The difference between the groups' test results was 25 to 50 milliseconds on average, the researchers found, too brief to be detected in social interactions with an infant. However, they showed that this measurable delay could be accounted for by differences in the structure and organization of actively developing neurological circuits of a child's brain.
Efficiently shifting attention early in infancy is thought to be important for later social and cognitive development. Split-second delays, the researchers suggested, could be a precursor to such well known symptoms of autism as difficulty making eye contact or following a parent's pointing finger, problems that generally emerge after a child turns 1. Typically, autism spectrum disorder (ASD) is not diagnosed until after 3 or 4 years of age.
"This study ties a difference in reaction times to differences in the developing brain, which may shape the way babies take in and respond to their environment in more noticeable ways over time," said Alice Kau, Ph.D., of the Intellectual and Developmental Disabilities Branch of the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), the institute that funded the research. "The brain's pathways for communication are forming rapidly in early infancy, and small differences at this stage could foretell greater difference at a later age."
First author Jed T. Elison, Ph.D., of the University of North Carolina at Chapel Hill (UNC) and California Institute of Technology, Pasadena, collaborated with senior author Joseph Piven, M.D., of UNC, and researchers from The Children's Hospital of Philadelphia and the University of Pennsylvania, Philadelphia; the University of Texas at Dallas; Washington University, St. Louis; the University of Washington, Seattle; the University of Utah, Salt Lake City; McGill University, Montreal; and the University of Alberta, Canada.
The study appears in the American Journal of Psychiatry.
The research is part of the ongoing Infant Brain Imaging Study, which is supported through the NICHD's Autism Centers of Excellence Program.
To measure shifts in gaze and visual attention, the researchers used sophisticated eye tracking equipment to capture the precise timing of eye movements. The infants sat on their parent's laps and watched images appear on a computer monitor. The test procedure used in the study is known as the gap/overlap task. In one part of the test, an image would appear in the center of the screen to attract the infant's gaze, and would then disappear. After a brief delay, or gap, another image would appear at the edge of the screen.
In another part of the test, the central image remained on the screen, and an image appeared at the periphery of the screen. The researchers measured the time it took infants to initiate an eye movement to the image in the periphery. In addition to the eye tracking task, the 7-month-old infants took part in a type of magnetic resonance brain imaging called diffusion weighted imaging, which measures the organization of neural circuits in the brain.
Fifty-seven infants had an older sibling diagnosed with autism, and so were considered at higher risk for developing autism themselves. The study also included 40 infants who did not have an older sibling with autism and so were considered at low risk for developing autism. All of the children returned to the study facility after their second birthdays for clinical assessments. By this time, 16 of the high-risk children were classified as having ASD. Based on the classification during the clinical assessment visit, the researchers compared the brain imaging data and the eye tracking data collected at 7 months across three groups:
- Children with an older sibling with ASD who themselves were classified with ASD (high-risk positive)
- Children with an older sibling with ASD who were not classified with ASD (high-risk negative)
- Children who did not have an older sibling with ASD (low risk)
The researchers uncovered evidence that the functioning of a key brain structure may account for the differences in gaze shifting between the groups. The brain structure is called the splenium of the corpus callosum. This structure is considered to be an important neural connection between the two hemispheres of the brain.
In the low-risk infants, the researchers found that the speed with which the infants shifted their gaze was closely associated with the size of the splenium. The greater the size of the splenium, the more rapidly the infants were able to switch their gaze.
However, in the infants who later were found to have autism, the researchers did not find any correlation between splenium size and the speed at which an infant shifted gaze. The researchers theorize that the differences in gaze shifting between the two groups may not be due directly to differences in the splenium between the groups, but to differences in a brain circuit that connects the splenium to visual areas of the brain.
Ultimately, differences in gaze detected at 7 months of age might help doctors identify children likely to develop autism later on, the authors suggested.
"By refining the gaze test and coupling it with other assessments, we hope to improve the ability to identify ASD in the first year of life," Dr. Elison said.