Rates of autism diagnosis continue to rise, with an increasing number of diagnoses made in adulthood. ^{Reference Russell, Stapley, Newlove-Delgado, Salmon, White and Warren1} Although many report receiving a diagnosis as a positive milestone, the diagnostic process can be challenging owing to the pathologising and deficit-based language of diagnostic criteria, including ‘deficits in social-emotional reciprocity’, ‘abnormal social approaches’ and an ‘absence of interest in peers’. ^{Reference Kiehl, Pease and Hackmann2}

Psychology, linguistics, neuroscience and philosophy have led a recent shift in autism research. Instead of autistic communication being studied at the individual level, interactions are examined at the dyadic level within a ‘double empathy’ framework,^, which refers to the ‘disjuncture in reciprocity between two differently disposed social actors who hold different norms and expectations of each other,’ such as autistic and non-autistic people. ^{Reference Milton, Heasman, Sheppard and Volkmar3} By comparing how autistic and non-autistic people interact with people of the same (i.e. autistic pairs, non-autistic pairs) and different (i.e. one autistic person, one non-autistic person) diagnostic statuses, we can learn more about how the diagnostic status of an interaction partner shapes the way that people communicate.

Research shows two key findings. First, communication difficulties are not exclusive to autistic people; non-autistic individuals also struggle to understand autistic social behaviours, leading to a ‘double’ problem in mixed-neurotype interactions. ^{Reference Edey, Cook, Brewer, Johnson, Bird and Press4} Second, autistic people prefer interacting with other autistic people and share information just as effectively with other autistic people as do non-autistic people with other non-autistic people. ^{Reference Crompton, Ropar, Evans-Williams, Flynn and Fletcher-Watson5} However, information-sharing breaks down in mixed-neurotype interactions. ^{Reference Crompton, Ropar, Evans-Williams, Flynn and Fletcher-Watson5} Autistic people disclose more to, feel closer to and empathise more with other autistic individuals than with non-autistic people. ^{Reference Komeda, Kosaka, Fujioka, Jung and Okazawa6}

What does this mean for autism, as clinically defined by deficits in social and communicative behaviours? The current research shows that many social difficulties experienced by autistic people are not due to an individual deficit or an intrinsic autistic failure of social behaviours, but rather a mismatch between the way that autistic people behave, experience the world and communicate and the way that the non-autistic majority do. Autistic people can interact very well with autistic others, and, during autistic interactions with non-autistic people, both parties experience similar difficulties. ^{Reference Crompton, Ropar, Evans-Williams, Flynn and Fletcher-Watson5} Understanding autism social difficulties contextually rather than individually is important for three reasons. First, for autistic self-understanding: the way autistic people interact is not deficient, it is different to that of the non-autistic majority. This understanding can reduce self-stigma and help autistic people find ways to communicate that work for them, without resorting to masking their autistic behaviours to adhere to non-autistic norms, which can be damaging to their mental health. Second, for designing support systems for autistic people: encouraging ways to communicate that support autistic communicative styles and provide opportunities to engage with other autistic people. Finally, for non-autistic people: to understand that the non-autistic way is one way of communicating, but it has limitations. It is not superior to autistic ways of communication (as indicated by studies of how autistic people share information and build rapport with other autistic people just as well as non-autistic people do with non-autistic people ^{Reference Crompton, Ropar, Evans-Williams, Flynn and Fletcher-Watson5} ), merely different. Rather than the myriad ‘social skills’ interventions that encourage autistic people to learn non-autistic social norms, it is time to consider how non-autistic people can meet autistic people half-way.

How can these findings be embedded into clinical practice? First, it is important to consider the pathologising nature of language used in diagnosis ^{Reference Bottema-Beutel, Kapp, Lester, Sasson and Hand7} and understand that although autistic and non-autistic communication styles may be different, that does not mean one is superior to the other.

Second, given the relative ease with which some autistic people interact with other autistic people, peer support has been suggested as a promising model for post-diagnostic support. ^{Reference Crompton, Hallett, McAuliffe, Stanfield and Fletcher-Watson8} Although post-diagnostic peer support pilots have shown promising results, provision is geographically inequitable. ^{Reference Crompton, Hallett, McAuliffe, Stanfield and Fletcher-Watson8} However, encouragement to engage with an autistic community after diagnosis may be beneficial for autistic adults ^{Reference Watts, Crompton, Grainger, Long, Botha and Somerville9} in the absence of formal peer support.