Eating Disorder Study Shows How the Brain Dictates Food Choices

Walyce Almeida
November 14, 2017

Joanna Steinglass has worked with anorexia nervosa patients since 1999, when she was a resident psychiatry at the College of Physicians & Surgeons. “With eating disorders, it feels like the absence of eating is the problem,” she said. “It’s been challenging to figure out how best to study that.”

Anorexia is a psychiatric illness characterized by an inability to maintain a healthy weight. One of the least common eating disorders (binge eating disorder is far more prevalent), it has the highest mortality rate of any mental illness, according to the National Association of Anorexia Nervosa and Associated Disorders.

Diagnosis can be tricky and is often complicated by the fact that many patients hide their disorder until severe weight loss causes medical problems. Combatting it often requires a team of mental health experts, dietitians and medical doctors.

“It’s very scary and confusing for patients and their families to deal with this illness,” Steinglass said. “For most people, we choose food based on how tasty it is. But patients with anorexia nervosa restrict what they eat. This illness has a biological basis. Studying food choice is a really nice new pathway to find the brain mechanisms of restrictive eating.”

Steinglass, an associate professor of clinical psychiatry in Center for Eating Disorders, a joint venture between the New York State Psychiatric Institute and Columbia University Medical Center, reached out to psychology professor Daphna Shohamy, whose Learning Lab is based in Columbia’s Mortimer B. Zuckerman Mind Brain Behavior Institute.

“My lab focuses on how our choices are driven by how we learn from experience,” said Shohamy, who studies the cognitive neuroscience behind decision-making. “We’re interested in value or reward-based decisions, where you’re trying to make a choice that will maximize a reward.”

Together, they set up a study using two groups of women. Only one group had been diagnosed with anorexia nervosa, but both were asked to make food choices while undergoing an fMRI scan. Shohamy wanted to see what happens in the striatum, an area of the brain involved in decision-making, amongst other things.

That’s where the fMRI showed distinct differences between the study participants, with the anorexia patients showing an association between activity in the dorsal striatum and decisions about food, similar to those who suffer from substance abuse, Shohamy said.

Based on the findings, Steinglass formulated a hypothesis that an anorexia patient’s brain learns early on that eating less food is somehow rewarding (rewards may differ for different individuals). A healthy person’s decision on what to eat can be updated with new data - such as when a positive behavior is no longer positive - enabling the possibility of a different choice. But for someone with anorexia nervosa, the striatum has created a routine based on rewarding restrictive eating.

“We could see that patients appeared to be using different neural systems,” Steinglass said. “These brain mechanisms are more related to automatic behaviors and less related to deliberative choices that people make.”

The study also involved observing participants’ eating habits outside of the clinic. “Those patients who showed the biggest differences in brain activity were also the patients who showed the lowest consumption of calories,” noted Shohamy. “It’s a correlation, but we don’t know what’s driving it. It’s a good first step.”

Now Steinglass and Shohamy are developing treatments that focus on intervention and disrupting the cycle in the striatum.

“We did a treatment study where we interrupted cues that trigger the automatic behavior,” Steinglass said, adding that re-nourishment is a crucial step in treating anorexia nervosa. “But we don’t understand enough of the biology behind it to develop medication.”

Their study does not solve the mystery of when and how anorexia develops. The professors have received grants from the National Institutes of Health for two further studies based on Steinglass’ hypothesis. One will focus on teenagers and the other on people who fall on different points along the along the spectrum of restrictive eating—from healthy to pathological.

Shohamy said that people take it for granted that we function by improving our behavior based on past experiences. She cautions that “you might not see how your decisions are affected by your brain and not your feelings.”