Figure 6From: Online detection of error-related potentials boosts the performance of mental typewritersErrP Classification. Performance of the online ErrP classifiers. The AUC for each participant and for the mean are depicted in the top left plot for classifier A (magenta) and classifier B (blue). The hit rates and false alarm rates for each participant and classifier are shown underneath. The right plots show the ROC curves of each participant for classifier A (top) and classifier B (bottom). Classifier B has higher AUCs in all but one participant. The mean AUCs differ more than 10%. This can also be seen in the ROC curves, which reach higher hit rates at lower false alarm rates for classifier B. False alarm rates could be kept around 5% in most participants. In two participants, the false alarm rate approached 10%.Back to article page