Intersectional Barriers to Autism Diagnosis

Section 1

Diagnosis as Recognition and Unseeing

In the case of autism, diagnosis usually means that a clinician observes a person’s behaviour, compares it to diagnostic criteria, and decides whether the label applies. In this project, I am thinking about diagnosis as a form of recognition beyond a traditional medical label. To be diagnosed is to become visible in a particular way. It can make someone’s experiences more understandable and digestible to themselves and to others. It can open access to accommodations, services, support, and community. It can also offer language for experiences and behaviours that have been misunderstood for years. But because diagnosis is tied to institutional recognition, it is also political.

Crenshaw’s (1989) theory of intersectionality helps illustrate how diagnostic recognition is shaped by overlapping systems of power. Race, gender, class, disability, and coloniality are never considered separately, as people do not move through diagnostic systems as only one part of themselves. A person is never only autistic, or only racialized, or only gendered, or only low-income. These categories overlap and shape how their behaviour is read. This means that, in autism diagnosis, the question is not only whether autistic traits are present, but also whether those traits appear in a form that clinicians, schools, families, and institutions have been trained to recognize.

This is where Afana’s (2023) critique of “unseeing” becomes important. I understand unseeing as more than simply not noticing. It is a structured way of failing to recognize what exists in front of you. In relation to autism, many people are made invisible because their distress, communication, masking, sensory needs, or ways of coping do not match the dominant image of autism, despite those needs being consistently present. A person can be seen as anxious, dramatic, defiant, gifted, difficult, lazy, shy, or “fine,” while still not being recognized as autistic.

The significance of this distinction is that misrecognition is not the same as absence. Often, the signs can be present, but they are interpreted through the wrong framework. A child’s distress may be understood as bad behaviour, a teenager’s exhaustion may be understood as anxiety, and a high-achieving student’s burnout may be understood as perfectionism. A racialized or Indigenous child’s difference may be read through deficit, discipline, or another diagnostic category before autism is even considered. In these examples, an individual is not unseen because nothing is happening. They are unseen because the system is trained to see something else.

Hacking’s idea of classification is also applicable because diagnostic categories help organize how people are treated, what services are accessible to them, and sometimes, how they come to understand themselves. In this sense, diagnosis does not simply “discover” autism in a neutral way. It also produces a relationship between the person and the institution. This is the contradiction at the centre of the project: diagnosis can be validating and necessary, but it also depends on systems that have not recognized everyone equally or equitably.

Some people are diagnosed early because their autism fits the expected image. Others are delayed, misdiagnosed, or forced to prove themselves and their needs repeatedly, because their autism appears through masking, internalized distress, racialized assumptions, colonial misrecognition, or institutional barriers. This makes diagnosis a social justice issue rather than only a clinical one. Each section asks a version of the same question: what does a diagnostic system have to be trained to see, and what does it continue to unsee?

Section 2

The Dominant Image of Autism

If diagnosis is a system of recognition, then it is important to consider what image that system has been trained to recognize. Autism is not identified in a vacuum. Before a clinician, teacher, parent, or institution recognizes autism in a specific individual, there is often already an assumed image of what it is supposed to look like and how it should present. That image is biased, and it has been shaped by histories of autism research, clinical practice, media representation, and educational systems that have often centred young, white, male, and more visibly externalizing presentations of autism.

This dominant image becomes a diagnostic template, filtering who is most easily seen as fitting that image. When someone fits that image and its expectations, recognition happens earlier and with far less resistance. When someone contradicts that image, they run into difficulty. This is one way unseeing functions: visibility does not guarantee recognition when what is observed is filtered through a framework that already incorrectly interprets an individual’s experience.

Bargiela, Steward, and Mandy’s (2016) study of late-diagnosed autistic women shows the consequences of this narrow template. Many women in their study described being misunderstood for years and receiving other explanations for their distress before autism was recognized. Their experiences were often read through anxiety, depression, eating disorders, personality difficulties, or social sensitivity rather than autism. This does not mean their autistic traits were absent, but rather, that their autism appeared in forms that clinicians and others were less prepared to recognize.

