Is ADHD Real?

By Natalia Zdorovtsova

ADHD is not real. Or is it? You can find out the answer in a talk I’m giving at Luminomelia Cambridge on the 28th of February, 2023. The talk will be based on this article, and there will be a round-table discussion afterwards.

Art by Greg Dunn

ADHD: fact or fiction?

As a developmental cognitive neuroscientist, the goal of my research is to understand how inattention and hyperactivity, two traits that exist in varying degrees across the entire population, might coincide with differences in the structure and function of the brain. I mainly work with children who have high levels of both of these traits—those who, perhaps, might be classed as having a neurodevelopmental condition. If you fall into this category—of developing ‘atypically,’ or being ‘neurodivergent’—you can expect to encounter some challenges, particularly when you’re at school.

Neuroscientists are interested in studying neurodevelopmental differences for a number of reasons. The first, I think, is that we want to help people who are struggling. This can come in the form of developing pharmacological and therapeutic interventions, sure, but we also study how specific environmental factors—like early life adversity, ineffective learning environments, and systemic barriers—contribute to the negative outcomes that we often see among neurodivergent children. Neurodevelopmental conditions also offer us a window into what kinds of genetic differences, neurological differences, and environmental differences coincide with—or perhaps even cause—differences in cognition and behaviour. Tracking the complex interplay between genes, the brain, cognition, and behaviour can help us extract some fundamental principles of development itself, and maybe even help us understand what makes every person unique.

For those not in the know, neurodevelopmental conditions have been formally classified in a set of medical labels, like Attention Deficit Hyperactivity Disorder (ADHD), autism, dyslexia, and dyspraxia. These labels are contained in diagnostic manuals, which help clinicians make informed judgements about their patients’ symptoms. If you look in a diagnostic manual, you’ll find a list of symptoms underneath each category. In some cases, symptoms overlap across categories—for instance, ‘constant fidgeting’ and ‘interrupting conversations’ are listed as two main symptoms of ADHD, but are also present in autism as 'repetitive behaviours’ and ‘not understanding social cues.’ The core assumption here is that these similar behaviours are caused by different underlying mechanisms in ADHD and autism. In other words, we assume that people who have these conditions are fundamentally different from each other in some way, despite displaying some similarities, because there are different things causing those behaviours to arise.

In this article, I’m going to discuss the belief that a particular diagnostic label—ADHD—doesn’t constitute a real disorder. I’m not interested in ‘debunking’ this belief—I think some of the criticisms it raises, particularly of our ill-defined medical classification systems, are quite valid. I’m mainly interested in what it could possibly mean for any mental condition to be ‘real,’ and what there is to gain from knowing which ones are ‘grounded in reality.’ Who decides what’s real and what’s not? And how do we go about ascertaining the reality of mental conditions?

Before we get into these questions, let’s cover some statistics. ADHD is one of the most common neurodevelopmental disorders in childhood, and it often persists into adulthood—though the exact way in which symptoms present might change over time. The global prevalence of ADHD is estimated to be around 5%, although prevalence estimates in the United states range from 8% to 10% due to much higher rates of diagnosis. Its profile of symptoms includes restlessness, a pervasive inability to focus or complete tasks with planned intentionality, poor impulse control, memory deficits, and difficulties with perceiving and tracking the passage of time.

The fact that ADHD is so common is what makes some people question whether it’s a ‘real’ neurodevelopmental condition. If a set of traits occurs so frequently in the population, then it could be that powerful people are crafting pathologies around normal childhood behaviour in order to meet their nefarious goals. Many claim, for instance, that ADHD was invented and popularised by pharmaceutical companies in order to sell more Ritalin to unwitting families, or that ADHD is the result of poor parenting and systemic complacency in schools. These claims don’t necessarily question the existence of ADHD-like profiles of cognition and behaviour, but they certainly doubt that ADHD traits constitute a genuine neurodevelopmental trajectory—one which would arise in the absence of lazy parenting, or one that would even be regarded as problematic in the absence of pharmaceutical incentives.

For a moment, let’s assume that about 5% of children do match the clinical description for ADHD—that they’re inattentive, hyperactive, and have all of the other features that you’d associate with this condition. What reifies—or makes real—the fact that they have ADHD?

