Anorexia and Reward

Recovery from anorexia often takes years, and new brain scan research shows that even fully recovered anorexics are distinct from the general population in the way their brains process reward.

According to a study published in the American Journal of Psychiatry, women who never had anorexia showed highs and lows in brain activity that corresponded with winning or losing. But former anorexics showed no difference in their brains' reward centers.

"People with anorexia had a hard time distinguishing win or loss," says Walter Kaye, a psychiatrist and coauthor of the new study.

In addition, the study suggests that the brain circuits of recovered anorexics are overactive in areas of the brain concerned with long-term consequences.

Kaye says these findings in recovered anorexics match what he observes in his work with people currently battling anorexia. "People with anorexia have a hard time living in the present with immediate gratification," says Kaye. As the brain scans suggest, anorexics are "over-concerned with future consequences, which goes along very much with the clinical picture that we see in people with anorexia," he adds.

Reward processing is emerging as important theme in anorexia research, says Kaye. For one thing, people with anorexia tend to worry a lot and find little pleasure in life. Also, studies have shown that the neurotransmitter dopamine, which plays a role in reward prediction, has altered activity in anorexic people.

Kaye, who has treated anorexic patients for 30 years., now views the disorder as one of altered reward perception. "Most people would find long term starvation to be stressful and unpleasant, and they would get very hungry," he says, "and people with anorexia find that's probably the only thing in life that's really rewarding to them."

Looking back, Erin Kennedy says what made her happy years ago now produces more mixed feelings. One Christmas, when she was in the hospital, her parents had brought new size One clothes for Kennedy, who then weighed 80-some pounds. The Christmas gifts were too big on Kennedy's emaciated frame. "In that moment, I was like, 'Good… This doesn't fit me,'" she recalls. But today, Kennedy says she feels guilty that her parents had to return the gifts. "I feel bad about it, you know? Because here it was Christmas time. My parents had to come see me in a hospital," she says.

Guessing Games and Brain Scans

In the new study, Kaye and colleagues from Germany, Austria, and the US had their subjects play a simple guessing game while their brain activity was being recorded through functional magnetic resonance imaging (fMRI).

The game went like this: subjects were asked to predict whether the value of a hidden number—one through nine—was greater or less than 5. Correct answers netted the subjects two dollars, but incorrect answers cost them one dollar. The researchers assured their subjects that the numbers were generated randomly, so guessing would be a sensible strategy. But the game was actually controlled so that no matter how subjects guessed, each would get the same number of wins and losses, and take away $40 in wins.

The researchers observed that control subjects—women who never had an eating disorder--simply made each guess and proceeded to the next question. But the recovered anorexics took a different approach to the task: "They were trying to figure out some strategy to get things right, to not make mistakes, to do this perfectly, which is pretty typical of people with anorexia nervosa," says Kaye.

Kaye and his colleagues found differing activity patterns between the recovered anorexic women and the controls in two brain regions. In a brain region called the anterior ventral striatum, which is important in immediate reward processing, the control subjects showed distinct responses to positive and negative stimuli, as expected.

"If you go to Las Vegas and you gamble $100,000, and you win a $100,000, you're going to feel an awful lot better than if you lose $100,000. So clearly there's some way that your brain is able to determine good and bad, win and loss," he explains. But the recovered anorexics' brains responded similarly to both wins and losses, suggesting that in people with anorexia, this circuit may be altered.

The fMRI scans also revealed a difference in the caudate brain region, which is involved in planning for future outcomes. During the guessing game, this region was overactive in recovered anorexics. The researchers linked this fMRI finding with the subjects' own reports that they were trying to use a strategy to guess the right numbers. People with anorexia "tend to have exaggerated and obsessive worry about the consequences of their behavior, looking for rules when there are none, and they are overly concerned about making mistakes," the researchers wrote.

But Kaye cautions that it's still not known whether the unique patterns of brain activity they found contribute to causing anorexia. Other research has shown that people who develop anorexia often suffer from childhood anxiety symptoms, suggesting that underlying brain differences may predate the onset of disordered eating.

Kaye says new therapies need to be developed and evaluated. "We're really still at a very early stage in terms of developing more effective treatments for this disorder," he says.

And Kennedy's long process of recovery is not uncommon. These days, she enjoys time with family and friends, her home near the beach, and her job helping psychiatric inpatients make the transition back to independent living. "Everyone's there for different reasons, but the feelings are similar, so it's just being able to say, 'I know kind of what you're going through,'" she explains.

Through Eating Disorders Anonymous (EDA), she also leads an eating disorders support group. She hopes that sharing her story will prompt people to take action. "Someone's going to be able to open their eyes and say, 'You know what? My daughter, or my son, my niece or my nephew, doesn't eat. Maybe I can get them some help and stop it,'" she says.

This research was published in the American Journal of Psychiatry, December, 2007 and was funded by the National Institute of Mental Health and the Price Foundation.