We enthusiastically support Jaswal & Akhtar's (J&A's) argument that autistic people show their desire to engage with others in unconventional ways. Social psychology has long established that actors and observers may have differing perspectives on the extent to which behavior is social (Jones Reference Jones1976). J&A demonstrate this insight in the context of autism: Low levels of eye contact, infrequent declarative pointing, motor stereotypies, and echolalia are often misconstrued as social disengagement by observers.

But J&A fail to question the corollary of their insight: Do autistic people make the same assumptions and read autistic behavior in the same way as neurotypical observers? Much of the literature reviewed by J&A is focused on interaction between autistic and non-autistic people. Indeed, their core argument is that non-autistic people may misconstrue the social significance of certain behaviors displayed by autistic people. We suggest that examining how autistic people appraise autistic behavior can provide solutions to improving neurotypical-to-autistic interaction.

Autistic testimonies indicate that autistic people share better social alignment with other autistic people than non-autistic people (Chown Reference Chown2016; Milton et al. Reference Milton, Heasman, Sheppard and Volkmar2018). Perceiver-target familiarity facilitates shared understanding (Ickes Reference Ickes1993), and research has shown autistic people interpret and understand the narratives of other autistic characters better than typically developing characters (Komeda et al. Reference Komeda, Kosaka, Saito, Inohara, Munesue, Ishitobi, Sato and Okazawa2013). Contrary to the observation that “little consideration has been given to how (mis)interpretations of autistic behavior may contribute to those problems” (J&A, sect. 5.2, para. 7), empirical studies have identified a lack of experiential overlap (Heasman & Gillespie Reference Heasman and Gillespie2018b) as a contributing factor to why non-autistic people encounter difficulties in understanding autistic perspectives (Chambres et al. Reference Chambres, Auxiette, Vansingle and Gil2008; Edey et al. Reference Edey, Cook, Brewer, Johnson, Bird and Press2016; Sasson & Morrison Reference Sasson and Morrison2017). However, despite evidence on perceiver-target familiarity influencing social understanding, autistic-to-autistic social interactions have received relatively little consideration.

Motivated to find out how autistic people interact with each other in ideal conditions, we recently published a study of autistic people jointly playing video games, a popular naturally occurring social activity (Heasman & Gillespie Reference Heasman and Gillespie2018a). Across 20 gaming sessions involving 30 autistic adults, we observed the types of behavior identified by J&A, such as echolalic speech where participants would repeat sounds made by characters in the game, often disrupting the flow of conversation. But we also observed that these behaviors were not interpreted as indicative of social disengagement by autistic partners. In addition to echolalic speech, autistic partners tolerated behaviors such as shouting, obscure topic shifts, long detailed monologues, ignored turns in speech, and unreciprocated jokes – behaviors that non-autistic people might construe as disengaged or disruptive.

Autistic-to-autistic interactions, we suggest, are less constrained by neurotypical conventions; revealing a different type of sociality, one that permits periods of incoherent and fragmented dialogue in favor of pockets of intense rapport, reciprocation, and humor. Autistic-to-autistic interactions reveal a dynamic in which actors can explore personal and specific interpretations about the situation within a norm of reduced social judgment regarding the behaviors identified by J&A. Interestingly, the relaxed social norms that we observed did not seem to constrain effective interaction: 19 of 20 gaming sessions were completed efficiently. Indeed, research on creativity shows that the ability to quickly move past potential misunderstandings is an adaptive feature of creative collaboration (Hawlina et al. Reference Hawlina, Gillespie and Zittoun2017).

Examining autistic-to-autistic interactions reveals a well-developed set of social norms for handling and interpreting the very behaviors that J&A identify as potentially misconstrued by non-autistic people. Accordingly, we suggest that a solution to the problem identified by J&A might already have been found by autistic people, namely, broadening the norms around expected communication styles.

Taking seriously the idea that autistic people might not only be the source of ostensibly problematic social behavior, but also provide insights into how to appraise this same behavior, raises methodological questions. Namely, how do researchers challenge neurotypical assumptions embedded in the research process itself? Valuing autistic testimony, as J&A demonstrate, is one important step to toward this goal. However, the suggestion that “research is also urgently needed to identify and characterize the range of behaviors that can signal social interest” (J&A, sect. 5.2, para. 6) risks reinforcing the same issue that J&A argue against: that social behavior cannot be interpreted from the external observer view of the researcher alone. What counts as signaling social interest will depend on whether one is an autistic actor or a neurotypical observer. A more interactional intervention would be re-educating neurotypical people through learning from autistic social appraisal, so as to diversify neurotypical understanding of autistic social behavior. This is why more autistic involvement in research (Fletcher-Watson et al. Reference Fletcher-Watson, Adams, Brook, Charman, Crane, Cusack, Leekam, Milton, Parr and Pellicano2018; Pellicano et al. Reference Pellicano, Dinsmore and Charman2014b), such as seeking autistic consultation on research proposals, conducting autistic interrater reliability for data gathered, and supporting autistic academics is important; the literature also needs to diversify its understanding of social behavior.