This leads to internalized distress, careful self-monitoring, social exhaustion, and masking, connecting directly to the idea that misrecognition is not the same as invisibility. Many late-diagnosed autistic women and girls are not ignored because no one notices them struggling. Often, people do notice, but they name the struggle incorrectly. A girl who is overwhelmed may be labelled dramatic. A student who is socially exhausted may be labelled shy. Someone who is constantly monitoring themselves may be praised as mature or high-functioning. These readings of behaviour can become barriers to diagnosis because they disguise autism under more familiar labels.

However, it is also important not to treat women and girls as a complete correction to the dominant image. Lovelace, Robertson, and Tamayo (2022) describe how Black autistic women and girls have been largely missing from autism research across decades. This means that even when autism research expands beyond boys, it can still reproduce whiteness if race is not placed at the centre of the analysis. The category of the late-diagnosed autistic woman can become another limited template if it mainly reflects white women’s experiences.

Silberman’s (2016) discussion of the persistent invisibility of Black autism helps illustrate this further. Black autistic people have always existed, but dominant autism narratives have never been built to recognize them. This is not simply a problem of representation after diagnosis. It affects who is diagnosed in the first place, because diagnostic expectations are shaped by whose behaviours, families, childhoods, and forms of distress have historically counted as evidence of autism.

The dominant image is therefore not only gendered. It is also racialized, classed, and shaped by institutional ideas of normalcy. Autism may be more easily recognized when it appears in a person who is already granted innocence, credibility, and access to professional attention through their identity. In contrast, racialized, Indigenous, low-income, masked, gender-diverse, or psychiatrically labelled people may be read through deficit, discipline, trauma, defiance, or emotional instability before they are read as autistic.

The same behaviour can carry different meanings depending on whose body is performing it. Diagnostic systems have been built and trained around a narrow prototype, and people outside that prototype are made harder to see. The dominant image of autism does not just leave people out; it actively shapes the conditions under which they are believed.

Section 3

Gender, Masking, and the Cost of Being Legible

If the dominant image of autism creates a narrow template, masking is one of the ways people learn to survive outside of it. Masking, or camouflaging, refers to the work of hiding, suppressing, or compensating for autistic traits in order to appear more socially acceptable. Cook, Hull, Crane, and Mandy (2021) describe camouflaging as an exhausting process that can involve rehearsing conversations, forcing eye contact, copying facial expressions, suppressing stimming, monitoring tone, and constantly adjusting one’s behaviour in relation to other people’s expectations.

I want to understand masking as both a survival strategy and a cost of conditional inclusion. Autistic people do not mask because their autism is less real or because they are choosing to be deceptive, but because unmasked autistic traits are often punished, misunderstood, or treated as socially unacceptable. In this way, masking is not separate from the dominant image of autism. It is produced by it. When only certain presentations of autism are recognizable or acceptable, people outside that image may learn to perform in order to be safe, liked, included, or left alone.

This is especially relevant in relation to gender. Bargiela, Steward, and Mandy (2016) show that many late-diagnosed autistic women spent years managing their behaviour through self-monitoring before autism was recognized. Their participants often described trying to copy others, hide confusion, and appear socially competent, even when doing so was physically and emotionally taxing. As a woman on the spectrum, this is very reflective of my own experience as well. This connects to the previous section because masking can make someone appear to contradict the dominant image of autism; if a person seems socially fluent or high-achieving, the amount of work required for them to appear that way is ignored.

The problem is that masking also changes how distress is interpreted. When someone has spent years performing, their burnout may be misread as anxiety, depression, perfectionism, laziness, or emotional instability rather than as the cost of long-term camouflaging. Cook et al. (2021) connect camouflaging to exhaustion, anxiety, depression, and burnout, which shows that masking is not a harmless social skill. It becomes a form of hidden labour that makes an individual more digestible to others, while making their own needs harder to believe.

Milton’s (2012) “double empathy problem” helps further this point. If communication difficulties between autistic and non-autistic people are relational, rather than simply caused by autistic deficit, then masking reveals an unequal burden. Autistic people are often expected to do most of the translating: to make their tone, face, body, sensory needs, and communication style easier for neurotypical people to understand. The work of mutual understanding becomes one-sided. The autistic person becomes responsible for making themselves readable to a world that rarely has to do the same in return.

