A meta-analysis of 66 studies found that autistic individuals tend to show differences in the microstructure of many of the brain’s white matter tracts. Compared with typically developing individuals, they showed moderately lower fractional anisotropy, a measure often interpreted as reflecting the organization of nerve fibers, in nine tracts, and higher mean diffusivity, indicating freer water movement, in eight tracts. The paper was published in Biological Psychiatry: Cognitive Neuroscience and Neuroimaging.

Researchers have long investigated whether differences in brain connectivity contribute to the characteristics of autism spectrum disorder. Some theories propose that autism involves weaker communication between distant brain regions, particularly between the front and back of the brain, alongside stronger connections between nearby regions. These differences may contribute to the cognitive and behavioral characteristics of autism.

Scientists increasingly use magnetic resonance imaging (MRI) techniques to investigate the structure of brain white matter. Brain white matter is composed of nerve fibers that connect different brain regions. One widely used technique to study white matter of the brain is diffusion tensor imaging. This technique examines how water molecules move through brain tissue to provide information about its organization and structural properties.

Studies using this technique have identified differences in several white matter pathways in autistic individuals, typically indicating less directionally organized water movement and greater overall freedom of water movement. However, findings have been inconsistent, with some studies reporting widespread differences in white matter structure and others identifying only localized differences or no differences at all.

Study author Christy D. Yoon of the University of Wisconsin–Madison and colleagues aimed to pool diffusion tensor imaging data from studies of autistic individuals to clarify the findings on white matter alterations. More specifically, they wanted to quantify the size of differences in two indicators of white matter structure, fractional anisotropy and mean diffusivity, across published studies. They also examined whether factors such as age, sex, IQ, publication year, and brain hemisphere influenced the size of these differences.

Fractional anisotropy measures how strongly water molecules move in a particular direction within brain tissue, which provides information about the organization and integrity of white matter nerve fibers. Mean diffusivity measures how freely water molecules move through brain tissue in general, potentially reflecting differences in tissue density and structure.

The researchers searched four databases of scientific literature (MEDLINE, PsycArticles, PsycINFO, and PubMed) for studies comparing autistic and typically developing individuals using quantitative brain imaging techniques. To be included, studies had to be published in English in peer-reviewed journals, include both an autistic group and a typically developing comparison group, derive a quantitative measure of white matter microstructure, and report enough data to calculate the size of the difference between groups.

Although the researchers initially aimed to include several imaging techniques, there were too few data from techniques other than diffusion tensor imaging, so the final analysis focused on fractional anisotropy and mean diffusivity. This left 66 studies, contributing 881 effect sizes from 2,231 autistic and 1,856 typically developing participants, with study average ages ranging from 2 to 50 years. The analysis covered 12 white matter tracts.

Results showed that autistic individuals had, on average, moderately lower fractional anisotropy in nine white matter tracts: the corpus callosum, which connects the brain’s two hemispheres, the corticospinal tract, the thalamic radiation, the arcuate fasciculus, the inferior fronto-occipital fasciculus, the inferior longitudinal fasciculus, the superior longitudinal fasciculus, the uncinate fasciculus, and the cingulum. For two of these tracts, the superior longitudinal fasciculus and the cingulum, the difference reached statistical significance only after the researchers removed a few unusually influential results, so the authors urge caution in interpreting them.

The researchers also found moderate-to-large differences in mean diffusivity in eight tracts: the corpus callosum, corona radiata, corticospinal tract, arcuate fasciculus, inferior fronto-occipital fasciculus, inferior longitudinal fasciculus, superior longitudinal fasciculus, and uncinate fasciculus. In all eight, autistic individuals had higher mean diffusivity than typically developing individuals, which may indicate lower tissue density, although other explanations are possible. The largest difference appeared in the arcuate fasciculus, a pathway involved in language, but the researchers note that an effect this large may have been inflated by factors unrelated to white matter microstructure, such as enlarged fluid-filled spaces in the brain.

Many of the differences appeared in long-range pathways connecting distant brain regions, which the authors say supports the idea that autism involves atypical long-range connectivity. No differences were found in the cerebellar peduncles or the internal capsule, though the researchers suggest this may reflect limitations of the imaging technique in those regions rather than a true absence of differences.

The size of the differences also varied with study characteristics. Fractional anisotropy differences were smaller in studies with higher average IQs in seven tracts, and smaller in older samples in two tracts. Samples with a higher proportion of males tended to show smaller differences in several tracts, a pattern the authors link to theories proposing that females require a greater genetic or environmental load to develop autism. Publication year did not affect the results, suggesting that the differences have remained stable across nearly two decades of research.

“Our findings highlight the complex, multidimensional nature of white matter microstructure alterations observed in ASD [autism spectrum disorder],” the study authors concluded.

The study contributes to the scientific understanding of white matter differences in autism. However, the authors note that males were overrepresented in the included studies, limiting the generalizability of the findings to females. The age range of participants also excluded people in later adulthood, when white matter declines, limiting the generalizability of the findings to older adults.

In addition, the findings mostly represent autistic individuals with higher IQs and potentially lower support needs. Diffusion tensor imaging measures are also nonspecific, so they cannot pinpoint the biological changes underlying the differences, and the researchers could not rule out that differences in head movement during scanning inflated some effects. The meta-analysis was not preregistered, and the included studies were cross-sectional, limiting conclusions about how these differences change with age.

The paper, “Widespread alterations in white matter microstructure in autism: A multilevel meta-analysis,” was authored by Christy D. Yoon, Andrew L. Alexander, Brittany G. Travers, Janet E. Lainhart, and Douglas C. Dean III.