ADHD is supported by converging genetic, developmental, brain and treatment evidence. There is no single cause and no diagnostic brain or gene test — but the core machinery is well understood.
1 · Quick understanding
ADHD is one of the most studied conditions in psychiatry. Several independent lines of evidence point the same way: it is strongly heritable, it is linked to differences in how the brain develops, it involves the systems that regulate attention and motivation, and medicines that act on those systems reduce symptoms in trials.
An international consensus of 80 authors from 27 countries, drawing only on large studies and meta-analyses, set out 208 evidence-based conclusions about ADHD's nature, course, causes and treatment.
None of this gives a test for an individual. Diagnosis remains clinical.
2 · Possible explanations
Explore the levels below. Select any part to see its connections and how strong each one is; switch on a medicine to see where it acts.
From genes to everyday experience — a six-level teaching model
Teaching model, not a diagnosis. Real systems involve hundreds of genes, many cell types and feedback loops. Solid = established mechanism · dashed = association in groups · dotted = hypothesis.
Genes
Many common variants: Twin studies estimate heritability around 74%. Genome-wide studies find many common variants, each with a very small effect; 27 loci reached significance in the largest study, with 76 candidate genes enriched for expression in early brain development. R7, W5
Rare variants: Rare protein-truncating variants and copy-number changes add risk in some people. The 2023 study implicated SORCS3 through both common and rare variants. W5, R7
Shared with other conditions: An estimated 84–98% of ADHD-influencing variants are shared with other psychiatric conditions — one reason ADHD so often co-occurs with others, and why genes cannot diagnose an individual. W5
Components
Tyrosine hydroxylase: The enzyme that makes L-DOPA from tyrosine — the usual rate-limiting step in making dopamine and noradrenaline. R8
VMAT2 (vesicle loader): Pumps dopamine from the cell fluid into storage vesicles, ready for release. Amphetamines disturb this storage and redistribute dopamine into the cell fluid. R8
DAT (dopamine transporter): Recaptures released dopamine, ending its signal. Methylphenidate blocks it; amphetamines can reverse it so it pushes dopamine out. R8, R2
NET (noradrenaline transporter): Recaptures noradrenaline — and in the prefrontal cortex, where DAT is sparse, much of the dopamine too. Atomoxetine blocks it. R9, W6
MAO and COMT: Enzymes that break dopamine and noradrenaline down inside and outside cells. R8
D1 and D2 receptors: Receive the dopamine signal. D2 receptors on the releasing cell also act as a brake (autoreceptors), feeding back to reduce release. R5, R9
Alpha-2A receptors: Noradrenaline receptors in prefrontal cortex that strengthen task-relevant network signals. Guanfacine stimulates them. R9
Store: Dopamine is packed into vesicles; unpacked dopamine is vulnerable to breakdown. R8
Release: Vesicles release their contents when the cell fires. Amphetamines add a second, firing-independent release route. R8
Clear: Transporters pull the signal back in; enzymes break it down. Most medicines act by slowing clearance or forcing release. R8, R2
Self-tune: Autoreceptors and adaptation adjust the system over time. Transporter levels appear to shift after stimulant exposure — which complicates brain-scan comparisons. R6
Circuits
Prefrontal regulation: Prefrontal cortex regulates attention, behaviour and emotion; imaging shows these regions less active, with weaker connections, in ADHD on average. Its function depends on finely tuned dopamine and noradrenaline levels. R9
Cortical maturation timing: In 223 children with ADHD versus 223 controls, half of cortical points reached peak thickness by about 10.5 years versus 7.5 years — a delay most marked in prefrontal regions. R4
Subcortical volumes: Across 23 sites, accumbens, amygdala, caudate, hippocampus and putamen were slightly smaller on average (effect sizes d −0.11 to −0.19), mainly in children; differences in adults were not significant. Medication did not explain them. R3
Reward pathway markers: In 53 unmedicated adults, PET showed lower dopamine transporter and D2/D3 receptor binding in reward regions, correlating with inattention. But a meta-analysis found transporter levels lower in drug-naive and higher in previously medicated people — so findings depend on history. R5, R6
Functions
Executive control: Holding a goal, inhibiting distractions, switching and planning. Common-variant ADHD risk is associated with weaker performance on several executive measures. W5, R9
Reward and motivation: How strongly future or delayed outcomes drive effort now. R5
Arousal and state: Matching alertness to the task — too little or too much both degrade control. R9
Experience
Starting tasks: Knowing what to do and not being able to begin — one of the most commonly described ADHD difficulties. W6
Holding intentions: Losing track of a plan between deciding and doing; forgetting why you entered a room. W6
Time awareness: Under-estimating elapsed time and time needed; 'now' and 'not now' as the only two clocks. W6
Acting before deciding: Speech or action running ahead of evaluation. W6
Connections and evidence strength
Many common variants → Cortical maturation timing: association — Risk genes are enriched for early brain development expression; the route to cortical timing is not demonstrated.
Many common variants → Reward pathway markers: association — Common-variant risk is associated with midbrain dopaminergic neuron expression (Demontis 2023).
