UMMS autism expert says new eye tracking research ‘an important indicator’

Teresa Mitchell believes study could lead to earlier interventions

By Ellie Castano

UMass Medical School Communications

November 27, 2013
mitchell-teresa
Teresa V. Mitchell, PhD

New research showing that infants who spent less time looking at people’s eyes were more likely to be diagnosed with autism is an important indicator that autism spectrum disorders (ASDs) may not stem from congenital abnormalities and therefore may be modified by early intervention, according to Teresa V. Mitchell, PhD, a researcher at UMass Medical School’s Eunice Kennedy Shriver Center who performs similar studies with older individuals with ASDs.

The study, published on Nov. 6 in the online edition of Nature, reports that researchers using eye-tracking technology found that infants later diagnosed with ASDs exhibit a “decline in eye fixation from 2 to 6 months of age, a pattern not observed in infants who do not develop autism spectrum disorders.”

“This research documents the earliest evidence of a diverging developmental path for infants who go on to receive a diagnosis of an autism spectrum disorder,” said Dr. Mitchell, who was not involved in this study.

“Fixating on another’s eyes is a critical early foundation for the development of socialization, language and attention. When presented with a human face, typically developing babies spend the bulk of their time looking at the eyes, and this behavior increases slightly over the first two years of life,” said Mitchell, assistant professor of psychiatry. “By contrast, babies who later go on to develop an autism spectrum disorder behave like their typical counterparts at age two months but their behavior quickly diverges and time spent looking at the eyes declines steadily until age 24 months.”

The research results point toward the possibility of intervening earlier when infants begin to show a decline in eye-gaze.

“The fact that eye gaze behavior in these babies starts out within the normal range is an important indicator that their divergent behavior may not stem from a congenital abnormality and therefore may be modified by early intervention,” said Mitchell.

While Mitchell calls the results of the research “particularly powerful” because it tested the infants ten times between the ages of 2 and 24 months and then maintained contact with them long enough to see which ones received diagnoses of an ASD, she hopes to see further research that replicates the findings.

“The behaviors they measured were highly variable between subjects so following a larger at-risk cohort would be very important for replicating the findings,” said Mitchell. “Furthermore, the variability between subjects suggests that continued follow-up of the original cohort may reveal more about individual differences in the ASD population, such as whether those infants whose behavior was closer to the normal range early on have better outcomes in later childhood or are more amenable to intervention. Jones and Klin were able to show that those infants whose gaze toward the eyes declined the most across the age range studied showed the greatest social deficits.”

Mitchell’s own lab, as well as others using similar eye-tracking technology, have found that older individuals with autism look less at the eyes than their typical peers do, but these findings are one piece in a complex puzzle.

“As more research is published we are learning that it is more complicated than we thought. In some instances this population can and does spend a normative amount of time fixating on the eye region, such as when they are viewing an audio-visual clip of a person talking. My colleague Ruth Grossman [PhD, adjunct assistant professor of psychiatry] and I conducted a study in which adolescents with autism were asked to view two video clips side-by-side and identify which one was in synchrony with a matching audio clip. The adolescents with autism fixated on the eyes as much as their typical counterparts did, but less at the mouth, which is where they needed to look to perform the task.

“The picture that is beginning to emerge across our study and others is that children on the autism spectrum are less able to select the most important information available on the face for a given task; it just happens that for most tasks the eyes are the critical source. Because individuals with ASDs are less adept at focusing their gaze and attention on the most relevant information, they are less adept in complex social interactions. The Jones and Klin paper documents a very early precursor to this effect that may be amenable to intervention because it starts within the normal range.”