Three ways to argue that ADHD is real

I can think of three major routes that we could take in attempting to demonstrate the reality of ADHD: formal medical classification (Medical Authority Route), people’s inability to control their ADHD behaviours (‘Lack of Agency’ Route), and the presence of biomarkers which indicate that ADHD is grounded in some kind of genetic or biological materiality (Biological Essentialist Route). My argument is this: ADHD can be seen as ‘real’ in the sense that it is medically recognised, associated with a lack of behavioural control, and tied to a range of neurological and genetic differences (compared to controls). However, the issue with ADHD is that it ultimately lacks clinical specificity—it doesn’t capture a consistent phenotype, and the label isn’t applied reliably in clinical contexts because of its relatively fuzzy definition. Instead of demanding a fundamental ‘ground truth’ from our psychiatric labels, we should test how ‘real’ ADHD is based on whether the label is useful—and personally, I’m not too sure if it makes the cut. At the end of this article, I’ll discuss alternative ways of thinking about developmental atypicality, which might help us break free from our overdependence on diagnostic labels.

The American Psychiatric Association (APA) released its first Diagnostic and Statistical Manual of Mental Disorders (DSM) in 1952. This manual listed all recognised mental disorders—but ADHD was not included at the time. It was only in 1968, in the third edition of the DSM, that ‘Hyperkinetic Reaction of Childhood’ was introduced as a formal diagnostic category. In 2000, the fourth edition of the DSM recognised the label ‘ADHD’ with three subtypes: inattentive-type, hyperactive-type, and combined-type, which is the taxonomy we still use now. Obviously, medical models are constructed gradually, and new diagnostic pipelines are created as institutions adjust to new information. I’ve left ‘information’ intentionally vague here, because it’s not always entirely clear—particularly in the case of mental conditions—what motivates the inclusion of each particular diagnostic label in our formalised taxonomies. But, in any case, we could say that the codification of ADHD within a diagnostic manual lends some credential weight to its existence as a genuine condition. If you’re arguing with the average person about whether or not a particular condition ‘exists,’ my bet is that you could cite the DSM, show them that ADHD is listed in big, glossy letters, and convince them that your evidence is sufficient to show that ADHD is real—or, at least, that the argument isn’t worth continuing. The medical authority route to showing that we should take the term ‘ADHD’ seriously ultimately rests on our trust in rigorous, evidence-based practice within psychiatry. And to be honest, that’s fair enough. Centralised institutions play a huge role in allowing us to access information quickly and efficiently. The knowledge they offer is often underpinned by self-regulating processes, like science, which have a relatively high level of commitment to things like open frameworks, peer review, and experimental replication.

Of course, there are some issues here. Apply a bit of critical thinking, and you’ll notice that there are many institutions which claim to do good, rigorous, evidence-based work, but are operating alongside other interests that hinder science’s ability to moderate, check, and openly share its own findings. The scientific publishing industry is one such hindrance, because experimental results are often hidden behind a paywall, and null results—ones where an experiment’s findings don’t show any statistically significant effects—are seldom published. On the surface, this means that we might see an overwhelming number of papers being published in support of a scientific theory, which makes its predictions seem a lot more replicable, and therefore credible. We don’t become aware of all the null findings, claims to the contrary, and indications that we might be in the wrong. A few years ago, in the realm of clinical publishing, this problem reached concerning new heights: investigations revealed that only about 50% of clinical trials are published, with a heavy bias towards ‘positive’ or ‘significant’ results. You can probably imagine how this might impact our scientific theories and systems of clinical classification. Our operational taxonomies, procedures, and treatments are functioning on a relative paucity of evidence.

In the realm of classifying neurodevelopmental conditions, this can present a few problems. Even though our scientific publishing practices are becoming better, clinical manuals don’t change their definitions that quickly. Combined with various biases that exist in our diagnostic practices, as well as the fact that diagnosis is often expensive and inaccessible to those from lower-income families, this means that it can be difficult to understand what ADHD ‘really is’ out in the real world.

In other words, our understanding of ADHD is limited to our study of people who have access to a formal diagnosis. And the set of people who have a diagnosis of ADHD isn’t representative of the much larger set of people who have all the formally-recognised symptoms of ADHD, but might have additional traits that aren’t picked up on by diagnostic manuals, since those people aren’t getting diagnosed as often, and therefore aren’t informing research about ADHD. It’s all a feedback loop. As a consequence of diagnosis-based sampling in neurodevelopmental research, certain symptoms of ADHD become ‘cardinalised’ over time, while others become marginalised and forgotten. And this, in turn, constrains our clinical definitions in various unhelpful ways.

