
Hello, and welcome back to The Autistic Writer. Certain theories in science and research are so groundbreaking, so insightful, that they represent a sea change in their respective fields. Relativity, evolution, and quantum physics are among those major ideas.
In the field of autism, the double empathy problem, a hypothesis put forward by social psychologist Damian Milton, has become one of the most influential ideas in contemporary autism theory.
“A lack of social reciprocity has often been cited as a deficit contained within the minds of autistic people […] Yet such theories psychologise what is essentially a socially negotiated interactive event.”
~Damian Milton, 2014.
What is the Double Empathy Problem?
The double empathy problem, proposed by Damian Milton in 2012, suggests that communication difficulties between autistic and non-autistic people are reciprocal. Rather than seeing misunderstanding as a deficit located solely within the autistic person, it argues that both people bring different ways of experiencing and interpreting the world, creating a mismatch in mutual understanding.
The term double empathy can be misleading. Milton’s theory is not primarily about empathy in the psychological sense of cognitive or affective empathy. It is a theory of reciprocal misunderstanding and mutual interaction between people with different dispositional experiences.
The double empathy problem is fundamentally a socio-philosophical theory of social interaction, rather than a psychological theory of individual cognition. Milton says his theory “produces a critique of autism being defined as a deficit in ‘theory of mind’, re-framing such issues as a question of reciprocity and mutuality.”
Roots
Milton acknowledges that his theory builds on observations made by other writers in the field of autism, citing Jim Sinclair and Ian Hacking as examples.
In his landmark paper, Don’t Mourn For Us, Sinclair makes this particularly relevant point:
“You try to relate as parent to child, using your own understanding of normal children, your own feelings about parenthood, your own experiences and intuitions about relationships. And the child doesn’t respond in any way you can recognize as being part of that system.
“That does not mean the child is incapable of relating at all. It only means you’re assuming a shared system, a shared understanding of signals and meanings, that the child in fact does not share. It’s as if you tried to have an intimate conversation with someone who has no comprehension of your language. Of course the person won’t understand what you’re talking about, won’t respond in the way you expect, and may well find the whole interaction confusing and unpleasant.
“It takes more work to communicate with someone whose native language isn’t the same as yours.”
~Jim Sinclair, Don’t Mourn For Us, 1993.
Implications of the Double Empathy Problem
The Neurodiversity Paradigm
Anyone familiar with the neurodiversity paradigm will immediately recognise the overlap with the double empathy problem.
The neurodiversity paradigm challenges the assumption that neurological differences should be understood as deviations from supposedly normal or ideal neurological functioning. Instead, it views neurological variation as a natural feature of humanity.
In the neurodiversity paradigm, all types of neurological functioning fall under the neurodiversity umbrella (even though we might label clusters of functioning as autistic, ADHD, or typical, etc). While the concept of neurodiversity had its roots in autism self-advocacy, the neurodiversity paradigm includes everyone: the full neurological diversity of the whole human race.
The social othering of autistic people that results from the medicalised abnormality paradigm is questioned by the neurodiversity paradigm, and then even more forcefully by Milton’s insight.
ABA
In his 2012 paper, Milton points to Applied Behaviour Analysis (ABA) as an example of how “children are socialised into what are deemed appropriate behaviours of socially functional future roles.” In the context of the double empathy problem, the supposed abnormality of autistic behavioural and interaction styles is normalised away from the medicalised paradigm that enables ABA practice, and seen as a natural variation, a difference rather than an internal deficit. This shifts the abnormality to the medicalised paradigm itself that others autistic behaviour as something to be corrected.
The Shared Burden
The double empathy problem emphasises the bidirectionality of communication differences between autistic people and non-autistic people: Both have limited understanding and insights into each other’s experience and expression. This means that the burden of adaptation required to achieve functional understanding is now shared between the two.
Theory of Mind
One of the most startling aspects of the double empathy problem is that it raises questions and offers alternative explanations for so much historical autism theory that has come to be seen as problematic as understanding of autism has increased. For example, the double empathy problem offers an explanation for why early Theory of Mind research may have interpreted communication failures as evidence of autistic deficits. If mutual misunderstanding is real, then researchers observing autistic–non-autistic interactions could mistakenly attribute the breakdown entirely to the autistic participant, because the non-autistic participant’s contribution to the interaction was assumed to be normative and therefore invisible.
