Does an Eating Disorder Affect Decision-Making?

While the capacity for decision-making has been reported to be reduced among some persons with eating disorders, a team of French researchers led by Dr. Sebastien Guillaume found this was not the case in their study of patients with anorexia nervosa (AN) and bulimia nervosa (BN) (Neuropsychology 2010 Sept. 10, published ahead of print].

Two groups of patients with eating disorders (49 with AN, 38 with BN) were compared with 83 healthy controls. The authors used the Iowa Gambling Task to assess the three groups. In this test, participants are shown 4 virtual decks of cards on a computer screen and told that each time they choose a card they will win some game money. However, occasionally when they choose a card they lose money. The goal of the game is to win as much money as possible—thus some decks become “bad decks” while others are “good decks.” Most healthy participants sample cards from each deck and after 40 to 50 selections they become good at sticking to the “good” decks. The overall idea is that “bad” decks carry larger rewards but also larger losses, while “good” decks have smaller rewards but smaller losses and thus are better in the long run.

All participants were euthymic and not taking any psychotropic medication. After the card task, they completed the Eating Disorder Inventory-2 and the Eating Attitude Test.

No significant differences in decision-making performance were seen between patients with eating disorders and healthy controls or between patients with restrictive and purging types of AN. The authors suggest that previously reported changes could be related to other clinical characteristics of patients with eating disorders.

The effect of metabolic state on decision-making

In a second study, MKael Symmonds, a clinical fellow at the Wellcome Trust Centre for Neuroimaging at University College, London, and colleagues at Yale University, reported that metabolic state had an effect on decision-making among 24 healthy normal-weight male volunteers; data from 19 of the men were included in the final analysis (PLoS One 5:2010; published online June 16, 2010).

Visual analog scales were used to assess hunger, fullness, prospective food consumption, illness, and anxiety after fasting. The men were tested immediately after fasting, immediately after a meal, and 60 minutes before a meal. The subjects performed one of three different decision-making tasks within each hour to make sure that cognitive demand was the same. The authors used a multiple paired lottery choice task, presenting a sequence of 200 paired lotteries; subjects were required to select one preferred option per pair. Each week, the men were exposed to the same set of lotteries, but the position of the 4 cards was changed—none of the subjects reported realizing that the lottery sequences were the same across the three weeks.

The researchers measured suppression of circulating ghrelin and circulating leptin levels, and showed that both immediate and delayed effects could be seen on risky decision-making following a meal and that these changes correlated with an individual’s baseline leptin and ghrelin levels.

Dr. Symmonds and colleagues found that human risk preferences were exquisitely sensitive to the current metabolic state, and feel that this has significant implications for aberrant decision-making in persons with eating disorders or obesity.