a breakdown in reciprocity and mutual understanding that can happen between people with very differing ways of experiencing the world. If one has ever experienced a conversation with someone who one does not share a first language with, or even an interest in the topic of a conversation, one may experience something similar (albeit probably briefly).

This theory would also suggest that those with similar experiences are more likely to form connections and a level of understanding, which has ramifications in regard to autistic people being able to meet one another.

The double empathy problem

Our interim findings can be summarised as follows

1. Autistic people share information with other autistic people as effectively as non-autistic people do.
2. information sharing can break down when pairs are from different neurotypes – when there is an autistic and a non- autistic person.
3. Feelings of rapport between people of the same neurotype accompany these information-sharing benefits – autistic people have higher rapport with other autistic people, and non-autistic people have higher rapport with non-autistic people.
4. External observers can detect the lack of rapport apparent in mixed autistic/non-autistic interactions.

In essence, what we are demonstrating for the first time is that autistic people’s social behaviour includes effective communication and effective social interaction, in direct contradiction of the diagnostic criteria for autism. We have, for the first time, uncovered empirical evidence that there is a form of social intelligence that is specific to autistic people.

Diversity in Social Intelligence

The Double Empathy Problem suggests that communicative difficulties between autistic and non-autistic people are due to bi-directional differences in communicative style and a reciprocal lack of understanding. If true, there should be increased similarity in interaction style, resulting in higher rapport during interactions between pairs of the same neurotype. Here, we provide two empirical tests of rapport, with data revealing whether self- and observer- rated rapport varies depending on the match or mismatch in autism status within a pair.

In summary, autistic people experience high interactional rapport when interacting with other autistic people, and this is also detected by external observers. Rather than autistic people experiencing low rapport in all contexts, their rapport ratings are influenced by a mismatch of diagnosis. These findings suggest that autistic people possess a distinct mode of social interaction style, rather than demonstrating social skills deficits. These data are considered in terms of their implications for psychological theories of autism, as well as practical impact on educational and clinical practice.

The results indicate that participants, regardless of diagnostic status, give poorer ratings of rapport for mixed neurotype pairs than for matched neurotype pairs. This suggests a mismatch between neurotypes results in lower ratings of rapport, and that subtle verbal and non-verbal cues to rapport are similarly perceptible by autistic and non-autistic individuals. Interestingly, rapport scores were significantly higher for the autistic pairs than non-autistic pairs, indicating that the autistic dyads may display even greater social signals of shared enjoyment and ease when interacting with one another, as viewed by an external observer.

An exploratory comparison between participants’ own judgments of rapport and an observer’s ratings, suggests autistic participants’ self-rating of rapport are more in line with others’ ratings of rapport. There was a greater discrepancy between non-autistic participants’ estimates of their rapport with a partner compared with observers’ rating of the same social interaction.

Frontiers | Neurotype-Matching, but Not Being Autistic, Influences Self and Observer Ratings of Interpersonal Rapport | Psychology

And this is where the neurotypical belief in theory of mind becomes a liability. Not just a liability – a disability.

Because not only are neurotypicals just as mind-blind to autistics as autistics are to neurotypicals, this self-centered belief in theory of mind makes it impossible to mutually negotiate an understanding of how perceptions might differ among individuals in order to arrive at a pragmatic representation that accounts for significant differences in the experiences of various individuals. It bars any discussion of opening up a space for autistics to participate in social communication by clarifying and mapping the ways in which their perceptions differ. Rather than recognize that the success rate of the neurotypical divining rod is based on mere statistical likelihood that the thoughts and feelings of neurotypicals will correlate, they declare it an ineffable gift, and use it to valorize their own abilities and pathologize those of autistics.

A belief in theory of mind makes it unnecessary for neurotypicals to engage in real perspective-taking, since they are able, instead, to fall back on projection. Differences that they discover in autistic thinking are dismissed as pathology, not as a failure in the neurotypical’s supposed skill in theory of mind or perspective-taking.

Ironically, constantly confronted with the differences in their own thinking and that of those around them, and needing to function in a world dominated by a different neurotype, autistics are engaged in learning genuine perspective-taking from the cradle on. The perceived failure in that perspective-taking is thus based on the fact that autistics do not rely on and cannot rely on neurological similarities to crib understanding by projecting their own thoughts and feelings onto others.

As such, autistics talk about themselves rather than others, a feature of autistic narrative that has been pathologized as “typically autistic” by researchers like Ute Frith. The fact that much of autistic writing is dedicated to deconstructing neurotypical fallacies about autistic thinking set in the world when they spoke about (or for) us, and to explaining differences in autistic thinking in order to broker mutual understanding remains unremarked upon, as it would have required adequate perspective-taking to have identified this.

Thus, if we were to summarize the effect of neurotypicals sitting in wells that are structured in much the same way, delimited in much the same way, oriented in the same general direction and located in the same geographic location, manifested as an unassailable belief in their natural gift of theory of mind, we would have to conclude that this belief in theory of mind severely impairs neurotypicals’ ability to perceive that there is sky or even the great sea outside the narrow limits of their purview. It also necessarily impacts their cognitive empathy vis-à-vis autistics and, sadly, their affective empathy as well.

This deficit in neurotypicals needs to be remediated if autistics are to have a chance to participate as equals, because the truth is, in this regard, autistics suffer and are excluded from social communication not because of our own disability, but because of neurotypical disability.

The belief in a theory of mind is a disability — Semiotic Spectrumite

The 20th Century political scientist Karl Deutsch said, “Power is the ability not to have to learn.”

I quote this statement often, because I think it’s one of the most important truths ever articulated about privilege, oppression, and social power relations.

When a social system is set up such that one particular group is almost always in a position of social power or privilege over another group, the members of the privileged group never truly need to learn or practice empathy or understanding for the members of the disempowered, oppressed group. Nor do the members of the privileged group need to learn to adapt to the communication style of the oppressed group.

Neurotypical privilege means that neurotypical people interacting with autistic people—particularly when the neurotypical people in question are in positions of professional authority—have the luxury of never having to address or even acknowledge their own empathy deficits or poor communication skills, because they can blame all failures of empathy, understanding, and communication on the alleged deficits of the autistic people.

Power—or privilege, as we now more commonly call the particular kind of power to which Deutsch was referring—is the ability not to have to learn. There’s a phrase, “check your privilege,” that’s often repeated but rarely understood or heeded by those privileged persons at whom it is directed. If we start from Deutsch’s definition of power or privilege as the ability not to have to learn, we can understand “check your privilege” to mean, at least in part, “Learn! Be quiet, pay attention, and learn. Learn, even though the learning process, and the level of profound humility it requires, is going to be uncomfortable. Learn even though, because of your privilege, this sort of learning and humility is a discomfort that you have the luxury of being able to avoid—a luxury that we didn’t have, when we had to learn your ways. Learn even though you don’t have to.”

Unfortunately, as members of all oppressed groups discover, most privileged people just won’t do that. The states of profound mindfulness, humility, openness to correction, and tolerance for uncertainty that such learning demands are too far outside of most people’s comfort zones. Most human beings simply won’t go that far outside of their comfort zones if they don’t have to. And privilege means they don’t have to.

NEUROTYPICAL PSYCHOTHERAPISTS & AUTISTIC CLIENTS • NEUROQUEER

I don't want to know
I don't want to know what they're saying about me
I don't want to know
I don't want to show that it devastates me

I'm living somewhere nobody goes to
I'm speaking in a language nobody talks
The window broken, a cold wind blows through
My soul a series of electrical shocks

--Trans Mantra by Ezra Furman

Further reading,