International Journal of Obesity (2015) 39, 353–360 © 2015 Macmillan Publishers Limited All rights reserved 0307-0565/15 www.nature.com/ijo ORIGINAL ARTICLE Attentional impulsivity in binge eating disorder modulates response inhibition performance and frontal brain networks MA Hege1,2, KT Stingl1,3, S Kullmann1,4,5, K Schag6, KE Giel6, S Zipfel6 and H Preissl1,4,5 BACKGROUND: A subgroup of overweight and obese people is characterized by binge eating disorder (BED). Increased impulsivity has been suggested to cause binge eating and subsequent weight gain. In the current study, neuronal correlates of increased impulsivity in binge eating disorder during behavioral response inhibition were investigated. METHODS: Magnetic brain activity and behavioral responses of 37 overweight and obese individuals with and without diagnosed BED were recorded while performing a food-related visual go–nogo task. Trait impulsivity was assessed with the Barratt Impulsiveness Scale (BIS-11). RESULTS: Specifically, increased attentional impulsiveness (a subscale of the BIS-11) in BED was related to decreased response inhibition performance and hypoactivity in the prefrontal control network, which was activated when response inhibition was required. Furthermore, participants with BED showed a trend for a food-specific inhibition performance decline. This was possibly related to the absence of a food-specific activity increase in the prefrontal control network in BED, as observed in the control group. In addition, an increase in activity related to the actual button press during prepotent responses and alterations in visual processing were observed. CONCLUSIONS: Our results suggest an attentional impulsiveness-related attenuation in response inhibition performance in individuals with BED. This might have been related to increased reward responsiveness and limited resources to activate the prefrontal control network involved in response inhibition. Our results substantiate the importance of neuronal markers for investigating prevention and treatment of obesity, especially in specific subgroups at risk such as BED. International Journal of Obesity (2015) 39, 353–360; doi:10.1038/ijo.2014.99 INTRODUCTION on both dimensions using two different eye-tracking paradigms.8 Behavioral and cognitive control of eating behavior is one of the Furthermore, activation differences in neuronal networks involved 9,10 hallmarks for preventing and treating obesity.1 Binge eating in reward processing have been reported for BED. In particular, 9 disorder (BED), with a prevalence of ~ 3% in the general Schienle et al. evaluated reward sensitivity in the context of brain population and more than 30% in patients enroled in weight- responses to visual food cues. BED patients reported enhanced control programs, is characterized by recurring episodes of eating reward sensitivity, which was related to stronger medial orbito- large amounts of food with subjective sense of loss of control frontal cortex activation in comparison with control participants. over eating and subsequent distress.2,3 BED is strongly associated Rash-spontaneous behavior is assessed with psychometric self- with obesity as, in contrast to bulimia nervosa, no weight- report trait measures as, for example, the Barratt Impulsiveness compensatory behaviors are taken.3,4 Furthermore, BED is Scale (BIS-11).7 Neuronal and behavioral correlates of impulsivity characterized by increased impulsivity, a personality-related can be assessed by a variety of tasks such as go–nogo or stop construct that correlates with deficits in control over food signal tasks. In a go–nogo task, participants have to inhibit intake.5–7 A recent factor analysis of impulsivity as a personality behavioral responses that are inappropriate in a current context trait by Dawe and Loxton7 found two main factors, namely ‘reward and prepotent response inhibition is assessed. Participants are sensitivity’ and ‘rash-spontaneous behavior’. Reward sensitivity is required to perform speeded responses on go trials and to described as a purposeful drive to obtain rewarding stimuli and withhold responses on nogo trials. Aichert et al.11 showed that rash-spontaneous behavior as a tendency to act rashly and performance (number of inhibition errors) during a go–nogo task without consideration of consequences, which is associated with is related to trait impulsivity (evaluated by BIS-11). Furthermore, disinhibited behavior and loss of control. With regard to BED, individuals with BED have been reported to have lower inhibitory reward sensitivity is considered to determine food preferences control than obese and non-obese control participants.12 On the and to have a major part in the initiation of binge eating episodes, neuronal level, activation of