ADHD and the unseen mind · Article 3 of 6
After the broadcast: what the scan argument can actually support
The documentary made this question concrete through Professor Katya Rubia’s discussion of MRI research. The important scientific point was that average differences between groups are small, distributions overlap substantially, and current imaging cannot diagnose ADHD in an individual. As Article 2 argues, the difficulty is the inference drawn from that fact. A biomarker need not classify individuals accurately for a clinically meaningful, partly biological syndrome to exist.
Cross-cultural epidemiology also makes a simple “modern Western life created ADHD” account difficult to sustain. A worldwide meta-regression of 102 studies found a pooled childhood prevalence of about 5.3 per cent and concluded that much of the variation between studies was explained by methodology rather than geography. A later meta-regression similarly found no evidence that prevalence measured with standardised procedures had increased across three decades. More recent reviews produce somewhat different absolute estimates, which reinforces rather than weakens the methodological point: criteria, informants and impairment thresholds matter greatly, but ADHD-like syndromes are not confined to one contemporary culture.
That does not make culture irrelevant. Culture and environment help decide which behaviours are noticed, how impairing they become and where a clinical threshold is placed. It does mean that screens, rigid schooling or modern lifestyles are better treated as possible modifiers and amplifiers than as a sufficient explanation for the existence of the phenotype.
Additional evidence: Polanczyk et al., worldwide prevalence and meta-regression, American Journal of Psychiatry 2007; Polanczyk et al., updated three-decade prevalence meta-regression, International Journal of Epidemiology 2014.
Article 3 of 6
ADHD cannot be diagnosed with a brain scan. Its traits lie on continua, and environments can determine how disabling they become. None of that makes the condition imaginary. “Neurodevelopmental disorder or social construct” is a false choice.
Audio edition
An audio version is planned. The player and episode notes can be added here without changing the article’s permanent link.
There is a version of the sceptical argument about ADHD that deserves to be taken seriously.
Human attention is not divided into two natural kinds. Children vary in their capacity to remain still, inhibit impulses and sustain effort. Adults vary in organisation, working memory, restlessness and tolerance of delayed reward. Sleep, stress, anxiety, depression, trauma, substance use, physical illness and digital interruption can affect all of those capacities. Schools and workplaces impose very different demands on them. No blood test, scan or single cognitive task can determine whether one person has ADHD.
Those are not heretical observations. They are part of competent assessment.
The error begins when uncertainty about an individual diagnosis is treated as uncertainty about whether a meaningful disorder exists. Pemberton has written that “even the existence of childhood ADHD as a condition is up for debate”.1 Channel 4’s publicity asks whether ADHD is a “genuine neurodevelopmental disorder, or a social construct”.2
That opposition sounds decisive only because neurodevelopmental is made to mean objectively biological while constructed is made to mean invented. Neither meaning is adequate.
What a brain scan can — and cannot — show
Large imaging studies have reported average differences between groups with and without ADHD, including small differences in some subcortical volumes and measures of cortical structure or function.3 The size and consistency of findings vary. Large samples are needed precisely because the signal is subtle and individuals differ.
A group difference is not an individual diagnostic test. If two distributions have slightly different averages but overlap extensively, the difference may help us understand populations while doing little to classify the person in front of us. Recent reviews describe many structural effects in ADHD as small — often standardised differences around 0.1 to 0.3 — and not clinically diagnostic.4
This should restrain advocacy. There is no single “ADHD brain”, and “differently wired” can imply a biological uniformity that the evidence does not support. A colourful scan cannot authenticate somebody’s diagnosis. Commercial claims to diagnose ADHD from imaging offer more certainty than current science permits.
The reverse inference is no better. The absence of a sufficiently sensitive and specific scan does not demonstrate the absence of a disorder. Migraine and many psychiatric conditions are diagnosed clinically; pain has no scan that verifies one person’s experience. Even where biological mechanisms are known, diagnosis often depends on a pattern of development, symptoms, function and exclusion of better explanations.
For ADHD, the relevant pattern includes childhood onset, persistence, symptoms across important settings and significant impairment, supported by collateral or documentary evidence where available. No component is infallible. A converging history can still provide useful clinical knowledge.
Genetics: strong evidence, limited individual prediction
ADHD runs in families. A review of family, twin and adoption research estimated heritability at about 74 per cent.5 A major genome-wide association study identified 27 statistically significant risk loci and 76 candidate risk genes, with enrichment in genes expressed during early brain development.6 More recent work has also identified rare variants associated with substantially higher risk in a minority of people.7
Together, these findings are powerful evidence that inherited biology contributes to ADHD.
They are often misunderstood.
Heritability is not the proportion of one person’s condition “caused by genes”. It estimates how much variation in a population, living within its particular range of environments, is statistically associated with genetic variation. It can change across places and time. High heritability does not mean that a trait is fixed, untreatable or uninfluenced by context.
Nor is there one “ADHD gene”. Common genetic risk is highly polygenic: many variants each alter probability by a small amount and overlap with risks for other traits and conditions. Rare variants may carry larger effects but account for only part of the picture. Neither a common-variant polygenic score nor current genetic testing can diagnose routine ADHD in an individual.
The evidence supports biological contribution, not genetic destiny or a laboratory test.
