(PS7-59) Inflated Subjective Costs of Cognitive Effort Are State-specific in Bulimia Nervosa

Abstract Text: Bulimia nervosa (BN) is characterized by deficits in cognitive control processes. However, because individuals with BN oscillate between out-of-control binge eating and over-controlled dietary restriction, these deficits in control might not represent a stable trait. After eating begins in most binges, individuals with BN report thinking, "Now that I have done it, I might as well go all the way." Here, we aimed to examine how the internal state change introduced by eating may abnormally influence value-based decisions about cognitive-control-related effort in women with BN. We hypothesized that for individuals with BN, food intake may abnormally inflate the subjective costs of cognitive control, thus reducing willingness to expend effort on control and promoting out-of-control binge eating episodes. To test this hypothesis, thirty-one women with and without BN completed the Cognitive Effort Discounting task after a 16-hour fast and after a standardized meal. This task assesses the discounting of cash offers to complete more demanding levels of an N-back cognitive control task. In a fed relative to fasted state, women with BN showed steeper discounting than healthy comparison women (B = 0.11, p = 0.014), indicating that cognitive effort was abnormally costly for the BN group after eating. However, there were no group differences in self-reported measures of task difficulty (B = -0.76, SE = 0.77, t = -0.99, p > 0.05) or effortfulness (B = -0.74, SE = 0.77, t = -0.96, p > 0.05) in either state; groups did not differ on N-back task accuracy in either state (B = 0.51, SE = 0.36, t = 1.41, p > 0.05), and there was no association between accuracy on the N-back task and the subjective costs of cognitive effort in women with BN (r = -0.01, p > 0.05). Therefore, the steeper discounting in a fed vs. fasted state seen in women with BN suggest that individuals with BN have inaccurately inflated subjective costs of cognitive effort after eating. Further, this state-specific subjective cost of control varied with symptoms in women with BN: women who discounted cognitive effort more steeply in the fed state reported more episodes of out-of-control binge eating (rho = -0.64, p = 0.001), and women who showed the greatest increases in subjective effort costs when moving from a fasted to a fed state spent more days fasting (i.e., completely avoided eating any amount of food) between binge episodes (r = -0.54, p = 0.037). Together, our data indicate that state-specific alterations in effort-cost computations may perpetuate out-of-control binge eating episodes that alternate with periods of dietary restriction in BN. While existing interventions for BN have focused on enhancing cognitive control in general, this study supports the potential investigation of new and state-specific treatments that help individuals with BN to decrease the subjective costs and re-evaluate the benefits of cognitive effort before the maladaptive engagement of control occurs. Our ongoing analyses are exploring whether the observed effect of eating on cognitive effort costs can be explained by altered postprandial hormone levels and neural connectivity in women with BN.