202 recordings were segmented into conditions by pasting together all epochs of a single condition. Segmentation was done using markers of the following conditions: baseline moderate-pain stimulations (10 seconds each), baseline high-pain stimulations (10 seconds each), first eyes-closed rest (ECR1; 5 minutes), control acquisition stimulus (10 seconds each), nocebo acquisition stimulus (10 seconds each), control evocation stimulus (10 seconds each), nocebo evocation stimulus (10 seconds each), and second eyes-closed rest (ECR2; 5 minutes). Exclusion of certain segments (for example, segments that were too short for DFA computation) resulted in a varying number of participants across analyses and figures. Statistical analysis EEG biomarkers were computed and tested per EEG-channel for all 32 channels. Non-parametric paired Wilcoxon signed-rank test was used to test for differences between each two conditions. Multiple-comparison corrections were performed using a False Discovery Rate procedure (FDR) with q = 0.05 35,36. For the Wilcoxon signed-rank test, we reported the median of the two conditions tested, the Z-value and the P-value. To test for associations between EEG biomarkers and behavioral outcome measures, we calculated Spearman’s rank correlation coefficient (rs). On all spatial topographies, open white circles reflect statistical significance at P < 0.05, whereas closed white circles indicate statistical significance after FDR correction. Since some statistical effects were widespread across the cortex and others were localized above specific brain areas, we report statistics of the wholebrain average and additionally report statistics of specific electrodes in case of localized effects.
RkJQdWJsaXNoZXIy MTk4NDMw