Abstract : This study proposes a novel noninvasive ECG-based approach to accurately quantify differences in the atrial activity (AA) signals of atrial fibrillation (AF) patients characterized by very similar global AF organization. High-density body surface potential maps were recorded in 63 patients in persistent AF prior to electrical car-dioversion (32 recurrences occurred after 4-6 weeks). AF substrate complexity was quantified as the degree of overall temporal organization of the AA pattern, determined from multi-variable unthresholded recurrence plots (RP). Results showed that non-recurrent patients were characterized by temporally more stable propagation patterns than recurrent patients, suggesting a lower complexity of the AF substrate in the former. This was used to discriminate between recurrent and non-recurrent AF patients following electrical cardioversion with an accuracy of 72%. Noninvasive assessment of AF substrate complexity in patients characterized by very similar levels of AF progression was improved by applying RP-based analysis on extracted AA signals. The proposed approach may help improving AF treatment selection and success.