About me
Unraveling Neural Mechanisms of Binge Eating Through Spectral Dynamic Causal Modeling
Location: 23
Mentor: Dr. Roger McIntosh
Binge eating is associated with altered brain connectivity, yet the directional patterns underlying these changes remain unclear. We analyzed the effective connectivity (EC) of brain regions implicated in binge eating. Resting-state functional magnetic imaging data were analyzed in 158 healthy participants (age: 38.02 ± 12.82, 112 females). We used the Eating Disorder Examination Questionnaire to identify a binge eating group (N = 30) and a control group (N = 128). We then used spectral dynamic causal modeling to measure EC differences within an a priori food-processing network between these two groups. Finally, we used leave-one-out cross-validation to predict group membership based on significant connections and conducted an area under the receiver operating characteristic (AUC-ROC) analysis to assess their predictive strength. Results revealed significantly stronger EC patterns in the binge eating group in regions implicated in cognitive control, reward processing, and homeostatic regulation. Among these significant connections, right medial orbitofrontal cortex to nucleus accumbens (AUC = .65, 95% CI [.53, .76]), left posterior cingulate cortex to left dorsolateral prefrontal cortex (AUC = .62, 95% CI [.51, .74]), and the left posterior cingulate cortex self-connection (AUC = .60, 95% CI [.50, .71]) significantly predicted group membership, as their 95% confidence interval lower bounds exceeded .5. These findings reveal directional intrinsic connectivity patterns linked to binge eating, with some associations robust enough to predict group membership. Such connectivity patterns may serve as biomarkers for binge eating susceptibility, offering valuable insights to inform targeted interventions and treatment strategies for binge eating disorder.
