ARRA project examines everyday reflexes, reveals unusual patterns of emotional processing

By Susan Johnson

April 28, 2011

Gabriel Dichter, Ph.D.
Assistant Professor of Psychiatry, University of North Carolina (UNC) at Chapel Hill School of Medicine Investigator, Carolina Institute for Developmental Disabilities Core Faculty, Duke–UNC Brain Imaging and Analysis Center

A volunteer participating in the study. Two electrical sensors taped under her eye record the muscle contractions of each blink. Two electrical sensors behind her right ear (not shown) measure another reflex. (The sensor on the forehead is the electrical ground.) When a startling noise plays through the headphones, the strength of these two reflexes reveals the volunteer’s emotional responses to pictures on a computer screen.

With support from the American Recovery and Reinvestment Act (ARRA), collaborators Gabriel Dichter, Ph.D., and Stephen Benning, Ph.D., have gained insight into the atypical but unconscious emotional responses that occur in people with autism. They are planning to apply this knowledge to improve treatment for people with this potentially debilitating disorder.

Since his years in graduate school, Dichter, an assistant professor of psychiatry at the University of North Carolina (UNC) at Chapel Hill School of Medicine, has been eager to study an ignored aspect of autism—blinking. This innocuous reflex has long been an important tool in studying emotional, or “affective,” states: the stronger a person’s blink when startled, the more unconsciously frightened or threatened she or he is by the sights, smells, and other stimuli at that moment. Because the brain circuitry underlying this phenomenon already is mapped out in detail, researchers have used the startle-induced eyeblink for more than two decades to improve understanding and treatment of dysregulated affective responses in mental illnesses, such as depression.

However, this noninvasive and informative research tool had never been applied to autism spectrum disorder (ASD), which now is thought to affect as many as 1 in every 110 children in the United States. ASD is characterized by impairments in communication and social abilities as well as restricted interests or repetitive behaviors. These unusual behavioral characteristics can prevent many children with autism from learning the most basic skills they need to function in society—such as speech and self-care—if they are not addressed early with intensive therapy.

“I had always thought that the startle-induced eyeblink would be a beautiful measure to understand social deficits and repetitive behaviors in autism,” Dichter says. “But I never had the opportunity to try it.”

Dichter found the opportunity in a research partnership with Benning, an assistant professor of psychology at Vanderbilt University. Benning was one of the first researchers to describe what is known as the postauricular reflex, which also varies in response to unconscious affective states. In this reflex, a tiny muscle behind the ear contracts when a person is startled—and contracts more strongly when a person is immersed in pleasant stimuli. (Although invisible in humans, you can see this reflex pull back your dog’s ears when you call him.) Together, these two startle-induced reflexes—eyeblinking and the postauricular reflex—can provide a rich picture of how the brain’s affective processing systems work.

ARRA Opens a Door

With these two complementary approaches in hand, Dichter and Benning applied for funding from the National Institute of Mental Health (NIMH), which has been supporting Dichter’s training as a new investigator. The collaborators planned to explore whether they could use these measures to reveal atypical, unconscious affective states in people with autism, hoping that this information could lead to new research directions and treatment progress in autism as it had in other disorders.

Their plans, however, hit a wall. In the tight competition for research funding, their application just barely missed the cutoff. But their bad luck suddenly was reversed when ARRA funding arrived at NIMH.

NIMH used roughly 25 percent of its ARRA money to fund studies that had already scored well in review but were beyond the Institute’s ability to support. Grants selected for ARRA funding were deemed highly mission-relevant and scientifically meritorious, including this project by Dichter and Benning.

“We felt really lucky about that,” Dichter says. “We had been thinking about resubmitting our application, but there really was no guarantee that it would have ever been selected.”

The grant allowed the collaborators to hire a full-time project coordinator—“a new college graduate desperate for a job,” Dichter recalls. ARRA also provided portions of the salaries of six other individuals, including a graduate student and a biostatistician.

With ARRA support, Dichter and Benning also purchased a key piece of equipment, an electrophysiology recording system. This apparatus is capable of recording and processing 64 different bioelectrical signals at once. By capturing the electrical impulses created by contracting muscles, this machine can detect the slight eyeblink and the invisible postauricular reflex that are activated when a person is startled. By measuring the size of these electrical pulses, the recording system can reveal someone’s underlying affective states, providing insights into brain circuits.

