Patients with binge eating disorder (BED) suffer from regular food binges with loss of control. This may be due to dysfunctional approach vs. avoidance tendencies towards food in BED. We applied an approach-avoidance task (AAT), in which n = 24 patients with obesity and active BED (OB-BED), n = 32 patients with obesity without current BED (OB), and n = 25 healthy controls (CO) either approached ("pulled") or avoided ("pushed") high (HC) vs. low calorie (LC) food pictures. We tested the hypothesis that OB-BED patients would show an approach bias (measured as different response times RT) towards HC food compared to the other groups. While there was no main effect for group or direction of movement, a significant main effect for calorie (p < 0.001; RT for HC significantly slower than for LC) was found. Repeated measures ANOVA (rm-ANOVA) for comparison of OB-BED vs. OB vs. CO revealed a significant three-fold interaction group × direction × calorie (p = 0.02). Against our hypothesis, the OB-BED group showed an avoidance bias for LC. In explicit ratings, OB-BED reported a significantly reduced urge to consume LC food compared to the OB group. Similar to OB-BED, CO also showed an avoidance bias for LC. The implications of our results are discussed and future directions in this field of research are presented.