Many common variants → Executive control: association
NET (noradrenaline transporter) → Clear: established
MAO and COMT → Clear: established
D1 and D2 receptors → Self-tune: established
Store → Release: established
Make → Store: established
Release → Prefrontal regulation: established
Release → Reward pathway markers: established
Alpha-2A receptors → Prefrontal regulation: established
Self-tune → Reward pathway markers: contested — Whether transporter differences are a cause or an adaptation to past medication is unresolved (Fusar-Poli 2012).
Prefrontal regulation → Executive control: established
Cortical maturation timing → Executive control: association
Subcortical volumes → Reward and motivation: association
Reward pathway markers → Reward and motivation: association
Prefrontal regulation → Arousal and state: established
Executive control → Starting tasks: association
Executive control → Holding intentions: association
Executive control → Acting before deciding: association
Reward and motivation → Starting tasks: association
Executive control → Time awareness: association
VMAT2 (vesicle loader) → Reward pathway markers: hypothesis — Individual differences in vesicle storage capacity shaping stimulant response is a research question, not an established ADHD mechanism.
Medicine overlays
Methylphenidate: Blocks dopamine and noradrenaline transporters, so released transmitter stays active longer. First-line in children and one of two first-line options in adults (NICE). R2, W1
Amphetamines (dexamfetamine): Enter the terminal, disturb vesicle storage and run transporters in reverse, releasing dopamine and noradrenaline independent of firing; they also block reuptake. In adults' clinician-rated trials, amphetamines had the largest average symptom effect, with lower tolerability than placebo. R8, R2
Lisdexamfetamine (Elvanse): An inactive prodrug absorbed intact and converted to d-amphetamine, mainly by red blood cells. Same active drug as dexamfetamine, delivered more gradually — a different exposure profile, not an identical one. R10, R1
Atomoxetine: Non-stimulant that blocks the noradrenaline transporter, raising noradrenaline and prefrontal dopamine. W6, R2
Guanfacine: Non-stimulant that stimulates alpha-2A receptors in prefrontal cortex, strengthening network signalling. R9
The useful way to think about it: the parts are few, the regulation is complex. Most people with ADHD do not have a broken part; they have many small differences in how parts are built and tuned, adding up across development.
Because the medicines act on well-understood parts of the supply chain, their benefit in trials is strong evidence that the chain matters for symptoms — even though it does not show that every person's ADHD starts there. See How ADHD medicines act.
5 · Deeper explanation
Genes
The largest genome-wide study (38, 691 cases) found 27 significant loci and 76 candidate genes enriched for expression in early brain development and in midbrain dopaminergic neurons; 84–98% of ADHD-influencing variants are shared with other psychiatric conditions.
Children with ADHD reached peak cortical thickness at a median of about 10.5 years versus 7.5 years in controls, with the delay most marked in prefrontal regions.
A meta-analysis found transporter levels lower in medication-naive people but higher in previously medicated people — so brain-scan findings depend partly on treatment history.
The overall picture is well supported; how it applies to any single person is not measurable with current tests. No single universal "dopamine deficit" explains everyone.
Sources cited
W6 Faraone SV, et al. The World Federation of ADHD International Consensus Statement: 208 evidence-based conclusions about the disorder. Neuroscience & Biobehavioral Reviews 2021;128:789–818. link(Abstract read; full text open access)
R7 Faraone SV, Larsson H. Genetics of attention deficit hyperactivity disorder. Molecular Psychiatry 2019;24(4):562–575. (Abstract read (PubMed))
W5 Demontis D, et al. Genome-wide analyses of ADHD identify 27 risk loci, refine the genetic architecture and implicate several cognitive domains. Nature Genetics 2023;55:198–208. 38,691 cases, 186,843 controls. (Abstract read (PubMed))
R4 Shaw P, et al. ADHD is characterized by a delay in cortical maturation. PNAS 2007;104(49):19649–54. 223 children with ADHD, 223 controls, 824 scans. (Abstract read (PubMed))
R5 Volkow ND, et al. Evaluating dopamine reward pathway in ADHD: clinical implications. JAMA 2009;302(10):1084–91. 53 unmedicated adults with ADHD, 44 controls. (Abstract read (PubMed))
R6 Fusar-Poli P, et al. Striatal dopamine transporter alterations in ADHD: pathophysiology or adaptation to psychostimulants? A meta-analysis. Am J Psychiatry 2012;169(3):264–72. Nine PET/SPECT studies. (Abstract read (PubMed))
R3 Hoogman M, et al. Subcortical brain volume differences in participants with ADHD in children and adults: a cross-sectional mega-analysis (ENIGMA). Lancet Psychiatry 2017;4(4):310–319. 1,713 ADHD, 1,529 controls. (Abstract read (PubMed))
R2 Cortese S, et al. Comparative efficacy and tolerability of medications for ADHD in children, adolescents, and adults: a systematic review and network meta-analysis. Lancet Psychiatry 2018;5(9):727–738. 133 double-blind RCTs. (Abstract read (PubMed))