A category can be constructed without being fictional
The traits associated with ADHD are continuous. The diagnosis is categorical because clinicians and systems must decide when a pattern warrants a clinical name, treatment or particular support.
That threshold is constructed: experts define criteria, decide how much impairment matters and revise classifications as evidence and values change. It is not written into nature as a sharp border. But the fact that a category is made for a purpose does not make the phenomena it organises unreal.
Blood pressure is continuous; hypertension remains a useful category. The chosen threshold reflects evidence about future harm and treatment benefit. ADHD is harder because there is no single measurement equivalent to blood pressure. Multiple traits, informants and contexts must be interpreted, so reliability and assessor judgement matter more.
A useful ADHD diagnosis should identify a developmentally coherent and significantly impairing pattern; help distinguish plausible alternatives and coexisting conditions; and lead to support or treatment with a reasonable chance of benefit. If it merely redescribes a personality or common lapse, it has failed. That is a criticism of how the category was applied, not proof that all applications are arbitrary.
This is also why isolated criteria are misleading. Everybody sometimes loses things or interrupts. Not everybody shows a persistent pattern, beginning during development, across settings and at meaningful cost. The diagnostic work lies in that whole sentence.
Environment changes expression, impairment and recognition
Pemberton’s most persuasive point is that an environment helps determine how much a trait matters.
A child who struggles with attention may function better in an active, flexible classroom than one organised around prolonged sitting and repeated written work. An adult may excel in urgent, stimulating tasks while struggling with unstructured administration. Quiet space, movement, external reminders, task variety and supportive supervision can reduce impairment. Sleep loss, stress, sensory overload and chaotic demands can magnify it.
That interaction is compatible with neurodevelopmental disorder. Disability often arises from the relationship between a person and their environment. A ramp reduces the disabling effect of a mobility impairment without proving that the underlying difference was socially invented. Environmental adjustment is a real intervention, not a concession that the diagnosis was false.
Modern environments may also influence attention across the population. Longitudinal research supports reciprocal associations between problematic digital-media use and ADHD symptoms: media use may worsen symptoms, while people with more symptoms may also be drawn to or have more difficulty regulating that use.8 Association does not identify one universal causal direction.
School context supplies a particularly important warning. Children who are young within their school year are more likely to receive an ADHD diagnosis than their older classmates.9 Developmental immaturity may be mistaken for disorder, or classroom demands may expose genuine vulnerability. Either way, clinicians should compare a child with age-appropriate development and examine function beyond one classroom.
Environment also shapes who is noticed. A disruptive, visibly hyperactive boy is more likely to prompt referral than a quiet, inattentive girl whose effort and distress remain private. Able children and adults may compensate through rigid systems, anxiety, overwork or occupational fit, only becoming visibly impaired when demands rise or structure falls away. Success does not prove absence of ADHD, but neither does hidden effort prove its presence. Both require a developmental and functional assessment.
These examples argue for contextual diagnosis. They do not establish that screens, schooling or parenting manufacture ADHD from nothing.
Biology and biography are not rival explanations
The language of “brain disorder” can itself mislead. Every thought, habit and learned strategy has a biological substrate; finding a correlate does not tell us whether it is cause, consequence or compensation. Equally, describing the social meaning of a diagnosis does not dissolve the person’s enduring difficulties.
For an individual, several levels of explanation may all be useful:
- genetic and developmental predispositions;
- learned strategies and past experiences;
- sleep, physical and mental health;
- present family, school and workplace demands;
- cultural expectations and access to support; and
- the consequences of receiving or not receiving a diagnostic label.
Good formulation asks how these interact. It does not force one level to defeat the others.
Validity, reliability and service quality
Public debate repeatedly substitutes one question for another.
- Validity: does ADHD identify a clinically meaningful pattern?
- Reliability: do competent assessors reach sufficiently similar conclusions?
- Individual accuracy: is this person’s diagnosis justified?
- Service quality: did the assessment and follow-up meet a defensible standard?
- Population distribution: which groups are being missed or labelled incorrectly?
Evidence that ADHD has genetic associations does not certify one assessment. Evidence that one clinic assessed badly does not invalidate the construct. A national shortfall in recognised diagnosis can coexist with false positives in particular people or services.
The conclusion permitted by the evidence is less theatrical than either “a scan proves ADHD” or “no scan means ADHD is a myth”. ADHD is a clinically defined neurodevelopmental disorder with substantial genetic and developmental evidence. Its traits are dimensional and heterogeneous. Biological findings are real but not individually diagnostic. Social conditions shape expression, impairment, recognition and the consequences of the label.
All of those statements can be true at once.
The important scrutiny belongs at the points where evidence becomes judgement: whether an individual’s history is coherent, whether impairment is substantial, whether alternatives were considered and whether diagnosis opens a proportionate route to help.
That is more demanding than choosing between biology and society. It is also closer to the truth.
Series disclosure: I am a doctor diagnosed with ADHD in adulthood and take lisdexamfetamine. This is context, not evidence; the clinical claims in this series should stand or fall on the sources cited.
Sources and further reading
Series navigation
← Previous: Article 2 — Did the documentary make its case? · Series index · Next: Article 4 — Can you buy an ADHD diagnosis? →
Leave a Reply