Skip to main content
. Author manuscript; available in PMC: 2016 Aug 1.
Published in final edited form as: J Neural Eng. 2015 Jul 14;12(4):046027. doi: 10.1088/1741-2560/12/4/046027

Figure 7.

Figure 7

The transfer learning classifier achieved higher classification accuracy than the traditional classifier for all tested ROIs thought to be related to the P300 ERP. Left: P300a ROIs included middle frontal gyrus and inferior frontal sulcus (light and dark green, respectively). P300b ROIs include intraparietal sulcus and transverse parietal sulci, superior parietal lobule, supramarginal gyrus, and angular gyrus (increasingly darker shades of blue, respectively). Right: Accuracy of the transfer learning classifier ± SEM as a function of pool size for these P300 ROIs averaged over 21 subjects repeated 25 times. Accuracy of the subject-specific classifier is indicated by the circular markers, left. Chance accuracy is 50%. Note scale difference from other BCI ROI figures.