Summary: New residential research shows that people with eating disorders such as bulimia nervosa do not necessarily lose control and binge-eat as a direct result of acute stress.
Source: University of Cambridge
A controlled residential study led by researchers at the University of Cambridge finds that stress does not simply trigger loss of self-control and binge-eating in people with eating disorders, challenging a long-standing theory.
The Cambridge-led findings appear in the Journal of Neuroscience and are based on carefully controlled observations combining behaviour, brain imaging and hormonal measures.
Bulimia nervosa and the binge-purge subtype of anorexia nervosa share core symptoms — recurrent binge-eating episodes followed by compensatory behaviours such as vomiting — but differ in body mass index (BMI). Adults with anorexia nervosa typically have a BMI below 18.5 kg/m2. In the UK, more than 1.6 million people are estimated to have an eating disorder, most of them women.
A prominent hypothesis has held that stress undermines self-control and thereby precipitates binge-eating, but this idea had not previously been tested directly in clinical populations under carefully controlled conditions.
To test the hypothesis, Cambridge researchers and clinicians from the Cambridgeshire and Peterborough NHS Foundation Trust invited 85 women—22 with anorexia nervosa (binge-purge subtype), 33 with bulimia nervosa, and 30 healthy controls—to stay for two days at the Wellcome Trust–MRC Institute of Metabolic Science Translational Research Facility (TRF). The TRF includes an Eating Behaviour Unit and is configured so diet, environment and metabolic status can be controlled and recorded while keeping the setting as naturalistic as possible.
Each morning participants received standardised meals planned by a nutritionist and then completed a fasting period prior to brain imaging. While fasting, the volunteers performed cognitive tasks inside the adjacent Wolfson Brain Imaging Centre while undergoing functional MRI (fMRI) scans.
The primary cognitive measure was a stop-signal anticipation task. Participants tried to stop a moving bar before it crossed a centre line by pressing a key; on some trials a stop-signal halted the bar automatically, requiring participants to withhold their response. This task assesses both proactive inhibition (slowing or preparing to stop in anticipation) and reactive inhibition (outright stopping).
To induce stress on one of the study days, volunteers completed mental arithmetic under pressure, received unpredictable mild electric shocks, and heard negative feedback such as “Your performance is below average.” They repeated the stop-signal task while stressed. After testing, participants returned to the Eating Behaviour Unit and were offered an unobtrusive “all you can eat” buffet to measure ad libitum food intake while still potentially affected by the experimental manipulation.
On the alternate day, the same tasks were carried out without shocks or evaluative pressure; for some participants the order of stress and control days was reversed to control for order effects.
Dr Margaret Westwater, lead author of the study during her PhD at Cambridge’s Department of Psychiatry, said the design allowed the team to examine whether stress altered brain regions linked to self-control and whether any change translated into increased food intake. “What we found surprised us and runs counter to the prevailing theory,” she said.
Behaviorally, women with bulimia nervosa showed reduced proactive inhibition: they were less able to slow responses in anticipation of needing to stop. This deficit was not observed in the anorexia nervosa group. The bulimia group’s impaired proactive control coincided with increased activity in a prefrontal cortex region, suggesting they may have been recruiting that area to compensate but still failed to engage other brain regions needed for optimal performance.
Crucially, inducing stress did not worsen task performance for either patient group or the healthy controls. Stress did, however, produce distinctive changes in prefrontal activity in both diagnostic groups, and these neural responses differed between anorexia and bulimia participants.
At the buffet, patients overall ate less than controls, and caloric intake did not differ significantly between stress and non-stress days for any group. Nevertheless, activity in two specific brain regions correlated with the number of calories consumed across all participants, indicating those regions have a role in dietary control regardless of diagnosis.
Hormonal measures collected during earlier work by the team revealed stress-related changes in appetite-related hormones that differ by diagnosis. In the anorexia nervosa group, stress increased levels of ghrelin (a hunger signal) and peptide YY (PYY, a satiety signal) simultaneously, creating conflicting physiological cues about appetite. In contrast, the bulimia nervosa group showed no consistent changes in ghrelin or PYY but did exhibit lower cortisol levels than healthy volunteers under stress — a pattern seen in conditions of chronic stress or depression.
Professor Paul Fletcher, joint senior author, emphasised that the relationship between stress and binge-eating is complex and cannot be reduced to a single behavioural mechanism. “Our results show that environment, psychological state and bodily hunger and satiety signals interact in ways that vary between disorders,” he said. “A clearer picture of how gut signals influence higher-level cognitive processes like self-control and decision-making will improve our ability to help people with these disabling illnesses.”
Funding: The study was funded by the Bernard Wolfe Health Neuroscience Fund, Wellcome, the NIH–Oxford–Cambridge Scholars Program and the Cambridge Trust, with additional support from the NIHR Cambridge Biomedical Research Centre.
About this eating disorder research news
Source: University of Cambridge
Contact: Craig Brierley – University of Cambridge
Image: The image is in the public domain
Original Research: Closed access. “Prefrontal responses during proactive and reactive inhibition are differentially impacted by stress in anorexia and bulimia nervosa” by Margaret L. Westwater [MPhil], Flavia Mancini [PhD], Adam X. Gorka [PhD], Jane Shapleske [MD], Jaco Serfontein [MD], Christian Grillon [PhD], Monique Ernst [MD, PhD], Hisham Ziauddeen [MRCPsych, PhD] and Paul C. Fletcher [MRCPsych, PhD]. Journal of Neuroscience
Abstract
Prefrontal responses during proactive and reactive inhibition are differentially impacted by stress in anorexia and bulimia nervosa
Binge-eating is a distressing, cross-diagnostic symptom often linked to impulsivity and negative mood. Neuroimaging studies suggest reduced fronto-striatal activity in bulimia nervosa, and a prevailing theory proposes that stress leads to failures in self-regulation that provoke binge-eating. Direct evidence for stress-induced impairment of self-regulation in clinical binge-eating populations has been lacking.
To address this, 85 women (33 with bulimia nervosa, 22 with anorexia nervosa binge-purge subtype, and 30 controls) completed repeated fMRI scanning while performing the stop-signal anticipation task under experimentally induced stress and neutral conditions across two days. This task measures proactive (anticipatory) and reactive (outright) inhibition.
Results showed reduced proactive inhibition in bulimia nervosa alongside increased prefrontal responses in both patient groups, while reactive inhibition remained intact behaviourally and neurally. Stress altered frontal activity during both forms of inhibition in the clinical groups, but it did not impair task performance. These findings indicate specific inhibitory control differences in bulimia nervosa and distinct stress-related neural responses in both disorders, and they argue against stress-induced self-regulatory failure as a comprehensive explanation for loss-of-control eating.
SIGNIFICANCE STATEMENT
Binge-eating often feels uncontrollable to those who experience it, and it has been theorised to arise from reduced self-regulation under stress. This study used repeated brain scanning to test that idea directly in people with anorexia nervosa and bulimia nervosa. Although action cancellation was intact, bulimia nervosa participants showed impaired anticipatory slowing of responses. Stress changed neural responses in both patient groups without degrading performance, suggesting that a simple model of stress-induced disinhibition is insufficient to explain binge-eating.