a prefrontal network during response whereas rash-spontaneous behavior leads to loss of control during inhibition is crucial for task performance in go–nogo tasks.13,14 these episodes and the inability to resist binge cravings.7 In a A decrease in activation in cerebral regions of this prefrontal recent behavioral study, we showed increased impulsivity in BED network has been shown to be associated with increased trait 1Institute of Medical Psychology and Behavioural Neurobiology/fMEG Center, University Tübingen, Tübingen, Germany; 2Graduate School of Neural and Behavioural Sciences, International Max Planck Research School, University Tübingen, Tübingen, Germany; 3Department of Neonatology, University Children's Hospital Tübingen, Tübingen, Germany; 4German Center for Diabetes Research (DZD), Neuherberg, Germany; 5Institute for Diabetes Research and Metabolic Diseases of the Helmholtz Center Munich at the University of Tübingen (IDM), Tübingen, Germany and 6Department of Psychosomatic Medicine and Psychotherapy, Medical University Hospital Tübingen, Tübingen, Germany. Correspondence: MA Hege, Institute of Medical Psychology and Behavioural Neurobiology/fMEG Center, University Tübingen, Otfried-Müller-Straϐe 47, Tübingen 72076, Germany. E-mail: [email protected] Received 6 February 2014; revised 15 April 2014; accepted 15 May 2014; accepted article preview online 9 June 2014; advance online publication, 8 July 2014 Response inhibition in binge eating disorder MA Hege et al 354 impulsivity.15,16 However, it is not known how increased Participants were asked to respond to go cues by pressing a button with impulsivity in BED relates to activation of the prefrontal control their right index finger before the appearance of the fixation cross. For a network during response inhibition. In a recent magnetoence- detailed description refer to Hege et al.17 phalographic (MEG) study, we established a visual food-related Reaction time for correct responses during go trials (hits) and go–nogo task.17 For lean participants, we reported a stable activity unsuccessful inhibition during nogo trials (unsuccessful withholds) were recorded. In addition, the percentage of correct responses during go trials increase in a right-lateralized prefrontal control network when and the percentage of successful inhibition of responses during nogo trials response inhibition was required during nogo trials. This network (successful withholds) were calculated (accuracy of response for go and was mainly involved in monitoring and gating of the response nogo, respectively). Participants with an accuracy level lower than 50% for inhibition. the nogo condition and 80% for the go condition were excluded from In the current study, we investigated obese and overweight further analysis. patients with BED and a body mass index (BMI)-matched control – group (OB) with this go nogo paradigm. We expected lower Data acquisition response inhibition performance for BED patients, potentially related MEG signals were recorded using a 275-sensor whole-head system (VSM to increased individual impulsivity scores (BIS-11). Furthermore, we MedTech Ltd, Port Coquitlam, British Columbia, Canada). Three coils expected decreased activity in the prefrontal control network during generating magnetic fields were attached at three fiducial points (nasion, response inhibition for highly impulsive individuals and BED patients. preauricular point on each side of the participant’s head). These coils were More specifically, we hypothesized that in particular response used for continuous recording of the head position in relation to the MEG inhibition to food stimuli would be affected in BED. sensor array. MEG data and behavioral responses were recorded in a continuous mode with a sampling rate of 586 Hz. MATERIALS AND METHODS Data analysis Participants Continuous MEG data were cut into trials of 700 ms (−100 to 600 ms) Thirty-seven female, overweight and obese participants volunteered to length according to the presented stimulus and the recorded response participate in this study. A total of 18 participants fulfiled the diagnosis for and filtered with a 1-Hz high pass and a 40-Hz low pass filter. We selected BED according to the Diagnostic and Statistical Manual of Mental trials with successful withholds during nogo trials (successful withholds- Disorders, Fourth Edition (DSM-IV) criteria. Exclusion criteria in all groups food, successful withholds-toy) and correct responses during go trials (hits- comprised somatic diseases or medication influencing weight or eating food, hits-toy). In order to limit the number of go trials, only go trials behavior, pregnancy or lactation, psychotropic medication use except (correct responses) that preceded nogo trials were considered. All trials