If you’re claiming that what makes ADHD ‘real’ is its recognition within diagnostic manuals, you have to consider the epistemic limitations of medical science and practice. That’s why the medical authority route to argumentation—while sensible in some ways—doesn’t quite cut it.

We can take a different approach, then, and ask questions about the symptomology of ADHD, rather than fixate on the label itself. We can also talk about what makes something a disorder, rather than just a neutral set of traits.

We tend to follow a certain narrative structure when talking about disorders of the mind, which I’m sure you’ve picked up on if you’ve read some of the works of Oliver Sacks, who is famous for captivating public audiences with his neuropsychological case-studies. A key feature of these case studies is that they demonstrate a lack of agency on the part of individuals who develop abnormal behaviours. Although their behaviours result in harm—either the direct, obvious kind of harm, or the kind that an eccentric-seeming person faces when they’re ostracised from society—they can’t simply choose to stop. There is something more innate, more neurologically and cognitively fundamental, that overrides any attempt at correcting behaviours that lead to harmful outcomes. Of course, neurobiology isn’t the only thing that characterises or produces a loss of agency in mental disorders. In her 2016 article ‘Human Agency and Mental Illness,’ Margarita Mooney writes that “…human agency can be diminished by biological illness, power structures in psychiatry, and cultural categories of mental illness diagnoses.” Sometimes, psychiatric labels and treatments can make us feel disempowered in overcoming our habits and taking control of how we behave. Nonetheless, I think that the lack of agency narrative is one which runs through our intuitive definitions of what makes a mental condition ‘real.’ It also grounds most of our moral arguments about mental disorders, which naturally follows from our desire to categorise people as ‘good’ or ‘bad’ and deliver punishment accordingly. Much of our legal discourse on whether mentally ill offenders should be punished, for instance, rests on the assumption that mental disorders reduce, or even remove, a person’s agency. Although our preference for punishment over universal rehabilitation is a feature of our society, and not of any fundamental ‘ground truth’ of morality, it affects how we come to think about mental conditions by framing them as a lack of agency.

In the case of ADHD, agency is more of a meta-issue, since poor impulse-control is one of the key symptoms. In other words, people with ADHD usually struggle with controlling their ability to control certain behaviours.

Let me give you an example. Sometimes, people with ADHD engage in deep states of sustained focus, where they look and feel very much ‘in control’ of themselves and their immediate environments. This is a state called ‘hyperfocus,’ and it shows us that ADHD isn’t just an attention deficit—there are other regulatory mechanisms at play, which is what gives ADHD both of its attentional extremes. At other times, those with ADHD might struggle to feel ‘in control’ of the things happening around them. Thoughts become rapid, jumping from one loosely-related topic to the other, and things in the external world become harder to manage. The thing is, people with ADHD can’t really choose when to engage in hyperfocus—although it comes with a profound feeling of agency, it only emerges when someone is engaged in something they’re deeply interested in. And, as anyone with ADHD will know, interests and ‘hyperfixations’ aren’t explicitly chosen (if only it were that easy). Often, hyperfixations are centred around things that seem random, inconsequential, or ‘useless’ to the outside observer, like artisan coffee, obscure historical events, or video games. There is a broader lack of agency around certain symptoms, even when those symptoms appear very agentic in the moment.

Other symptoms, like inattention, forgetfulness, and restlessness, are also incredibly difficult to control in ADHD. All of us, to some extent, experience these things. Overcoming them is usually a matter of exerting some willpower over our habits—for instance, forcing ourselves to pay attention to a very boring lecture with the knowledge that it has something useful to offer us in the future. ADHD is characterised by an inability to do this. No matter how earnestly they might try to regain focus, no amount of willpower is enough to overcome the ADHD mind’s strong, natural tendency to wander. This has nothing to do with general intelligence, and everything to do with the fact that people with ADHD exert their agency in an uneven way: by being able to hyperfocus and excel at tasks they’re drawn to, while simultaneously being unable to complete necessary tasks that appear incredibly straightforward to individuals without ADHD.

