Effects of Negative Affect in Individuals With Binge Eating Episodes

Binge-eating is characterized by recurrent episodes of eating large amounts of - typically high calorie - foods, eating much more rapidly than normal and until feeling uncomfortably full, as well as feeling disgusted with oneself, depressed, or guilty after those episodes. Two eating disorders are characterized by binge-eating as central diagnostic criteria, binge-eating disorder (BED) and bulimia nervosa (BN). Binge-eating episodes in BN, but not BED, are typically followed by compensatory mechanisms such as self-induced vomiting, and BED is typically associated with obesity, while BN is not. Behavior studies such as ecological momentary assessment (EMA) research of affect in an individual's naturalistic environment have shown that negative affect and negative urgency (the tendency to act rashly when distressed) often precede binge-eating.

The Investigators want to answer the following questions: Can negative affect in BN and BED be linked to 1) altered dopamine related brain reinforcement learning, 2) to food value computation and cognitive control circuit function, and 3) can dopamine related brain activation predict eating and negative affect, indicating a brain based neurobiological vulnerability. Answering those questions will help to define binge-eating based on regulation of brain reward, cognition, and emotion circuit function and point toward potential psychopharmacological interventions to normalize brain function and behavior.

Specific Aim 1: To test on two study days, one neutral affect and one negative affect day (affect induction) whether negative affect alters brain reinforcement learning in a dopamine system anchored taste reward learning paradigm (a task for prediction error and reward value computation) during functional magnetic resonance brain imaging (fMRI). Hypothesis 1: In response to negative and compared to neutral affect induction, binge eating disorder (BED) and bulimia nervosa (BN) will show increased positive prediction error and taste valuation regression with striatum and insula activation compared to obese (OB) and healthy control (HC) groups. This will be an indication that negative affect excessively activates dopamine related reward circuit response in BED and BN. Specific Aim 2: To study during a food choice paradigm the effects of negative affect on brain circuitry for cognitive control and food valuation. Specific Aim 2a: To study brain circuitry for food valuation in response to neutral or negative affect. Hypothesis 2a: In response to negative and compared to neutral affect induction, BED and BN groups will show greater striatal activation when rating food for taste, compared to OB and HC. This will indicate that BED and BN are associated with increased affect-regulated motivation to approach high calorie foods after negative affect induction and compared to OB and HC. Specific Aim 2b: To study brain control circuitry during food choice in response to neutral or negative affect. Hypothesis 2b: In response to negative and compared to neutral affect induction, BED and BN groups will show increased insula and striatal, but lower dorsolateral prefrontal cortex (DLPFC) activation compared to OB and HC groups. This will indicate that negative affect activates automatic food reward circuit response, but reduces cognitive control circuitry during food choice in BED and BN. Specific Aim 3: To test whether brain activation predicts food intake or negative affect. Specific Aim 3a: To test whether brain activation can predict food intake during a post-fMRI scan test meal. Hypothesis 3a: On negative- versus neutral-affect days, BN and BED will select more food in a post-fMRI scan test meal; lower DLPFC and higher striatal activation during food choice, and higher (prediction error) taste reward activation in insula and ventral striatum will predict greater food intake. Specific Aim 3b: To test whether brain activation during reward learning or food choice predicts negative affect measured by EMA on the study day, as well as measured between study days in the individual's naturalistic environment. Hypothesis 3b: Insula, ventral striatum and DLPFC activation (taste prediction error task, food choice task) during the negative affect condition will predict the intensity of negative affect immediately after fMRI, as well as intensity of negative affect episodes during the days between brain scans in the individual's naturalistic environment, suggesting a neurobiological affective vulnerability.