Fig. 6

Predicted alpha amplitude correlates with observed amplitude. a To assess how well PaA corresponds with alpha amplitude, we computed correlation values for each subject and electrode for each timepoint over the entire –750 to 1000 ms peri-stimulus interval. Histograms show correlation values concatenated over all subjects, timepoints and electrodes on correct (blue) and incorrect (red) trials in the contralateral and ipsilateral electrodes. Stars on each panel indicate that all correlation values shown in these histograms are significantly different from correlations obtained with shuffled LUTs (see Methods, Supp Fig. 4). Dots in histograms indicate the median correlations for that condition. Timecourses panels on the right show these correlations are relatively stable over time, where the y-axes of the plots run from 0.425 to 0.525, corresponding to the gray shaded area in the histogram. All dots indicate significant difference in the correlations between conditions, evaluated by comparing the obtained t-value with a null distribution of t-values computed by shuffling the condition labels 10,000 times. This analysis was done on a timepoint-by-timepoint basis from −500 to + 1000 ms, as indicated by the non-shaded areas. Main effects of are in black, whereas purple indicates an interaction. Gray dots indicate significance after FDR correction at 0.05. b Three example trials from three different subjects show that PaA shifts on single trials mirror those in alpha amplitude. The y-axis and traces for PaA are indicated in black, while those for amplitude are purple. Note that the y-axis range is different in each subplot to maximize visibility of amplitude and PaA (see Methods). c Trials were sorted according to mean pre (−350:−50 ms) and post (350:650 ms) stimulus frequency. Average amplitude and PaA were then computed on these trials and timepoints. Trials were further split by accuracy, as indicated by blue and red lines. Significant differences were evaluated using a two-way repeated-measures ANOVA. Black stars indicate a significant effect of frequency, whereas a purple star indicates a significant effect of accuracy in the two-way ANOVA