It's worth noting here that agency doesn’t exist in a vacuum. Our ability to control ourselves isn’t only guaranteed by our own mental machinery, but facilitated by a world that is capable of both empowering and disempowering us. Our expectations of ourselves are shaped, especially during childhood, by the assumptions and intentions of those around us. Having a label of ADHD can play into those assumptions, altering what beliefs a child might come to hold about themselves. Lower self-esteem and expectations from teachers can undermine performance at school, leading a child to feel an overall lack of agency and act accordingly. In my line of research, this is what we call a ‘bad outcome.’ Although many neurodevelopmental conditions come with differences in aptitudes and abilities, characterising them as innate deficits can impose harmful limitations on a person’s growth and sense of agency. Moreover, they fail to acknowledge the fact that agency develops in accordance with factors that are completely external to us.

For various reasons, both individual and societal, ADHD is often characterised by a lack of agency, which suggests that there might be an underlying reality to this diagnostic label after all. But where does this lack of agency originate? What underpins a person’s inability to engage in self-control?

In the case of Oliver Sacks’ patients, profound changes in brain structure—such as those resulting from tumours, strokes, and neurodegenerative diseases—resulted in their loss of behavioural and cognitive control. The innate biological mechanisms that typically ensure functional harmony in the brain were not enough to overcome even larger biological changes that resulted from disease or injury. Our discussions about agency therefore can be reduced, in some ways, to the biological nature of the mind.

This brings us to our last potential argument for the reality of ADHD: the biological essentialist route. Put simply, this argument rests on the assumption that there’s a clear mapping between genes, neurology, and behaviour. It proposes that behavioural profiles (like ADHD) correspond to particular biomarkers, and that ‘real’ disorders are underpinned by biologically ‘real’ causes. Optimistically, this would mean that we could go looking for ‘ADHD genes’ or ‘ADHD brains’ that enable us to diagnose the condition on the basis of something that feels more ‘objective’—like a brain scan or genetic sequence.

Unfortunately, this vision is complicated by the fact that ADHD is a psychiatric construct: a diagnostic category formed on the basis of subjective observations about behaviour and cognition in children. ADHD is also a spectrum condition, meaning that there’s a great deal of heterogeneity—or variability—among those who possess a diagnosis. ADHD is heterogeneous on the genetic, neurological, cognitive, and behavioural level. In the past, many studies have found genetic and neurological differences between people with ADHD and non-ADHD controls; however, these studies have failed to present diagnostic accuracy statistics. A clinically-useful biomarker, in practice, should have at least 80% sensitivity (the ability to detect that the disorder is there), 80% specificity (the ability to distinguish between the disorder of interest and other disorders), and be reliable and reproducible. Given a certain biological marker, can we predict whether or not someone has a diagnosis of ADHD (selectivity), rather than a diagnosis of a different condition (specificity)? In 2012, a taskforce on biological markers in ADHD found that none of the measures examined in previous studies met these criteria for clinical utility. Since then, machine learning classifiers have had limited success with identifying ADHD on the basis of brain imaging and genetic data. A recent systematic review on the matter concluded that “large multi-modal imaging datasets, and potentially their combination with other types of data, like cognitive data and/or genetics, will be essential to achieve the goal of developing clinically useful imaging classification tools for ADHD in the future.” In other words, our machine learning models currently need brain data in conjunction with genetic, cognitive, and behavioural data in order to accurately assess whether someone might have ADHD. Cognitive and behavioural assessments are already used to make diagnoses of ADHD, and at this stage, we can’t say that additional data from brain scans adds any value to that process. Indeed, another systematic review published in 2023, which included clinical and biological data from 300,000 children, found no viable diagnostic biomarkers for any neurodevelopmental condition. We simply can’t prove the reality of ADHD, or diagnose it, using brain scans and genetic sequencing.

This isn’t to say that neurodevelopmental disorders aren’t biological—of course they are. It’s more likely, however, that the label ‘ADHD’ describes a wide range of behavioural and cognitive phenotypes that aren’t captured by the same underlying genes, differences in brain structure, and patterns of brain activity. ‘ADHD,’ after all, is a diagnostic category with fuzzy boundaries—if there’s cognitive and behavioural heterogeneity within ADHD, then it’s less likely that we’ll be able to pick up on any consistent effects at the genetic or neurological level.

