Self-Control in Bulimia Nervosa

Frontostriatal activation modulated by the predicted need for inhibition (P(stop) (i.e., parametric modulation by p(Stop)), measured as the mean BOLD signal across voxels within the region of interest (ROI). Positive values indicate that neural activation increases as the predicted probability of needing to stop increases (i.e., stronger responses when stopping is predicted to be more likely), whereas negative values indicate that activation decreases as the predicted probability of needing to stop increases (i.e., greater activation when stopping is predicted to be less likely).

Frontostriatal activation modulated by unsigned inhibitory control prediction errors (i.e., parametric modulation by absolute prediction error magnitude), measured as the mean BOLD signal across voxels within the region of interest (ROI). Positive values indicate that neural activation increases with the magnitude of unsigned prediction errors (i.e., stronger responses to more surprising outcomes), whereas negative values indicate that activation decreases as unsigned prediction errors increase (i.e., relative deactivation for more surprising outcomes).

Frontostriatal activation modulated by signed inhibitory control prediction errors (i.e., parametric modulation by prediction error magnitude), measured as the mean BOLD signal across voxels within the region of interest (ROI). Positive values indicate that neural activation increases as signed prediction errors become more positive (i.e., stronger responses to the surprising need for control), whereas negative values indicate that activation decreases as signed prediction errors become more positive (i.e., stronger responses to the surprising lack of need for control).

Frontostriatal activation associated with successful inhibition , measured as the mean BOLD signal across voxels within the region of interest (ROI) for the contrast of Successful Stop vs. Go trials. Positive values indicate activation associated with successful inhibition.

Behavioral performance on the stop signal task, as measured by stop signal reaction time calculated as mean go reaction time minus the stop-signal delay at 50% successful inhibition (SSD₅₀). Higher SSRT means longer latency before failed inhibition.

Percent correct responses to stop trials on the Stop Signal Task measured by percent of stop trials on which participant successfully withheld a response (vs. failed to withhold response).