Measuring the Invisible

Thanks to preliminary research that Dichter and Benning had been conducting prior to receiving grant support—“begging, borrowing, and stealing time on other people’s equipment,” Dichter laughs—the research team “hit the ground running” when they received the news of their support.

With some funds from the grant, the NC Autism Research Registry helped the project team find adults and children with autism for the study. (The UNC-Chapel Hill–based registry is a huge database of people who want to participate in research.) People without autism also volunteered to serve as a comparison group.

Despite the advanced equipment that played a large role in the study, participants’ tasks were simple: look at pictures and then rate how pleasant and interesting they were.

The research team used pictures to elicit subtle affective states in volunteers, who, after getting comfortable with the study procedures, watched the images appear one by one on a computer screen. With each picture, a startling static noise triggered tiny unconscious reactions throughout each participant’s body. Sensors measured the eyeblink and postauricular reflex components of this startle response. Volunteers viewing the familiar images—a snake, a puppy, and a shoe—weren’t aware of the emotional signals that they were sending.

In participants without autism, affect-modified reflex responses followed the typical pattern. Their picture ratings also were consistent with their unconscious reactions. For example, they identified the most pleasant pictures of the set, not knowing that their muscle reflexes had already picked these pictures out when they were startled by the static noise.

The results from the participants with autism were different. These volunteers had fear responses, as detected via stronger eyeblink reflexes, when they were startled by the noise while looking at the pictures that were supposed to be the most pleasant, such as beautiful landscapes. And when they saw the pictures that had caused fear responses in the other group—such as car crashes—they tended to have an unconscious attractive response to the images, as measured by the electrical strength of the muscular contraction behind the ear. Participants with autism later rated the pictures “correctly,” even though this was at odds with their unconscious emotional reactions.

Dichter and his colleagues published these results in a July 2010 paper. They see these unusual startle reflex responses as evidence for a new way of looking at autism, one which ties together the disorder’s seemingly disconnected characteristics.

“Could it be the case that the abnormal social function and repetitive behaviors that are the hallmarks of autism are both explained by broader dysfunction in affective processing?” Dichter wonders, explaining his framework for this and other projects. Autism researchers often focus exclusively on the brain system controlling social interactions, which is just one part of the affective processing system. “This new model would be a way to conceptualize and guide research to look at these symptoms in different ways and to suggest new lines of research.”

Looking to the Future

This relatively small, exploratory project is contributing to a larger undertaking, as Dichter and colleagues pursue this new paradigm for understanding autism. Thanks to growing evidence, this framework is spreading among the leaders in the autism field.

The team now is using the affect-modified startle response approach in studies of the unusual, narrow interests common in autism, such as a fascination with fans or street signs. This measure also is contributing to the team’s imaging research, which is examining how the brain’s reward circuitry responds to pictures of human faces. (The brain’s reward system induces the feeling of pleasure as a “reward” for certain behaviors.)

The goal of using these methods to understand and improve autism treatments is getting closer. With a multidisciplinary team, Dichter is using the affect-modified startle reflex to study how people on specific drug or therapy regimens improve their affective responses. He hopes that treatment personalization isn’t too far away.

“This is going to be a nice assay for response to treatment and also—hopefully—a predictor of response,” Dichter explains. “Now, we can only take a trial-and-error approach to treatment. But if there is a measure to indicate that one child should start with treatment A and another start with treatment B, that would be enormously beneficial.”

This ARRA-funded project will have another long-lasting benefit—the electrophysiology recording system.

“This purchase helped to establish a new electrophysiology lab within the Carolina Institute,” Dichter says. “We have plans to turn that into a resource that’s available to other institute investigators,” studying both autism and other disorders.

Dichter reflects, “It’s really the gift that keeps on giving.”

Dichter wishes to acknowledge the support of the Carolina Institute for Developmental Disabilities for supporting aspects of this research.

Recovery Act Investment: “Emotion-Modulated Psychophysiology of Autism Spectrum Disorders”; Gabriel S. Dichter; University of North Carolina Chapel Hill; 2009: $258,981 (1R21MH085254-01A1); 2010: $156,781 (5R21MH085254-02). Funded by the National Institute of Mental Health.

Publications listing this Recovery Act Investment as providing grant support:
Dichter GS, Benning SD, et al. Affective modulation of the startle eyeblink and postauricular reflexes in autism spectrum disorder. Journal of Autism and Developmental Disorder, 2010;40(7):858–869.