In brief
The largest imaging study to date with brain scans (ENIGMA consortium, 1,571 people with autism and 1,651 controls from 49 sites) found only small differences on the group average in individual brain regions (effect sizes mostly between about −0.2 and +0.2). Earlier studies were, according to the authors, inconsistent. [1] A reliable biological marker for the diagnosis does not exist according to the 2016 S3 guideline; diagnosis is based on behavioural observation. [2, Ch. B.2.1] A brain scan is therefore not a test for autism and does not explain why an individual person is the way they are. The guideline has expired. [3]
This English translation was produced with AI assistance and has not yet been independently reviewed.
What the large collaborative study found
The study evaluated MRI scans of people aged 2 to 64 years using a uniform procedure. On average, people with autism had smaller volumes in some deeper brain structures (pallidum, putamen, amygdala, nucleus accumbens) and a changed thickness of the cerebral cortex (thicker in the frontal area, thinner in the temporal area). The differences were small (effect sizes Cohen’s d up to about 0.2). The development of cortical thickness followed a different course, with the biggest differences around puberty. [1] The authors interpret this as an indication of a changed interplay of several brain regions over the lifespan. [1]
What this does not mean: Group averages allow no statement about an individual person; small effects mean that the distributions overlap strongly. The study shows differences, not causes.
Why brain images do not replace assessment
The 2016 guideline stated that reliable biological markers for autism spectrum disorders do not exist and that the diagnosis is based on behavioural observation. [2, Ch. B.2.1] Instrument-based diagnostics such as EEG or MRI should only be carried out with a clear clinical indication, for example to clarify other neurological findings. [2, Rec. 31] More on assessment: How an autism assessment works.
Earlier indications on brain growth
Some studies found accelerated growth of head circumference and brain in the first years of life in children later diagnosed. The guideline put this in context: the abnormalities in head growth could not be repeated in representative samples; the comparison norms may have been outdated; the absence of macrocephaly (large head) does not rule out autism. [2, Ch. B.2.4] For a brain growth pattern, neither sensitivity nor specificity was calculated. [2, Ch. B.2.4]
What do we know – what is still open?
Finding (large mega-analysis, abstract read): On average small differences in some brain regions and in the development of cortical thickness. [1]
Classification by the guideline (2016): No clinically usable biological marker; early growth findings not replicated. [2, Ch. B.2.1, B.2.4]
Open:
- What the differences mean for experience and behaviour, the study does not show.
- The study’s sample is large, but it comprises people with access to imaging and studies; people with high support needs are often under-represented in brain studies (assessment by the editorial team, not backed in the abstracts read).
- We have not read more recent imaging work (after 2018).
- Whether a reliable diagnosis or choice of therapy can ever be derived from brain features is open.
What else you should know
- Check headlines: With news about “the brain”, pay attention to group size, effect size and whether differences on average or a diagnosis are meant; help: How good is a study?.
- Causes: What research says about causes: Causes of autism and Genetics and autism.
- Not covered: functional imaging (fMRI), EEG studies and animal models.
Note: This article offers general information and does not replace medical advice.
Sources and literature
- van Rooij D, Anagnostou E, Arango C, et al. Cortical and Subcortical Brain Morphometry Differences Between Patients With Autism Spectrum Disorder and Healthy Individuals Across the Lifespan: Results From the ENIGMA ASD Working Group. American Journal of Psychiatry 2018;175(4):359–369. https://doi.org/10.1176/appi.ajp.2017.17010100 – Mega-analysis
- DGKJP, DGPPN et al. Autismus-Spektrum-Störungen im Kindes-, Jugend- und Erwachsenenalter. Teil 1: Diagnostik. S3-Leitlinie, AWMF-Reg.-Nr. 028-018, Textstand 23.02.2016 (Kap. B.2.1, B.2.4; Empf. 31). – Guideline (in German) (expired; according to the document note under revision)
- DGKJP. Aktuelle Leitlinien (Eintrag 028-018: „Gültig bis 04.04.2021“). https://www.dgkjp.de/wissen/aktuelle-leitlinien/ – Professional society (website in German)
You might also be interested in
Knowledge articleScience & research
Participatory autism research: research with rather than about people on the spectrum
Knowledge articleScience & research
How good is a study? Evidence and recommendations explained
Knowledge articleScience & research
Causes of autism: what research knows
Knowledge articleScience & research
Genetics and autism: inheritance, genetic tests and recurrence risk
Found a mistake?
If something is factually wrong, out of date or unclear, let us know. You do not have to give any personal details.