DMN Connectivity Predicts Location-Learning-Related Attentional Capture Vulnerability

Individual differences in susceptibility to attentional capture may reflect stable differences in large-scale brain organization. Using task-based fMRI (N = 33), we tested whether default mode network (DMN) connectivity predicts distractor interference after spatial probability learning. Participants were exposed to spatial bias applied either to same-dimension distractors (SS group, n = 16) or to different-dimension distractors (DS group, n = 17). DMN connectivity predicted capture specifically for the distractor dimension associated with location learning. Cross-prediction analyses further showed that this relationship generalized across groups for the location-learning but not for the comparison dimensions. Although this asymmetry may partly reflect the greater reliability of the more frequently sampled biased-dimension measure, the overall pattern is consistent with vulnerability expressed in the location-learning context rather than with fixed dimension-specific susceptibility. These findings suggest that DMN connectivity indexes a trait-like vulnerability to learned suppression failure.