Follow AFI: The podcast that explores the intersection of autism and financial independence. Hosted by Darren Scothern (The Autistic Writer) and David Scothern (the Mortgage Advisor on FIRE). Available on YouTube and all major podcast platforms.


The Criticism That is not REally Criticism
One interesting aspect of the double empathy problem as described by Milton is that it rarely mentions empathy. This might seem surprising, given that empathy is part of the theory’s name.
In May 2026, a paper was published that appeared, at first glance, to be critical of the double empathy problem. The paper, authored by Sam Fellowes, was titled, Cognitive Empathy, Affective Empathy and Empirical Research on the Double Empathy Problem: A Critical Methodology Review.
Fellowes’s central argument is that much of the empirical research claimed to support the double empathy problem does not actually measure cognitive empathy or affective empathy, and therefore cannot establish claims about those psychological constructs.
However, Milton never presents the DEP as a theory of cognitive or affective empathy in the first place. Milton defines the DEP as “A disjuncture in reciprocity between two differently disposed social actors…”
He goes on to explain that it is “…a breach in the ‘natural attitude’ that occurs between people of different dispositional outlooks and personal conceptual understandings when attempts are made to communicate meaning […] it is not a singular problem located in any one person.”
Throughout his 2012 and 2014 papers, Milton consistently frames the double empathy problem as a sociological theory of reciprocal social interaction, centred on concepts such as reciprocity, mutuality, interaction, and dispositional differences.
While Milton does mention empathy, it is not in the sense of the psychological distinction between cognitive empathy and affective empathy. Instead, he refers to the way empathy is commonly understood within Theory of Mind research:
“The ‘theory of mind’ and ‘empathy’ so lauded in normative psychological models of human interaction, refers to the ability a ‘neuro-typical’ (NT) individual has to assume understandings of the mental states and motives of other people. When such ‘empathy’ is applied toward an ‘autistic person ’, however, it is often wildly inaccurate in its measure.”
~Damian Milton, On the ontological status of autism: the ‘double empathy problem’ 2012.
Milton concludes that the “empathy problem” is “a two-way street.” In other words, his concern is with reciprocal misunderstanding, not with defining or redefining cognitive or affective empathy.
Fellowes’ paper appears to evaluate the double empathy problem as though it were a theory of cognitive and affective empathy, whereas Milton presents it as a sociological theory of reciprocal interaction. He is asking a different question from the one Milton set out to answer. Fellowes asks whether the evidence for the double empathy problem demonstrates claims about cognitive and affective empathy. Milton, however, was primarily arguing that communication difficulties should be understood as reciprocal social interactions rather than deficits located solely within one person’s mind.
Livingston et al.
Fellowes was not the first to criticise the double empathy problem on the basis of its relationship to cognitive definitions of empathy.
Back in 2025, Livingston et al. argued that the double empathy problem is a promising idea, but that its theoretical foundations require greater conceptual clarity. They suggest that the literature discussing the double empathy problem has used the term empathy in ways that differ from its established meaning in cognitive neuroscience and psychology, while also using the theory to refer to a range of related concepts, including mutual misunderstanding, communication quality and social interaction. As a result, they argue, it is often unclear exactly what the DEP predicts.
They do not argue that the existing evidence is wrong, but that it is often uncertain which aspects of the DEP it actually supports. Until the theory and its terminology are more clearly specified, they recommend caution in drawing strong conclusions or applying the DEP to clinical practice or policy.
Irony
There is potentially a degree of irony here. It’s possible that the apparent criticism of Milton’s double empathy problem by Fellowes and Livingstone et al. may itself illustrate the kind of reciprocal misunderstanding described by the double empathy problem. Milton appears to be asking a sociological question about reciprocal interaction, whereas Fellowes and Livingston et al. approach the theory largely through the conceptual framework of cognitive psychology. If those really are different questions, then the two camps may be talking past one another rather than directly disagreeing.
Milton et al. Clarify the Double Empathy Problem
In June 2026, a new paper emerged online: What the Double Empathy Problem Is (and Is Not). In this paper, Elizabeth Kilgallon, Damian Milton and colleagues present a lengthy clarification of the double empathy problem, part of which addresses criticisms such as those from Fellowes and Livingston et al.
The authors argue that the double empathy problem has often been misunderstood, particularly when interpreted solely through the lens of cognitive psychology. They emphasise that the double empathy problem is not a theory of cognitive or affective empathy, but a broader account of reciprocal social interaction between people with different dispositional experiences.