Recently, some colleagues and I conducted a study where we looked at heterogeneity in brain structure among children with high levels of inattention and hyperactivity, which are the two most prominent behavioural symptoms of ADHD. The results were surprising: children with exactly the same behavioural profile (one that we’d associate with ADHD) can be grouped into two distinct ‘neurotypes’ (brain types). One of the neurotypes was characterised by more efficient information pathways across brain networks, as well as greater connectivity within modular clusters of white matter. These neurotypes had nothing to do with gender or age, as we initially suspected. Instead, they were tied to cognitive ability—those with more efficient brain networks performed better on tests of visuospatial reasoning. I suspect that all 232 of the children in our study would qualify for a diagnosis of ADHD, if given the chance to see a clinician, despite having very different brains. Nonetheless, the assumption persists that one set of brain differences, or genetic differences, must explain the emergence of ADHD symptoms. It’s just not true. In fact, there can be many different genetic and neurological ‘paths’ to developing the same features of ADHD. Moreover, symptoms emerge as an individual interacts with their unique environment over time—it’s important not to get too fatalistic, or too reductionistic, about the potential causes of complex neurodevelopmental conditions.

As our machine learning algorithms, brain imaging techniques, and computational abilities improve, I’m sure that the biological classification of ADHD will become slightly more viable—but for now, any argument from the perspective of biological essentialism is insufficient in demonstrating the ‘objectivity’ of ADHD. The label ‘ADHD’ does not seem to correspond to anything consistent within the realm of biology; indeed, even its core symptoms, divorced from the biases inherent to the diagnostic process, don’t capture anything specific about the brain.

Psychiatric realism doesn’t make sense

By the end of this article, you and I will have explored three ways of arguing that ADHD is ‘real’—the medical authority route, ‘lack of agency’ route, and biological essentialist route—but they all come with their own limitations, fundamental knowledge gaps, and biases. Where does this leave us?

Maybe we’re asking the wrong questions about neurodevelopmental conditions. After all, what does it mean for a condition to be ‘real’? When no argument or explanation fully satisfies our desire for truth and clarity, it’s easy to become extremely sceptical and conclude that ADHD isn’t real—and, indeed, that no neurodevelopmental condition can be considered ‘real enough’ to fit seamlessly within our intuitions about what diagnostic labels ought to represent.

The problem is that nothing will ever be ‘real enough’ to satisfy the psychiatric realist—someone who believes that there is some sort of fundamental ‘ground-truth’ to the labels we see in diagnostic manuals. But these labels weren’t created with any ground-truth reality in mind; they were made for clinical practitioners, who do their best to funnel all of the complexities of life into a set of helpful labels.

We shouldn’t be asking if ADHD is ‘real’—we should be asking if ADHD is a useful category.

Pragmatism and the ‘Transdiagnostic Revolution’

A more pragmatist approach to psychiatry calls upon us to stop looking for any kind of fundamental ‘truth’ to the labels that we apply in the clinical space, and instead to focus our energies towards finding categories that allow patients to meet their needs. After all, that is the purpose of diagnostic categories—the reason we’re making classifications, in the first place, is so that people who are struggling can access resources that will ease those burdens.

The best possible diagnostic system is one that’s effective in identifying patients’ struggles in a way that is clear, specific, and selective. Ideally, someone would be able to walk into a clinic, get a diagnosis that accurately maps onto their habits and experiences, and receive information about what to do next. If we don’t have diagnostic labels that reliably track patients’ experiences, then that’s the fault of our categories, and not the fault of patients.

But as I mentioned previously, neurodevelopmental disorders like ADHD are extremely heterogeneous—diversity is the rule, rather than the exception, within these diagnostic categories. Children with the same diagnostic label can vary considerably in the nature, scope, and implications of their symptoms. Efforts to subtype ADHD (into its inattentive, hyperactive, and ‘combined’ subtypes) haven’t been much help: these separate ADHD presentations aren’t stable over time. If we’re really interested in what ADHD might be, we should ditch static models of neurodevelopmental conditions in favour of ones that consider the actual dynamics of how the mind changes throughout childhood. A more useful question might be “what features of ADHD are relevant at what points in development?”. Similarly, clinical approaches that acknowledge that behaviour and cognition change over time—rather than staying within the limits of a diagnostic label—might be better at allocating the right resources to the right people. To be effective, our diagnostic labels need to be developmentally-appropriate and granular.