This is why I think of masking as a cost of being legible. Legibility can sound positive because it means being understandable to others, but in practice it can require a person to become smaller, quieter, more controlled, and more institutionally acceptable. The masked person may be rewarded for appearing “mature,” “easy,” “high-functioning,” or “not that autistic,” but those rewards can also become barriers. If people only recognize struggle when it becomes visible in the expected way, then the person who has learned to hide struggle may not be taken seriously until they are already burnt out.

This section also sits closest to my own experience, although I do not want the project to become only autobiographical. Diagnosis can bring recognition, but it does not undo the years before recognition happened. Being understood through other labels first can shape how you learn to explain yourself, how much you trust your own needs, and how much energy you spend trying to become easier for systems to read. For me, masking is not just an individual coping mechanism. It is a response to a world where support often comes only after someone has already become exhausted enough to prove they needed it.

Masking therefore complicates the idea of recognition. It shows that being included is not always the same as being understood. Sometimes inclusion depends on performance: appearing calm, capable, polite, flexible, and socially acceptable, even when that performance is unsustainable. In this sense, masking is not only about hiding autism. It is about how institutions and social spaces teach people which parts of themselves are safe to show.

Section 4

Race, Coloniality, Misdiagnosis, and Life Beyond Western Recognition

The previous sections focused on the dominant image of autism and the labour of becoming legible to that image. This section extends that argument by asking what happens when the diagnostic system itself is shaped by racial and colonial histories. Race and coloniality have a major role in defining whose behaviour is interpreted as disability, whose distress is treated as credible, and whose ways of knowing are permitted.

Gerlach, Matthiesen, Moola, and Watts (2022) are central to this section because they examine autism and autism services for Indigenous children and families within the settler-colonial context of Canada. Their review identifies a lack of population-level data, limited research on Indigenous autistic children, unmet health, social, and educational service needs, concerns about standardized diagnostic tools, and the overrepresentation of Fetal Alcohol Spectrum Disorder in discussions of Indigenous children. This often redirects a child into a different system of assumptions, services, stigma, and institutional response.

This connects back to unseeing, but in a more specifically colonial way. Indigenous autistic children may not be unseen because no one notices difference or distress. Rather, they may be seen through a colonial framework that already expects deficit, damage, family dysfunction, or pathology. When autism is interpreted through FASD, behavioural problems, or social risk before autism is even considered, children are misread through histories of colonial surveillance and institutional control. In that sense, misdiagnosis is part of a longer pattern in which Indigenous children and families are made legible to institutions through deficit.

At the same time, I do not want this section to only describe harm. That would keep the analysis trapped inside the same Western diagnostic framework I am trying to critique. Gerlach et al. (2022) do not simply call for Indigenous children to be added into existing autism systems as they are. They emphasize the importance of Indigenous-led research, culturally safe services, and approaches that respond to Indigenous families and communities on their own terms. This distinction is important because inclusion into a colonial system is not the same as transforming the assumptions that structure that system.

Ineese-Nash (2020) deepens this point by arguing that disability can function as a colonial construct when it is separated from culture, kinship, land, and community membership. Western disability frameworks often locate disability inside the individual bodymind, as a deficit to be classified, treated, or managed. Indigenous relational understandings may instead place children within networks of responsibility, belonging, family, land, and community. This changes the question: instead of only asking how Indigenous autistic children can be recognized more accurately by Western diagnostic systems, we can also ask what forms of recognition, value, and care already exist beyond those systems.

This is where I want to move beyond misdiagnosis and toward the idea of life outside Western recognition. Indigenous autistic people should not appear in this project only as examples of exclusion, lack, or diagnostic failure. They also show the limits of treating Western diagnosis as the only pathway to being understood. If colonial systems define value through independence, productivity, self-regulation, and institutional manageability, then existing outside those expectations can be read as deficiency. But from a relational perspective, a person’s value does not have to be proven through those terms. Care can begin from belonging, responsibility, and relation.

