The current pilot study investigated whether inhibitory control is especially decreased in patients with BED when angry vs. neutral facial expressions are presented. Data were compared with a BMI-matched control group (BMI-CG) as well as a normal weight control group (NW-CG) in a well characterized sample, while controlling for potential confounders like emotion perception and current mood, and by using an objective and validated laboratory task, the eSST. Like expected in hypothesis (1), all participants showed decreased inhibitory control in the angry vs. neutral condition, i.e. a faster SSRT and less correct reactions (%). Against our hypothesis (2), patients with BED did not show decreased inhibitory control in the eSST compared with the BMI-CG and the NW-CG, neither in the angry nor in the neutral condition. BMI-CG and NW-CG did also not differ from each other (hypothesis 3). Correlational analyses (4) yielded no significant associations between the eSST and negative urgency, disorder- or emotion-related sample characteristics.
The main effect concerning stimulus condition implies that all participants had decreased inhibitory control when angry vs. neutral facial expressions have been presented. This speaks for a general negative urgency effect in the eSST that concerns all participants, irrespective of eating behavior. It might be that this effect was so strong that it overrides a putative eating-disorder related effect in patients with BED. Additionally, as we have not included other negative facial expressions, it remains unclear if our results concerning decreased inhibitory control are specific for anger.
According to the evidence from several reviews and meta-analyses (3, 18, 25, 26), inhibitory control is decreased in patients with BED at least in comparison with normal weight controls. However, evidence in comparison with obese controls is heterogenous, especially in the Stop Signal task (25, 26). Moreover, the eSST considered negative mood, but did not take disorder-specific features like food into account. Thus, the eSST might not address the specific deficits in inhibitory control seen in patients with BED. Additionally, anger might not have been the appropriate emotion to decrease inhibitory control in patients with BED, although previous evidence suggests this role of anger in BED (11, 12). For instance, the patients with BED in our study did not show increased trait anger in comparison with the control groups. Though Zeeck et al. (11) reported that anger was most often preceding binge eating episodes, they also reported that the relationship between the desire to eat and binge eating was highest, when patients feel lonely, disgusting, exhausted or ashamed. Likewise, a current review delivers evidence especially for shame as a trigger for binge eating (38). Another explanation could be that the simple presentation of angry facial expressions did not induce anger in the participants and thus, did not affect inhibitory control. Results from the mood ratings before and after the eSST support this assumption as mood decreased after the eSST, but was still rated as highly positive.
Regarding correlational analyses, results have been sobering as well. Behavioral parameters from the eSST did not correlate with any of the proposed related variables, neither with eating disorder pathology, nor with negative urgency, trait anger or emotion regulation. Though self-reports and laboratory tasks are often not related to each other, especially concerning impulsivity concepts (21), and though the Stop Signal task is well validated to measure inhibitory control, it might be that the eSST does not assess inhibitory control in conjunction with negative urgency. Further, it might be that the perception of emotional stimuli is not problematic for patients with BED or that angry stimuli induce rather anxious and not angry feelings. In the current sample, negative urgency and perseverance were increased in patients with BED, whereas regulation of negative emotions was decreased, so that patients with BED might rather show decreased inhibitory control while they experience angry emotions themselves. Thus, it might be more promising to investigate the interplay of negative emotions and inhibitory control by inducing anger vs. neutral mood with a mood induction technique and present food as disorder-specific stimuli in an inhibitory control task. In this case, disorder-related as well as emotional factors would have been considered. However, mood induction techniques are not standardized and not easy to apply (39, 40).
With regard to the limitations of this study, we have to clearly state that this is a pilot study with a small sample size and smaller effects could be undetected due to poor statistical power. Further, even if not correlated with the outcomes of the eSST, the NW-CG reported a somewhat higher positive mood and lower depression scores in comparison with the patients with BED. Moreover, social emotional stimuli that are not disorder-related might not trigger impulsive reactions in patients with BED.