In Autism, Movements May Not Quickly Become Habit

by Faith Hickman Brynie

June 26, 2009

As their parents can attest, children who have autism seldom excel at sports. Although often skilled in manipulating numbers or memorizing facts, these kids often prove clumsy and awkward in athletic pursuits.

Stewart H. Mostofsky, a pediatric neurologist at the Kennedy Krieger Institute in Baltimore, may have found a reason why. In a study published online in the April 23 issue of Brain, Mostofsky reports that the brains of children with autism appear to work differently when performing a simple motor task, compared with the brains of typically developing children.

Mostofsky’s research team used functional magnetic resonance imaging (fMRI) to create images of the brain’s motor-control regions while 13 typically developing children and 13 high-functioning children with autism, all ages 8 to 12, performed a series of finger taps. Before they went into the brain scanner, the children were trained to touch, in sequence, their index finger, middle finger, ring finger, and little finger to their thumb. While in the scanner, they alternated left- and right-hand finger tapping with periods of rest.

During the finger-tapping exercise, brain activity increased in the primary sensorimotor cortex, thalamus, cerebellum and supplementary motor area (SMA) in both groups of children. Those brain regions are known to be associated with movements. However, activity in the cerebellum increased less in the children with autism than it did in the typically developing children. The cerebellum is the part of the brain that “takes over” when movements become habitual or automatic.

In addition, the children with autism showed a comparatively greater activation of the supplementary motor area, a higher cortical region that exerts “top-down,” voluntary control over movements.

These findings suggest that children with autism have trouble shifting responsibility for movement from the effort-evoking region, the SMA, to the cerebellum, where automatic control can take over. That would mean that while typically developing children are better able to rely on habitual motor systems; children with autism must continue to rely on the willful control of movement.

"We know that structural differences in the cerebellum occur in autism, but we don’t know much about how they affect the brain’s functions,” says Greg Allen, an educational psychologist at the University of Texas, Austin, who specializes in the diagnosis of autism but was not involved in this research.

“This study makes an important contribution: It confirms that the functioning of the cerebellum is compromised in individuals with autism, even when they perform a relatively simple motor task,” Allen says. “It also begins to shed light on how cerebellar changes affect connectivity with other brain systems."

Mostofsky and his colleagues also investigated connectivity, the ability of various brain regions to communicate with one another. Many anatomical, behavioral and imaging studies have suggested that localized, small-area connections “overgrow”—form more connections than normal—in the brains of people with autism, while connections among distant regions “undergrow.” These differences are termed “local overconnectivity” and “long-distance underconnectivity.” [See Dana story, “Autism and Motor Skills: A Matter of White Matter?” November 2007]

Mostofsky found evidence of long-distance underconnectivity in the children with autism. He measured faulty long-tract connections among the sensorimotor cortex, thalamus, SMA and cerebellum. “Decreased connectivity may reflect poor coordination within the circuit necessary for automating patterned motor behavior,” he says.

Mostofsky’s work suggests that imaging the brain during the performance of motor tasks may offer a way to diagnose autism earlier and may also help researchers find associations between autism and specific genes. The same neural processes that explain impaired control of movements could help explain how deficits in social and communicative skills arise, he suggests. His team plans to continue the research, looking at activation and connectivity while children are learning a new movement.