The First Nations Mental Wellness Continuum Framework also helps give language to this shift. Although it is not specifically about autism diagnosis, it understands wellness as connected to culture, language, Elders, families, creation, belonging, meaning, purpose, and community. This is useful because it shows how support can be imagined through a broader network of relations rather than only through an individual diagnosis. It also challenges the idea that care begins only after a professional category has been assigned. In this project, I use that framework carefully, not as a replacement for autism-specific services, but as an example of how Indigenous-led approaches can centre culture and community rather than deficit.

This also connects to the racial invisibility discussed in the Dominant Image section. Lovelace, Robertson, and Tamayo (2022) show that Black autistic women and girls have been largely missing from autism research, while Silberman (2016) describes the persistent invisibility of Black autism. Read together with Gerlach et al. (2022) and Ineese-Nash (2020), these sources show that racialized and Indigenous autistic people are not simply overlooked by accident. Their absence from dominant autism knowledge is produced through systems that have treated whiteness, settler norms, and Western medical authority as neutral.

The point is not that diagnosis has no value for racialized or Indigenous autistic people. Diagnosis can still provide access to services, accommodations, language, and support. But if diagnosis is the only way care becomes legitimate, then people are forced to seek recognition through systems that may already misunderstand them. A more just approach would not only ask how to diagnose better. It would also ask who gets to define disability, whose knowledge counts as care, and how support can be led by the people and communities most affected.

Section 5

Institutions, Access, and the Managed Mind

What does it mean to be “seen” by a system that requires paperwork before it will respond? Even when a person, family, teacher, or doctor begins to suspect autism, diagnosis still depends on moving through institutions that control assessment, documentation, accommodations, and services.

Access to diagnosis requires resources that are unevenly distributed: time, money, referrals, transportation, language, advocacy, professional trust, and the ability to keep navigating systems even after being dismissed. Aylward, Gal-Szabo, and Taraman (2021) use a biosocial-ecological framework to show that autism diagnostic disparities are produced through layered social and systemic factors in addition to individual clinical differences. Wiggins et al. (2020) similarly show that documented autism diagnosis is shaped by demographic, individual, and service-related factors, including race, timing of developmental concerns, timing of evaluation, support needs, and educational classification.

Documentation is often treated as neutral proof, when it is actually the result of access. A person may have autistic traits and still not have the paperwork required to be believed by a school, university, workplace, or healthcare system. Someone with financial resources may be able to pursue a private assessment. Someone with an informed parent or advocate may be more likely to receive referrals. Someone whose distress is legible to teachers or clinicians may be flagged earlier. In contrast, someone who is racialized, low-income, masked, Indigenous, gender-diverse, or already labelled through another diagnosis may be redirected away from autism before any assessment can even take place.

This is where institutions become gatekeepers of legitimacy. Diagnosis can open doors, but the doors often require the same resources that marginalized people are less likely to have. This creates a cycle where people without documentation are treated as less credible, while the barriers that prevented documentation in the first place remain invisible.

Titchkosky and De Welles’ (2021) concept of “university managed minds” helps me think about how institutions respond once distress appears inside them. They argue that universities often convert student struggle into an individual mental health problem to be documented, counselled, and managed, while leaving the institutional and colonial conditions of that struggle mostly unexamined. I extend this argument to autism diagnosis and documentation more broadly. Institutions may appear to provide support through counselling, wellness systems, accommodation letters, and disability services, but these systems can also individualize what are actually structural barriers.

Documentation can be both protective and necessary. It can make needs harder to dismiss and can create access to accommodations that may otherwise be denied. But the contradiction is that people are often required to translate their experiences into professionalized language before institutions will change anything for them. A person’s needs may only become legitimate once they are recorded in the right format, signed by the right expert, and submitted through the right office.

Then, access becomes bureaucratic legibility. The person must become a file, a report, a case number, an accommodation letter, or a documented exception. The institution can then respond to the individual without necessarily questioning the structures that made support difficult to access in the first place. A student who is struggling may be offered counselling, but not a more accessible classroom. A person who is burnt out may be asked for more documentation, but not met with more flexible systems. The individual mind is managed, while the institution remains positioned as neutral.

