Discover SPARK

Rethinking Head Size in Autism: Scientists Question Statistics on Early Brain Overgrowth

Marina Sarris

A version of this article first appeared in

For years, researchers have said that children with autism are relatively likely to have large heads, a phenomenon they attributed to “early brain overgrowth.”

Armin Raznahan, a scientist at the National Institute of Mental Health, repeated that scientific theory himself. “Like everyone else, I used the line that ‘autism is associated with early brain overgrowth,’” said Raznahan.

But now he and other researchers — including some using data from Simons Simplex Collection (SSC) families — are questioning assumptions about brain overgrowth and autism, especially how common it is. The Simons Simplex Collection was a SFARI-funded research initiative preceding SPARK. Their work spotlights some important ways science reconsiders evidence in the search for answers.


Many studies have found that from a tenth to a third of children with autism have unusually large heads.1,2 Scientists drew that conclusion by comparing the youngsters’ head-circumference measurements with those on American or international charts that plot normal growth by age and gender. At a typical “well baby” exam, a doctor or nurse will measure a baby’s head, weight and length, and record those numbers on a chart showing percentiles. 

An excessively large head could point to serious problems such as fluid in the brain. However, in the children with autism, scientists believed, a bigger head size was related to brain size. They theorized that the youngsters were experiencing an overgrowth of brain matter during their first year of life and wondered if the overgrowth was an early warning sign of autism.

Of course, they assumed the growth charts were correct. But what if they weren’t?

Two recent studies addressed this very concern. One, which appeared in the prestigious journal Pediatrics, examined the head-circumference measurements of more than 75,000 children in a large U.S. primary care network. Researchers found that two growth charts, one from the U.S. Centers for Disease Control (CDC) and the other from the World Health Organization (WHO), identified too many children as having abnormally large heads. In other words, head sizes overall were larger than the charts predicted.3

The medical term for an abnormally big head is “macrocephaly,” from the Greek words for “large” and “head.” Macrocephaly can be a sign of brain and genetic conditions, although a person can have macrocephaly and be otherwise normal.4 


Most research connecting macrocephaly to autism spectrum disorder (ASD) has relied upon standardized growth charts, and Raznahan’s research team decided to examine 34 such studies. “The notion of the big-headness in autism has massively influenced autism neuroscience,” he says — so the topic warranted further investigation.

His findings “shocked” even him: the growth charts used in many of the studies made it appear that more children have abnormally large heads than really do. Altogether, the studies reported an average macrocephaly rate of 14 percent in children with ASD. But when his team excluded studies that relied upon certain growth charts, the average rate fell to 4 percent.5 That’s not far from the 3 percent rate of macrocephaly in the general population.2

The team found 13 head-size studies that did not rely solely on growth chart standards. Of those, slightly more than half found evidence of head enlargement in ASD, while the others did not.5

Those conclusions do not mean that head size is unimportant in autism, however. “Our study doesn’t contradict the finding that there is a subgroup of children with autism with unusually large heads,” Raznahan explains. However, the overgrowth appears to occur later in early childhood that previous studies found, and to be subtler than the dramatic picture painted in some earlier reports.


In a separate study, a group of researchers approached the issue of head size from a different angle. Using data from families in the SSC, the research team compared head sizes of children with autism with those of their unaffected siblings and parents.

The researchers found a high rate of macrocephaly among the SSC families when their basis of comparison was standardized head-circumference charts. Almost 15 percent of the children with autism, 13 percent of the fathers, 20 percent of the mothers and 13 percent of the siblings had large heads for their gender and age according to such charts.1

However, the team created a formula for considering other factors that can affect the size of a person’s head. Head size, like other physical attributes, may be affected by the genes we inherit from our parents, along with height, weight and ethnic characteristics.1 When those factors were considered, the team found very few people whose heads were unexplainably large — just 3.6 percent of the children with autism.

They did find one small but intriguing difference. The SSC children with ASD had an average head size that was 2 millimeters larger than their unaffected brothers and sisters. That’s about the width of an apple stem or the metal tip of a ballpoint pen. 

They did find that children whose heads were significantly larger or smaller than expected, given their genetic potential, had lower intelligence than the others.


“Having autism doesn’t necessary mean having a big head. Having autism is less important to head circumference than other factors we examined, like height, weight and genetic ancestry,” says researcher Soo-Jeong Kim, a child psychiatrist at Seattle Children’s Research Institute.

The research paper calls for doctors to consider such factors when determining whether a child’s head size is truly abnormal. Growth charts “will classify individuals as macrocephalic far more often than is clinically appropriate” and will miss other people whose heads are larger than their genes would predict.1

Spokespeople for the CDC say they do not comment on research by scientists outside their agency, so they have no response to the head-circumference studies. The CDC recommends that American pediatricians use WHO growth charts, rather than its own, for children from birth to age 2. The WHO charts track the growth of babies in different countries who are breastfed for 12 months.6

According to the study in Pediatrics, using either the WHO or CDC charts in older youngsters could result in some children being misidentified with macrocephaly.


  1. Chaste, P., Klei, L., et al. (2013). Adjusting head circumference for covariates in autism: clinical correlates of a highly heritable continuous trait. Biological psychiatry, pii: S0006-3223(13)00389-2. View abstract.
  2. Dementieva, Y.A., Vance, D.D., et al. (2005). Accelerated head growth in early development of individuals with autism. Pediatric Neurology, 32(2),102-108. View abstract.
  3. Daymont, C., Hwang, W.T., Feudtner, C., & Rubin, D. (2010). Head-circumference distribution in a large primary care network differs from CDC and WHO curves. Pediatrics, 126(4). View abstract.
  4. Menounou, A. (2011). Head size: is it important? (2011). Advances in clinical neuroscience and rehabilitation, 11(2). Retrieved from
  5. Raznahan, A., Wallace, G.L., et al. (2013). Compared to what? Early brain overgrowth in autism and the perils of population norms. Biological psychiatry, pii: S0006-3223(13)00303-X. View abstract.
  6. Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report. (2010). Use of World Health Organization and CDC growth charts for children aged 0-59 months in the United States, Sep. 10. Retrieved from