Things are further complicated by the fact that co-occurring diagnoses and difficulties are extremely common among those with neurodevelopmental conditions. Additional learning difficulties are present in 44% of children with ADHD, and ADHD and autism have co-occurrence rates of 30-70%. Our diagnostic categories, for the most part, aren’t capturing anything particularly selective—many of them point to a very general profile of difficulties, while failing to specify exactly what an individual person needs in order to overcome these difficulties. And, because the rate at which ADHD is diagnosed is so high, the demand for general-purpose resources is becoming untenable.

On the basis of this evidence, I can’t say with certainty that ADHD is a particularly useful label. It doesn’t track how a child might change across time, it’s defined by an insufficiently specific set of symptoms, and it co-occurs with other conditions at such a high frequency that it might better be viewed as an appendage, rather than a category in its own right.

However, whether you agree with me or not, this doesn’t negate the fact that people’s experiences of neurodivergence are real. There are most certainly different trajectories of development, and unfortunately, some of these trajectories are reliably coupled with difficulties at school, work, and in interpersonal relationships. Even if our labels aren’t effective in distinguishing between forms of neurodivergence, this doesn’t preclude the possibility of creating better categories that actually capture people’s experiences in a meaningful way.

A recent review article entitled ‘The transdiagnostic revolution in neurodevelopmental disorders’ proposes a way of doing things differently. The review acknowledges that heterogeneity within neurodevelopmental conditions introduces a host of problems, both in developmental research and clinical practice. Here’s a quote I found particularly striking:

“The addition of new diagnostic options such as autism with intellectual disability or autism with language delay demonstrates the limits to which a categorical model can tolerate heterogeneity… In essence, the DSM-5 is a system used for assigning individuals to discrete diagnostic categories that appears to be straining at its limits.” (emphasis mine)

The term ‘transdiagnostic’ refers to a set of methods by which we can look beyond the diagnosis that an individual may or may not have. Instead of accepting diagnostic categories as fundamentally ‘real,’ the transdiagnostic approach to neurodevelopmental disorders seeks to understand how underlying dimensions of behaviour and cognition may contribute to differences in outcomes. These dimensions—like attention, impulse control, visuospatial reasoning, and verbal processing—are all very valuable in tracking neurodevelopmental differences between individuals. Cognitive and behavioural dimensions form a more basic, true-to-life way of looking at neurodevelopment than diagnostic labels, because they genuinely capture how a person performs in any given domain. Instead of reflecting assumptions and biases that are inherent to our diagnostic labels, they allow us to see how an individual child might be struggling, and create targeted interventions that actually help.

Dimensions are a useful way to track specific difficulties and predict future outcomes, and to me, that makes them far more ‘real’ than any diagnostic label in the DSM. I understand why labels like ‘ADHD’ exist—using diagnostic labels can expedite the process of labelling, reduce immediate costs, and give patients a term to describe their experiences. But I think that there are also costs to using labels that don’t accurately capture people’s needs and experiences. In the context of neurodevelopmental disorders, inadequate interventions can lead to a range of bad outcomes down the line—like psychiatric illness, academic difficulties, and a narrowing of opportunities. These outcomes aren’t inevitable, but I think they do result from a failure in the provision of resources to those who might need them most. Any effort to make resource provision for neurodevelopmental conditions more efficient is therefore a welcome one, and I think that dimensional, transdiagnostic approaches are the most pragmatic way to go.

In conclusion

In recent years, many have questioned the legitimacy of Attention Deficit Hyperactivity Disorder (ADHD). In this article, I covered three ways in which we could potentially demonstrate that ADHD is ‘real’: the medical authority argument, ‘lack of agency’ argument, and biological essentialist argument. Although there are merits to each of these arguments, there are also drawbacks—none of them quite satisfies our desire for proof that ADHD is either real or not real.

Instead of fixating on these questions of reality, I propose that our time would be better spent finding out if ADHD is a useful label. I define ‘usefulness’ in terms of a diagnostic category’s specificity and selectivity, which guarantee that an individual receives appropriate forms of care that are provisioned in an efficient manner. Unfortunately, ADHD does not meet my criteria for usefulness, because it fails to capture a reliable or specific developmental trajectory.

However, people’s experiences of neurodevelopmentally-relevant symptoms are still real, and clinicians have a duty to ensure the best quality of care for their patients. At the end of this article, I briefly argue that a transdiagnostic approach—one that looks at meaningful dimensions of behaviour and cognition—could be more useful than our traditional approach to categorising neurodevelopmental conditions.