This connects back to Afana’s idea of unseeing, as institutions can claim to see and support disabled or neurodivergent people while still failing to recognize the conditions they create that produce exclusion. They may see a student’s anxiety, a child’s behaviour, or a patient’s burnout, but not the racism, ableism, coloniality, cost, waitlists, inaccessible expectations, or bureaucratic exhaustion surrounding them. Diagnostic recognition does not automatically become care, since even after autism is suspected or named, people must still move through institutional systems that shape what diagnosis means in practice.

Section 6

Moving Toward Relational Care

The previous sections have traced how autism diagnosis can become a system of recognition, misrecognition, masking, racialized and colonial exclusion, and institutional gatekeeping. This final section asks what might come after that critique. If diagnosis can open doors, but those doors are unevenly built, guarded, and accessible, then the answer cannot only be better classification. It also has to involve a different understanding of care.

I do not want to end by rejecting diagnosis, since it can give language to experiences that were previously misunderstood, create access to accommodations, connect people to community, and help people reinterpret years of struggle with more compassion. For many autistic people, diagnosis can become a meaningful form of self-recognition. The main issue is that care is often made conditional on diagnosis, and diagnosis itself is shaped by intersecting systems of power that are flawed and unjust.

This is why relational care is a necessary final concept for this project. Relational care shifts the question away from only asking, “Does this individual fit the category?” and toward asking, “What does this individual need in order to belong, communicate, participate, and thrive?” It begins from the understanding that people exist in relation to families, communities, histories, lands, schools, bodies, and systems of power.

Rice, Dion, and Chandler’s (2021) work on decolonizing disability through activist art helps to guide me through this shift. Their discussion of disability, Indigeneity, non-normativity, land, language, embodied knowledge, and activist art offers a way of understanding disability beyond Western clinical classification. Their work is also a very important influence for my project because it shows that art is more than just a representation of theory after the fact. It can also be a method of thinking, refusing, remembering, and imagining otherwise. The visual components of my project tie into how I am trying to think beyond the limits of a diagnostic file.

Ineese-Nash (2020) also touches on the significance of relational care. If disability can function as a colonial construct when it is separated from culture, kinship, land, and community, then care cannot only mean placing someone into the correct Western category. Care has to include the relations that make life meaningful, and ask what forms of belonging already exist outside institutional recognition. We know the many limits of a system that treats Western diagnosis as the only path to being understood.

The First Nations Mental Wellness Continuum Framework’s emphasis on culture, language, Elders, families, belonging, meaning, purpose, and community helps show how care can be imagined through relationship rather than only individual treatment. I use this framework carefully, not as a substitute for autism-specific services, but as a reminder that wellness and support can begin with connection.

This final section also returns to masking and institutional access. If autistic people are expected to become readable, acceptable, and professionally documented before their needs are believed, then relational care asks something different. It asks institutions and communities to become more responsive instead of requiring individuals to endlessly translate themselves. It asks what would change if support was built into environments from the beginning, rather than treated as an exception granted only after proof. It asks what care could look like if the burden of legibility did not fall so heavily on the person already burdened with navigating a neurotypical world.

Relational care means refusing to let diagnosis become the only doorway to support, and understanding that an individual’s needs can be real before they are documented, that burnout can be meaningful before it becomes a crisis, and that neurodivergent individuals should not have blame placed on them for communication differences. It also means recognizing that no diagnostic system, no matter how improved, can fully capture the complexity of a person’s experience and needs.

This is where the project comes back to the central question: how do intersecting systems shape who is diagnostically recognized and supported? The answer is that recognition is never just about seeing. It is about power, who is believed, who is defined correctly, who has access to documentation, whose knowledge counts, and whose ways of being are accepted. Diagnosis can be one important form of recognition, but recognition alone is not care.

A more just approach would still improve diagnostic access, especially for those who have been historically misread and excluded. It would also move beyond the idea that people must first prove themselves through Western categories before they receive support. Relational care asks for systems that are flexible before crisis, communities that value people before classification, and forms of support grounded in accessibility, self-determination, and repair. It requires changing the conditions under which people have to fight to become visible to begin with.