They argue that misunderstandings between autistic and non-autistic people arise through differences in perspective and interaction, and contextual circumstances, such as social power dynamics, rather than from deficits located solely within autistic individuals. They also caution against reducing the double empathy problem to existing psychological constructs, arguing that it is better understood as a multidisciplinary framework drawing on sociology, phenomenology, disability studies and psychology, with reciprocal understanding at its core.
The paper relentlessly makes its point. One subheading is, “The Double Empathy Problem Is a Sociological, Not Social-Cognitive, Framework“, which is such a direct statement it hardly needs further explanation. It continues, “The DEP is not a social cognitive theory of autism; instead it is a sociological lens for understanding social interaction between differing people,” and “…[DEP] does not position itself as a social cognitive theory designed to explain autism ontologically.”
Later, we read: “It is named the ‘Double Empathy Problem‘ partly as a critique of social cognitive theories of autism which argue that there is a distinct social-communication deficit in autistic people that prevents autistic people from inferring the minds of others, rendering autistic people as less proficient in cognitive empathy.”
One statement that, I believe, serves as a response to critics who try to position the double empathy problem unsatisfactorily within cognitive/affective empathy models is this: “[The double empathy problem ] moves the subject of ‘autism’ into a social and sociological realm.”
There is one statement within the paper that I found absolutely startling. It’s this:
“[…] The (tongue-in-cheek) name ‘Double Empathy Problem‘ positions empathy at the forefront as a form of resistance to the dehumanization of autistic people by researchers.” This statement, perhaps more than any other in the paper, helps explain the cognitive and affective empathy-based criticisms of the double empathy problem. The theory was never intended as an account of cognitive or affective empathy; rather, the term double empathy problem was a deliberately ironic challenge to the longstanding assumption that autistic people alone have an “empathy problem”. We have returned to irony. Earlier, I suggested that the name can be misleading. Here we see exactly why. Stripped of its ironic context, it invites an interpretation that Milton never intended.
The Two-Way Street and Neuronormative Hegemony
An important point that Milton addresses in his 20212 paper, and which is expanded on in the 2026 clarification, is the issue of context in the bidirectional misunderstandings in communication between autistic and non-autistic people. In the 2012 paper, Milton frames the difference as between autistic and neurotypical (NT) people and says this:
“One could say that many autistic people have indeed gained a greater level of insight into NT society and mores than vice versa, perhaps due to the need to survive and potentially thrive in a NT culture. Conversely, the NT person has no pertinent personal requirement to understand the mind of the ‘autistic person’ unless closely related socially in some way.”
~Damian Milton, On the ontological status of autism: the ‘double empathy problem’ 2012.
This observation points to what we might call the neuronormative hegemony in social interaction styles that often leaves autistic people marginalised and othered as abnormal in a pejorative sense. The 2026 clarification goes further, stating, “We clarify the probabilistic framing of the DEP to emphasize that misunderstandings between autistic and non-autistic people shift with context and power.” The term probabilistic might seem jarring here, but the paper goes on to explain this.
It refers to the claim made in some research that communication between individuals of the same neurotype (NT-NT communication, or autistic-autistic communication) will be “superior” to communication between individuals of different neurotypes (autistic-NT communication), and makes the point that this claim is not foundational to the theory of the double empathy problem.
The clarification states that the bidirectional communication problems identified by the double empathy problem are both asymmetrical and context-dependent. It observes that autistic people will not automatically connect and empathise with each other more successfully simply because they are autistic, and the same is true of NT people. Instead, these are probable trends, but are subject to context, such as previous experience and social power dynamics.
Personal Reflection
I accept and respect the statement from the 2026 paper by Kilgallon, Milton, and colleagues, that “The DEP is not a social cognitive theory of autism; instead it is a sociological lens for understanding social interaction between differing people.” But the fact that the double empathy problem is not a social cognitive theory of autism does not mean it has no place in an overarching neuroinclusive theory of autism.
In terms of understanding autistic people on their own terms rather than as subjects of scientific conclusions based on, as the 2026 paper puts it, “weak evidence,” the double empathy problem changes the rules of the game.
That’s all for this time. Take care.
share this article:
The Autistic Writer is a labour of love, dedicated to promoting autism acceptance, understanding, and equality. If you’d like to help support this work, a small tip via Buy Me a Coffee would be greatly appreciated. Just click the button, below.

If you like this blog, I can also highly recommend the brilliant Mortgage Advisor on FIRE, which covers financial wellbeing, mental health, and autism issues, among much more.
