Since early 2026, one headline keeps coming back in social feeds: "Scientists discover 3 types of ADHD." It has been picked up by National Geographic, US public radio, health websites and a flood of posts and short videos, often with a colourful brain image and the suggestion that a scan can now tell which type you have. Behind it is a genuine and carefully done study. But the posts usually skip the details that matter most: who was scanned, what a "biotype" actually is, and what the study did not test.
The Study
The paper was published online on 25 February 2026 in JAMA Psychiatry by Nanfang Pan, Qiyong Gong, Ying Chen and colleagues at West China Hospital of Sichuan University, working with Monash University in Australia and the University of Cincinnati and University of California, Davis, in the United States. The team pooled brain scans from six research sites in the United States and China (446 children with ADHD and 708 children without ADHD), then checked their results in a separate US group, the Healthy Brain Network (554 children with ADHD and 123 without).
That adds up to 1,831 scans, the figure quoted in many posts. Two points are rarely mentioned: only 1,000 of those scans were of young people with ADHD, and every participant was a child or a teenager, aged roughly 6 to 18. No adults were included.
What Is a "Biotype"?
A biotype is a subgroup defined from biological measurements rather than from symptoms. Here, the researchers used standard structural MRI scans to estimate how similar each brain region is to the others in its anatomy, building what is called a morphometric similarity network. They then compared each child with "growth charts" built from the children without ADHD (a method called normative modelling) to see where each brain departed from what is typical for its age and sex.
Finally, an algorithm sorted the children with ADHD into groups using only these brain deviations. Symptoms were not given to the algorithm. That is what makes the result interesting: the groups it found turned out to differ in their symptoms too.
The Three Groups
- Severe combined, with emotional dysregulation (142 children): the most widespread differences, centred on links between the medial prefrontal cortex and the pallidum, a deep brain structure. These children had the highest levels of both inattention and hyperactivity/impulsivity.
- Predominantly hyperactive/impulsive (177 children): differences concentrated in a circuit linking the anterior cingulate cortex and the pallidum.
- Predominantly inattentive (127 children): more focal differences, mainly in the superior frontal gyrus.
The "about twice as many affected regions" line seen in some posts is a fair approximation: the first group showed 45 significant deviations, against 26 and 11 in the other two. The main patterns were reproduced in the independent Healthy Brain Network group, which is a real strength of the study.
The Emotional Piece
A subset of children from the Chengdu site, all unmedicated at the start, was followed every year for four years using a parent questionnaire. Emotional self-regulation problems fell markedly in the second and third groups but stayed more persistent in the first. The difference was statistically significant, though modest in size. Attention and behaviour problems did not follow different paths across the groups. The first group also had more mood disorders (25% versus about 10% and 5%), but that difference was not statistically significant.
This matters because emotional dysregulation, the quick, intense and hard-to-calm reactions many people with ADHD describe, is not part of the official diagnostic criteria. A separate, smaller study from Fordham University in Psychological Medicine (123 children aged 5 to 9), covered by PsyPost, also linked severe emotional outbursts in ADHD to differences in the prefrontal cortex. The two studies are sometimes blended together online: they are different pieces of work pointing in a similar direction.
How This Differs from the DSM "Types"
The DSM, the diagnostic manual used by clinicians, already describes three presentations of ADHD: predominantly inattentive, predominantly hyperactive/impulsive, and combined. They are based on counting symptoms, and a person's presentation often shifts over a lifetime. Two of the biotypes closely resemble these familiar presentations, so the study did not "discover" three new kinds of ADHD. What it adds is evidence that brain differences in children with ADHD are at least partly organised into patterns, and that the most severe profile carries an emotional component the DSM leaves out.
What the Study Does Not Show
- It is not a diagnostic scan. Brain deviations overlapped between children with and without ADHD, and the groups only emerge at the level of hundreds of scans. ADHD remains a clinical diagnosis based on history and behaviour. The authors stress that their biotypes are not new diagnostic categories.
- Medication response was not tested. Posts saying the study shows which medication suits which type go beyond the data. Better matching of treatments is a long-term hope voiced by the researchers. Most participants had not been medication-free, and the authors note that medication may have influenced the scans.
- Children only. Nobody knows yet whether the same groups exist in adults.
- Probably not separate boxes. The authors themselves write that the biotypes may be points along continuous dimensions rather than distinct conditions.
- Brain chemistry was inferred, not measured. Links with serotonin, dopamine or glutamate come from comparing the brain maps with reference atlases from other people, not from measurements in these children.
- A selected sample. Children with many co-occurring conditions were excluded, and all data came from the US and China, so groups may look different in everyday clinics.
What This Means for You
You do not need a brain scan to take your experience seriously, and no scan can currently tell you which "type" you are. If difficulties with attention, restlessness or impulsivity have been with you since childhood and affect work, studies or relationships, a validated questionnaire is a useful first step. The free ASRS-v1.1 adult ADHD screener has 18 questions and takes about 10 minutes.
If what you recognise most is the emotional side, such as feelings that flare up fast and take a long time to settle, or a strong reaction to criticism, you can also explore our free emotion regulation test (DERS) and our rejection sensitivity test. Screening tests are not a diagnosis, but your results can help you prepare a conversation with a professional.
Source: Pan N, Long Y, Qin K, Pope IZ, Chen Q, Zhu Z, Cao Y, Li L, Singh MK, McNamara RK, DelBello MP, Chen Y, Fornito A, Gong Q. "Mapping ADHD Heterogeneity and Biotypes by Topological Deviations in Morphometric Similarity Networks." JAMA Psychiatry 2026;83(5):478-490 (published online 25 February 2026). DOI 10.1001/jamapsychiatry.2026.0001. Multisite cross-sectional MRI study in children and adolescents aged about 6 to 18: discovery cohort of 446 with ADHD and 708 controls from 6 sites in the US and China; validation cohort of 554 with ADHD and 123 controls (Healthy Brain Network); 4-year follow-up in a medication-naive subset.