Play can boost relationships…but it won’t ‘reverse’ autism


Elise Stirling responds with a neurodiverse perspective to the idea of autism being a reversible condition.

Recently, headlines such as “Early baby therapy could reduce autism diagnoses“ and “Early signs of autism can be reversed if babies are offered therapy” were popping up all over the internet, following a study of play interventions for parents of autistic children (Ed Note. See our News Section for more on this story). As an autism specialist, an autistic individual, and mother of autistic daughters, this had me concerned. Being autistic is not something needing ‘reversal’ or ‘reduction’ – it is a natural, albeit different, state of being and exists within the person from pre-birth.

Autism has historically been a medicalised condition that requires an extensive assessment and diagnostic procedure, using criteria that (until recently) had not had any involvement from those who were themselves autistic. It is hoped that in the future, any changes to diagnostic criteria will be made with the full engagement of autistic individuals, to fully encapsulate the diversity of human minds (or ‘neurodiversity’) in a more positive light, whilst recognising the true challenges faced. Autism is simultaneously a biological reality and a social construct, and this should be recognised throughout the process of diagnosis, education, and care of autistic individuals. As a biological reality, there are differences in how autistic brains and bodies work. Porges’ Polyvagal theory proposes that there is a potential difference in the information flowing from the brain to the body in autistic individuals because of differences in the brain’s recognition of safety versus threat (neuroception). Some autistic individuals appear to live in perpetual fear of their environment and social experiences, obstructing their ability to perceive, understand and react in a way that an otherwise regulated person might. In safe contexts, the nervous system’s state is ‘soothed’, and the body becomes more regulated. The default state for autistic individuals, however, is more dysregulated, or ‘mobilised’, resulting in fluctuating social communication, irregular eye gaze, changes in receptive language ability, fewer externalised social gestures, and differences in social behaviours. This can account for individualised, context-based, levels of ‘functioning’ depending on the environment and experiences at the time.

I have proposed research in an education context that considers feelings of safety, regulation and authenticity at school. Although autism is not caused by relationships or lack thereof (the age-old ‘refrigerator mother’ argument), quality attachments can potentially have a positive impact on inclusion in schools.

The quality of the relationship between a child and their teacher is key to this, and emotional security in such relationships are central to a child’s later functioning. Forming attachments to key adults is possible, and is still beneficial to a child’s wellbeing at any stage of development. Attachment is a basic human need and something that all humans have a propensity and a capacity for, but this can be interrupted by external factors such as too much stress in the environment. It can be established or restored at any age, as long as the conditions are favourable. It follows that even with a dysregulated autistic individual, healthy attachments can be developed if the stressful environment that dysregulated them in the first place is removed or lessened to the point of comfort and security. It is never a removal of the ‘impairments’ (or cure, or prevention) of autism that I advocate for, but a removal of the negative stimuli that are causing the dysregulated state, before engaging in the development of attachment relationships.

The neurodiversity movement challenges us to rethink autism through the lens of human diversity. As part of this movement, autistic people have come to analyse autism and often arrive at very different conclusions about autism than the non-autistic professionals, including the development of play in children. The play behaviour of autistic children may be described by some as atypical and is often subject to scrutiny, where those involved have attempted to re-orientate these behaviours to something they deem more appropriate. The repetitive behaviours and ritualistic mannerisms, or stereotypies, have been attributed to many potential deficits, delays and difficulties in the past, but few people have considered these actions within the notion of play. It may be that these are not signifiers of delayed development, deficits, or otherwise lacking, but as playfulness and individuality which could be embraced by those around them instead of discouraged or extinguished. The concept of neurodiversity is not a new one, but it has evolved to become a description of difference in neurocognitive function, and includes neurotypical (as in, not significantly different from the norm in terms of cognition) and neurodivergent individuals, who are rendered different from the norm by differences in cognition or a condition that places them out with the neurotypical range of function.

Part of this movement is encouraging parents and educators of autistic children and young people to use responsive tactics, learning to speak the child’s language, so to speak, communicating on their terms and steering them away from dangerous treatments and therapies that can cause more harm than good. We can use this model to allow authenticity and individualisation during play activities and relationship building, with the aim of more successful inclusion being possible in schools.


Despite many ‘interventions’ for autism reportedly being play-based, they can still be influenced by a deficits model that may contain potentially problematic language and ideas such as ‘symptoms’, ‘disorder’, and ‘treatment’. Some may go as far as saying that autistic children are ‘missing out’ and not achieving vital milestones, although these concepts are based on typical, or non-autistic, development. The proposal of a play-based support for autistic individuals is desirable, but I would rather advocate for such a mechanism that is based on autistic norms and not limited by neurotypical expectations. For example, while one example I read in a recent article did acknowledge a more inclusive notion of play as one that reframes the solitary behaviour of autistic children (eg. restricted, repetitive behaviours) as play, it still presumes autistic individuals to be atypically developing on a normative scale, and that they need to be put on the correct pathway to be successful. I do not support this presumption, and advocate for a more individualised pathway through attachment play.

Attachment Play, popularised by Margot Sunderland, happens when an adult and child face each other to play where the relationship, not toys, is the focus. It enhances confidence, social skills, ability to trust, and brain development, particularly the frontal lobe and hippocampus, and Denice Tulloch- Johnson (who works in partnership with Margot Sunderland on attachment play in the early years) states that one to one attachment playtimes at home and at school are just as important for older children as for infants in terms of social, emotional and brain development. Attachment play optimises the chances of conversation and real ‘moments of meeting’ between adult and child. According to Margot Sunderland, science shows that attachment play helps develop higher brain function and may trigger new brain cell growth, as brains may have been previously damaged by stress such as shouting, shaming or negativity toward the child. Attachment play activates anti-anxiety and anti-stress hormones and can repair the damage done and the more these are activated, the more the brain benefits from them. Attachment play can create interpersonal synchrony: a delicious feeling of oneness with the other and it has a unique impact on the brain according to Denice Tulloch-Johnson.

Because of this, attachment play is going to be the focus of my doctoral research at which point I will deliver an attachment play package I have designed specifically for autism. It will encompass not only the important factors of attachment play itself, but caters for autistic individualism and authenticity, and provides a uniquely holistic ‘autism-friendly’ programme, pragmatically based on a neurodiversity model of autism in schools. If the conditions (environment, freedom of expression, acceptance, and so on) are favourable then there is a real opportunity for autistic individuals to match or even outperform their non-autistic peers. However, dismissing autistic approaches to communication or social-emotional aspects of life as inappropriate or wrong is failing to recognise them as important strategies for success, and it is vital to promote experiences that are compatible with their dispositions, without being forced to mimic non-autistic behaviours which could be potentially damaging. It is hoped that this research will show ways to support autistic individuals in school, without the unnecessary implications of uncomfortable and damaging neurotypical expectations. I would also like to think that, one day, we can live without such damaging headlines and change society’s views of autism to a more positive, neurodiverse frame of mind.

Elise Stirling
Author: Elise Stirling

Elise Stirling
+ posts

Elise Stirling is an autistic mother of autistic daughters, and an autism specialist teacher. She is currently studying for her PhD in Applied Autism Research. Elise is a passionate advocate for autistic, neurodiverse and disabled individuals, with a passion for parent activism and disability rights.


Please enter your comment!
Please enter your name here