Bulimia nervosa ranks second among eating disorders in both its overall prevalence and the profound burden it places on individuals and healthcare systems. Yet, despite its widespread impact, remarkably little has been understood about what the condition looks like physically within the human brain. Past neuroimaging studies have historically been constrained by small sample sizes, and their empirical findings frequently contradicted one another, leaving researchers and clinicians searching for definitive answers.
Now, a major new neuroimaging study published in JAMA Psychiatry has pooled MRI scans from 13 different research sites worldwide to finally establish a clear, consistent picture. The international team of researchers discovered consistent structural differences in specific brain regions tied directly to reward processing, sensory input, and social processing among individuals diagnosed with bulimia nervosa.
About the Study
To overcome the limitations of previous fragmented research, scientists participating in the ENIGMA Eating Disorders Working Group pooled magnetic resonance imaging (MRI) scans collected across 13 distinct academic and medical sites globally, encompassing 17 independent groups of participants. In total, the researchers conducted a rigorous comparative analysis of 369 females diagnosed with bulimia nervosa alongside 417 females without any history of an eating disorder.
To ensure the highest possible degree of scientific accuracy and comparability, all gathered scans were processed and subjected to rigorous quality checks using uniform protocols, allowing researchers to evaluate the numerical data side by side.
The participant pool consisted entirely of females, a demographic selection that aligns closely with epidemiological data showing that bulimia nervosa is approximately five times more prevalent in women than in men. The subjects included in the international pooling ranged widely in age, spanning from adolescence at 12 years old to adulthood at 53, with an average participant age of 23.6 years.
In examining the neuroimaging data, the research team focused on three primary physical metrics: the thickness of the cerebral cortex (the brain’s outer layer), the total surface area covered by that outer layer, and the physical volume of several distinct subcortical structures buried deep within the brain.
Reward and Sensory Regions Measured Smaller
When comparing the brains of women diagnosed with bulimia nervosa to those of healthy controls, the clearest and most pronounced differences manifested in the surface area of the cerebral cortex rather than its thickness.
Specifically, women with bulimia nervosa exhibited significantly less surface area in the superior temporal and transverse temporal regions. These anatomical areas of the brain’s outer layer are primarily responsible for processing auditory information, reading complex social cues, and managing other incoming sensory data from the external environment. Furthermore, participants with the disorder demonstrated a notably lower volume in the nucleus accumbens, a small, deeply situated structure that serves as a core engine for the brain’s reward and motivation circuitry.
In contrast, the overall thickness of the brain’s outer layer showed no meaningful statistical differences between the clinical group and the control group.
Importantly, the researchers noted that these structural differences mostly held steady even after performing statistical adjustments to account for potential confounding variables across the various study cohorts. These adjusted factors included body weight variations, illness duration, accompanying depressive symptoms, medication usage, and the frequency of purging behaviors.
Binge Eating Tracked With the Brain, Purging Didn’t
A particularly intriguing split emerged when the researchers shifted their analytical focus from the primary clinical diagnosis to specific behavioral symptoms. The frequency of binge eating episodes showed a direct correlation with structural variations across a wider network of brain regions.
More frequent binge eating was linked to reduced surface area in the insula—a region that helps individuals perceive internal physiological signals originating from inside their own bodies—as well as orbital and frontal cortical areas involved in judging how rewarding an experience will feel and in exerting cognitive control to rein in impulses. Additionally, these variations extended into parts of the cingulate cortex, which is intimately tied to the processing and regulation of emotion.
Conversely, the frequency with which individuals engaged in compensatory behaviors, such as purging following a binge, showed no measurable relationship to brain structure whatsoever. Clinicians currently rely heavily on the frequency of these compensatory behaviors to rate the clinical severity of a patient’s bulimia nervosa. The discovery that binge eating and compensatory mechanisms track differently with brain anatomy suggests that these two core features may operate on distinct neurological wiring. Over time, these insights could fundamentally reshape how clinical severity is measured and how targeted treatments for bulimia are developed.
What This Means
The study is widely regarded by experts as providing the clearest evidence to date that bulimia nervosa is associated with reproducible, measurable differences in brain structure. According to Laura A. Berner, PhD, Director of the Center for Computational Psychiatry at Mount Sinai and a key contributor to the research, the findings offer unprecedented clarity on a condition that has long been misunderstood.
The anatomical locations of these observed brain differences caught the research team somewhat by surprise. Because the surface area of the brain’s outer layer is a morphological feature largely determined early in human development, these findings raise important questions regarding whether some of these structural variations reflect an underlying vulnerability that was present long before any clinical symptoms ever manifested.
At the same time, Dr. Berner emphasized the importance of interpreting these neuroimaging results with care. She drew a firm line against deterministic thinking, stating that the findings should not be interpreted to mean that the brain is permanently fixed or that an individual’s long-term recovery outcome is predetermined. Instead, the research underscores that bulimia nervosa is a genuine biological mental illness characterized by measurable neurological changes, rather than a behavioral failing rooted in a lack of willpower or self-control, as persistent social stigma often implies.
The Takeaway
Because the participants in this large-scale study were scanned at a single point in time, researchers acknowledge that they cannot yet definitively establish causality—specifically, whether the observed brain changes preceded the onset of disordered eating behaviors or developed as a consequence of them.
Nonetheless, the empirical data generated by this international collaboration provides clinicians with a tangible, physical thread to follow in the ongoing effort to improve prevention strategies and therapeutic interventions. For the millions of individuals living with the illness, the research offers validation and a significant step forward in dismantling long-standing misconceptions about their health.
